[beginning of typical troll thread left away
  Discussion went then from Unix>NT to Multics>Unix and Tops-10>Multics
  this lead to an long good thread on Tops-10 Interna, recorded here]

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 21 Apr 99 08:49:34 GMT
Organization: UltraNet Communications, Inc.
Lines: 21
Message-ID: <7fk9i4$spj$1@ligarius.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu>
NNTP-Posting-Host: d8.dial-13.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 21 Apr 1999 10:33:40 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!masternews.telia.net!newsfeed1.swip.net!swipnet!newsfeed.cwix.com!18.24.4.11!newsswitch.lcs.mit.edu!news.ultranet.com!d8

In article <Rf5T2.194$W4.62887@srvr1.engin.umich.edu>,
   ftit@engin.umich.edu (Sergej Roytman) wrote:
>In article <FAI5rt.72p@midway.uchicago.edu>,
>Eric Fischer <eric@fudge.uchicago.edu> wrote:
>>Sergej Roytman <ftit@engin.umich.edu> wrote:

<snip>

>So, no, not deluded about Unix reliability---old or new.  I'm carrying
>on a correspondence with a friend who used to do Multics at Oakland
>University and GMR.  The conversation has touched on Multics, and how
>Unix is a step down from that, and after checking out the Multics site
>at best.com... what can I say? I'm impressed.  It's actually raised
>some questions with me; perhaps I'll ask them here once I think them
>through a bit better than I did that statement you flagged.
>
Sigh!  And TOPS-10 was better than them.  

/BAH

Subtract a hundred and four for e-mail.

######

From: roggblake@iname.com (Roger Blake)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 21 Apr 1999 13:10:42 GMT
Organization: Ministry of Silly Walks
Lines: 14
Message-ID: <slrn7hrjii.dbl.roggblake@unix5.netaxs.com>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net>
NNTP-Posting-Host: unix5.netaxs.com
X-No-Archive: Yes
X-Newsreader: slrn (0.9.4.3 UNIX)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!newscore.gigabell.net!newscore.ipf.de!newsfeed.ecrc.net!news-xfer.newsread.com!netaxs.com!newsread.com!netaxs.newsread.com!roggblake

On Wed, 21 Apr 99 08:49:34 GMT, jmfbahciv@aol.com <jmfbahciv@aol.com> wrote:
>Sigh!  And TOPS-10 was better than them.  

That seems to be the basic point.  Microshaft had much prior art to draw
upon in designing stable, efficient multitasking operating systems.  The
problems and issues inherent in doing so had been worked out decades
ago.  What's truly amazing is that with all of Unix's fundamental
flaws and shortcomings (hilariously documented in the "Unix-Hater's
Handbook"), it still looks like a stable, well-designed OS when put
up against anything coming out of Redmond.

-- 
  Roger Blake
  (Remove second "g" in name to send e-mail.)

######

Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net>
Organization: University of Michigan, College of Engineering
From: ftit@engin.umich.edu (Sergej Roytman)
Lines: 13
Message-ID: <BlpT2.3$MF5.1852@srvr1.engin.umich.edu>
Date: Wed, 21 Apr 1999 19:07:13 GMT
NNTP-Posting-Host: 141.212.106.44
X-Trace: srvr1.engin.umich.edu 924721633 141.212.106.44 (Wed, 21 Apr 1999 15:07:13 EDT)
NNTP-Posting-Date: Wed, 21 Apr 1999 15:07:13 EDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newspump.monmouth.com!newspeer.monmouth.com!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!ftit

In article <7fk9i4$spj$1@ligarius.ultra.net>,  <jmfbahciv@aol.com> wrote:
>In article <Rf5T2.194$W4.62887@srvr1.engin.umich.edu>,
>>  [Unix, Multics and (yecch!) NT] [oh my!]
>Sigh!  And TOPS-10 was better than them.  

Please tell.  I know that this would constitute topic drift, but isn't
the Cardinal Rule of Usenet, "threads should be allowed to roam free"?
Besides, there exist people like me, who have never heard first-hand
accounts of life on TOPS-10 (sad but true), and would appreciate the
stories.

-- 
Sergej Roytman

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Thu, 22 Apr 99 08:55:47 GMT
Organization: UltraNet Communications, Inc.
Message-ID: <7fmua2$591$1@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <slrn7hrjii.dbl.roggblake@unix5.netaxs.com>
NNTP-Posting-Host: d6.dial-17.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 22 Apr 1999 10:40:02 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Lines: 23
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!feeder.qis.net!feed1.news.rcn.net!rcn!news.ultranet.com!d6

In article <slrn7hrjii.dbl.roggblake@unix5.netaxs.com>,
   roggblake@iname.com (Roger Blake) wrote:
>On Wed, 21 Apr 99 08:49:34 GMT, jmfbahciv@aol.com <jmfbahciv@aol.com> 
wrote:
>>Sigh!  And TOPS-10 was better than them.  
>
>That seems to be the basic point.  Microshaft had much prior art to draw
>upon in designing stable, efficient multitasking operating systems.  The
>problems and issues inherent in doing so had been worked out decades
>ago.  What's truly amazing is that with all of Unix's fundamental
>flaws and shortcomings (hilariously documented in the "Unix-Hater's
>Handbook"), it still looks like a stable, well-designed OS when put
>up against anything coming out of Redmond.
>
That's what happens when a COBOL programmer thinks he can do
OS implementations.  Or when a kid logs into a TOPS-10 system,
does a DIR, K/F and thinks he knows everything about that
system.  That attitude is stamped all over the software.
One can tell a lot by a programmer's style :-).

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Thu, 22 Apr 99 09:08:44 GMT
Organization: UltraNet Communications, Inc.
Lines: 28
Message-ID: <7fmv2b$591$2@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <BlpT2.3$MF5.1852@srvr1.engin.umich.edu>
NNTP-Posting-Host: d6.dial-17.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 22 Apr 1999 10:52:59 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!Sprint!netnews.com!feed1.news.rcn.net!rcn!news.ultranet.com!d6

In article <BlpT2.3$MF5.1852@srvr1.engin.umich.edu>,
   ftit@engin.umich.edu (Sergej Roytman) wrote:
>In article <7fk9i4$spj$1@ligarius.ultra.net>,  <jmfbahciv@aol.com> wrote:
>>In article <Rf5T2.194$W4.62887@srvr1.engin.umich.edu>,
>>>  [Unix, Multics and (yecch!) NT] [oh my!]
>>Sigh!  And TOPS-10 was better than them.  
>
>Please tell.  I know that this would constitute topic drift, but isn't
>the Cardinal Rule of Usenet, "threads should be allowed to roam free"?
>Besides, there exist people like me, who have never heard first-hand
>accounts of life on TOPS-10 (sad but true), and would appreciate the
>stories.
>
Since I was the designated den mother of the group, I rarely tell
technical stories (I don't feel I'm qualified).  I wrote and posted
one on Dec. 10 that told how the SMP project got started.

I can tell you that the "feel" of an SMP TOPS-10 system had the
fastest response time than any I'd ever used, including this
pitiful 486 that can't seem to handle one user, let alone 500.
In other words, one got his work done on a -10 in a reasonable
time.

What kind of stories were you interested in?

/BAH

Subtract a hundred and four for e-mail.

######

From: "Carl R. Friend" <carl.friend@stoneweb.com>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Thu, 22 Apr 1999 15:35:14 -0400
Organization: as little as possible!
Lines: 73
Message-ID: <371F79F1.69EB3101@stoneweb.com>
References: <371be3e4.0@newsfeed.one.net> <7fk9i4$spj$1@ligarius.ultra.net> <BlpT2.3$MF5.1852@srvr1.engin.umich.edu> <7fmv2b$591$2@antiochus.ultra.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu>
NNTP-Posting-Host: zephyr.ultranet.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 22 Apr 1999 19:35:16 GMT
X-Mailer: Mozilla 4.5 [en] (X11; I; Linux 2.0.29 i586)
X-Accept-Language: en
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newspump.monmouth.com!newspeer.monmouth.com!news.ultranet.com!not-for-mail

Sergej Roytman wrote:
> 
> In article <7fmv2b$591$2@antiochus.ultra.net>,  <jmfbahciv@aol.com>
> wrote:
> >What kind of stories were you interested in?
> 
> Well, since this is a folklore newsgroup, perhaps something about what
> life under TOPS-10 was like?

   Life with the PDP-10 monitor was quite enjoyable. The monitor was
there to give you the tools you needed to perform your tasks (whatever
they might be) without the bloat one finds under so-called "modern"
OSes. It's just the basic stuff -- call programs and compilers into
execution, edit and modify source code, run required jobs, examine/
deposit memory, save core, and the like.
 
> How many users?

   A very large system might support up to 150 simultaneous users.
The nice thing was, though, you didn't really notice the other users;
response time was usually very good.

> What kinds of things did they do?

   All the same sorts of things people do with computers now, with
the exception of "fluffy" things like clip-art and cutesy bitmaps.

> High-level languages, or was that beneath any programmer who
> mattered?

   Lots of top-flight programmers wrote in languages like FORTRAN,
and ALGOL. A whole host of other languages were also available.

> Did people accumulate personal collections of shell scripts to make
> life easier?

   With the advent of MIC, one could easily "script" with the -10
monitor. So, quite likely, people would collect these.

> play games?

   Ever hear of "ADVENT"? "HAUNT"? "DECWAR"? Games were a large part
of the existance of a "-10 head".

> write goofy comments late at night like the tunafish one in
> the BSD manual?

   Or, "Step into a phone booth"? (for invoking kernel mode)

> email each other, even when they were sitting at the
> next terminal?

   The other individual might be at the next terminal 12 hours from
now.... To be honest, I don't recall e-mail in the -10 world. Did
it exist?

> Was there a TOPS-10 culture?

   Absolutely.

   I'll allow others who were more intimately involved with the
development of the monitor expound on your other questions. There
are still lots of people around who worked in the lab from whence
"TOPS-10" sprang who know all of this stuff first hand. Me? I just
fixed the hardware to pay the rent. :-)

-- 
+------------------------------------------------+---------------------+
| Carl Richard Friend (UNIX Sysadmin)            | West Boylston       |
| Minicomputer Collector / Enthusiast            | Massachusetts, USA  |
| mailto:carl.friend@stoneweb.com                |                     |
| http://www.ultranet.com/~crfriend/museum       | ICBM: N42:22 W71:47 |
+------------------------------------------------+---------------------+

######

Sender: marc@dumbcat.snafu.org
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <BlpT2.3$MF5.1852@srvr1.engin.umich.edu> <7fmv2b$591$2@antiochus.ultra.net>
From: Marco S Hyman <marc@snafu.org>
Date: 22 Apr 1999 15:45:18 -0700
Message-ID: <x7hfq8w9z5.fsf@dumbcat.snafu.org>
Organization: S.N.A.F.U. (www.snafu.org)
X-Newsreader: Gnus v5.5/Emacs 20.3
Lines: 24
NNTP-Posting-Host: 204.94.187.130
X-Trace: nntp1.ba.best.com 924821119 214 marc@204.94.187.130
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!uunet!ams.uu.net!ffx.uu.net!in5.uu.net!korova.insync.net!feed1.news.rcn.net!rcn!news2.best.com!news.idiom.com!news3.best.com!nntp1.ba.best.com!not-for-mail

jmfbahciv@aol.com writes:

> I can tell you that the "feel" of an SMP TOPS-10 system had the
> fastest response time than any I'd ever used, including this
> pitiful 486 that can't seem to handle one user, let alone 500.
> In other words, one got his work done on a -10 in a reasonable
> time.

Too subjective.  The Sparcstation 5 I'm using today seemed fast
when I first got it.  Had I stopped using it a year ago my memories
would be that it was reasonable.  Today I think it's slow and am
already planning to making it a secondary, single use machine and
replace it with something "faster".

This has been going on as long as I've had machines on my desk.
I remember when the 4 MHz CP/M machine was blazing fast -- at
least when compared to the 2 MHz machine it replaced.

As for UNIX -> NT... I'm still of the believe that applications
drive hardware/os sales.  As a software type I prefer UNIX because
it comes with most of the tools I need and the source.  My wife
has no need for any of those tool, nor the source.  She uses Win 95.

// marc

######

Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7fk9i4$spj$1@ligarius.ultra.net> <BlpT2.3$MF5.1852@srvr1.engin.umich.edu> <7fmv2b$591$2@antiochus.ultra.net>
Organization: University of Michigan, College of Engineering
From: ftit@engin.umich.edu (Sergej Roytman)
Lines: 37
Message-ID: <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu>
Date: Thu, 22 Apr 1999 17:19:53 GMT
NNTP-Posting-Host: 141.212.106.44
X-Trace: srvr1.engin.umich.edu 924801593 141.212.106.44 (Thu, 22 Apr 1999 13:19:53 EDT)
NNTP-Posting-Date: Thu, 22 Apr 1999 13:19:53 EDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!newsxfer.itd.umich.edu.MISMATCH!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!ftit

In article <7fmv2b$591$2@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
>What kind of stories were you interested in?

Well, since this is a folklore newsgroup, perhaps something about what
life under TOPS-10 was like?

Since TOPS-10 comes from the time before small computers on everybody's
desk, I assume that it ran on machines called The Computer (maybe Our
Computer, by people who liked to hack on it), and supported non-trivial
numbers of users.  From your stories in this newsgroup, it sounds like
TOPS-10 was interactive- rather than batch-oriented.

How many users?  What kinds of things did they do?  High-level
languages, or was that beneath any programmer who mattered?  Did people
accumulate personal collections of shell scripts to make life easier?
play games? write goofy comments late at night like the tunafish one in
the BSD manual? email each other, even when they were sitting at the
next terminal?  Was there a TOPS-10 culture?

Actually, since you seem to disagree with even my multician friend's
opinion that Multics was the best thing since bit-sliced ALUs, I
wondered why.  Her justifications were completeness, security,
reliability and lack of panicking, stuff like that (also, something
called Xadventure).  You worked on TOPS-10 and like it for that reason;
any other reasons?

Speaking for myself, my upbringing has been very Unix-centered.
Interactive model for users; text-centered user interface, pretty much
assuming a CRT terminal of some kind (though things like e/ed/ex date
from before those became common); a bit of a minicomputer outlook on
life as well, I think, with things like out-of-control forking ignored
because hey, someone can always come over and give the machine a boot
if the process table gets filled up and it wedges badly.  TOPS-10 was
different in some ways.  I was curious about how.

-- 
Sergej Roytman

######

From: Robert Billing <unclebob@tnglwood.demon.co.uk>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 23 Apr 1999 08:33:30 +0000
Organization: Tanglewood
Message-ID: <3720305A.C0167D93@tnglwood.demon.co.uk>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <slrn7hrjii.dbl.roggblake@unix5.netaxs.com>
NNTP-Posting-Host: localhost.demon.co.uk
X-NNTP-Posting-Host: tnglwood.demon.co.uk:158.152.132.30
X-Trace: news.demon.co.uk 924857112 nnrp-09:15771 NO-IDENT tnglwood.demon.co.uk:158.152.132.30
X-Complaints-To: abuse@demon.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.0.31 i586)
Lines: 13
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!dispose.news.demon.net!demon!news.demon.co.uk!demon!tnglwood.demon.co.uk!not-for-mail

Roger Blake wrote:

> That seems to be the basic point.  Microshaft had much prior art to draw
> upon in designing stable, efficient multitasking operating systems.  The

 Exactly. When DOS hit the 32M hard drive problem, I spent several days
wandering around saying, "This was fixed on the PDP-11 *years* ago".

-- 
I am Robert Billing, Christian, inventor, traveller, cook and animal
lover, I live near 0:46W 51:22N.  http://www.tnglwood.demon.co.uk/
"Bother," said Pooh, "Eeyore, ready two photon torpedoes and lock
phasers on the Heffalump, Piglet, meet me in transporter room three"

######

From: rob.nicholson@zetnet.co.uk (Rob Nicholson)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 23 Apr 1999 08:38:38 GMT
Organization: Rainow Systems Ltd.
Reply-To: rob.nicholson@zetnet.co.uk
Message-ID: <37222d81.1058246@news.zetnet.co.uk>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <slrn7hrjii.dbl.roggblake@unix5.netaxs.com>
X-Newsreader: Forte Agent 1.5/32.452
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Lines: 6
NNTP-Posting-Host: manr-013.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 924856727 15922 194.247.43.143
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!newsfeed.nacamar.de!newsfeed.nacamar.de!ayres.ftech.net!news.ftech.net!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail

>upon in designing stable, efficient multitasking operating systems.

I thought the NT kernel was designed by one of the primary VMS designers? Wasn't
this a good pedigree?

Rob.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 23 Apr 99 08:53:05 GMT
Organization: UltraNet Communications, Inc.
Lines: 21
Message-ID: <7fpih9$v7t$1@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <BlpT2.3$MF5.1852@srvr1.engin.umich.edu> <7fmv2b$591$2@antiochus.ultra.net> <x7hfq8w9z5.fsf@dumbcat.snafu.org>
NNTP-Posting-Host: d10.dial-17.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 23 Apr 1999 10:37:29 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d10

In article <x7hfq8w9z5.fsf@dumbcat.snafu.org>,
   Marco S Hyman <marc@snafu.org> wrote:
>jmfbahciv@aol.com writes:
>
>> I can tell you that the "feel" of an SMP TOPS-10 system had the
>> fastest response time than any I'd ever used, including this
>> pitiful 486 that can't seem to handle one user, let alone 500.
>> In other words, one got his work done on a -10 in a reasonable
>> time.
>
>Too subjective.  

<snip>

Of course it is subjective.  On the other hand, I was real
good at it :-).  I usually knew when the system was in trouble,
going to crash, etc.  

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 23 Apr 99 08:55:31 GMT
Organization: UltraNet Communications, Inc.
Lines: 17
Message-ID: <7fpils$v7t$2@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <slrn7hrjii.dbl.roggblake@unix5.netaxs.com> <3720305A.C0167D93@tnglwood.demon.co.uk>
NNTP-Posting-Host: d10.dial-17.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 23 Apr 1999 10:39:56 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!netnews.com!newsswitch.lcs.mit.edu!news.ultranet.com!d10

In article <3720305A.C0167D93@tnglwood.demon.co.uk>,
   Robert Billing <unclebob@tnglwood.demon.co.uk> wrote:
>Roger Blake wrote:
>
>> That seems to be the basic point.  Microshaft had much prior art to draw
>> upon in designing stable, efficient multitasking operating systems.  The
>
> Exactly. When DOS hit the 32M hard drive problem, I spent several days
>wandering around saying, "This was fixed on the PDP-11 *years* ago".
>
That seems to be a habit (we used to call it NIH).  It's what
our ANF-10 guys would say whenever they attended DECnet design
reviews.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 23 Apr 99 08:56:12 GMT
Organization: UltraNet Communications, Inc.
Lines: 12
Message-ID: <7fpin4$v7t$3@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <slrn7hrjii.dbl.roggblake@unix5.netaxs.com> <37222d81.1058246@news.zetnet.co.uk>
NNTP-Posting-Host: d10.dial-17.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 23 Apr 1999 10:40:36 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!newspump.monmouth.com!newspeer.monmouth.com!news.ultranet.com!d10

In article <37222d81.1058246@news.zetnet.co.uk>,
   rob.nicholson@zetnet.co.uk (Rob Nicholson) wrote:
>>upon in designing stable, efficient multitasking operating systems.
>
>I thought the NT kernel was designed by one of the primary 
>VMS designers? Wasn't this a good pedigree?

Exactly.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 23 Apr 99 09:09:50 GMT
Organization: UltraNet Communications, Inc.
Lines: 86
Message-ID: <7fpjgn$v7t$4@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7fk9i4$spj$1@ligarius.ultra.net> <BlpT2.3$MF5.1852@srvr1.engin.umich.edu> <7fmv2b$591$2@antiochus.ultra.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu> <371F79F1.69EB3101@stoneweb.com>
NNTP-Posting-Host: d10.dial-17.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 23 Apr 1999 10:54:15 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newspeer.monmouth.com!news.ultranet.com!d10

In article <371F79F1.69EB3101@stoneweb.com>,
   "Carl R. Friend" <carl.friend@stoneweb.com> wrote:
>Sergej Roytman wrote:
>> 
>> In article <7fmv2b$591$2@antiochus.ultra.net>,  <jmfbahciv@aol.com>
>> wrote:
>> >What kind of stories were you interested in?
>> 
>> Well, since this is a folklore newsgroup, perhaps something about what
>> life under TOPS-10 was like?
>
>   Life with the PDP-10 monitor was quite enjoyable. The monitor was
>there to give you the tools you needed to perform your tasks (whatever
>they might be) without the bloat one finds under so-called "modern"
>OSes. It's just the basic stuff -- call programs and compilers into
>execution, edit and modify source code, run required jobs, examine/
>deposit memory, save core, and the like.
> 
>> How many users?
>
>   A very large system might support up to 150 simultaneous users.

It depended on how many CPUs were running.  Our 5-CPU site had
a JOBMAX of 500.  To accomodate their users they hacked LOGIN to
maintain a "waiting-for-the-next-open-job-slot" queue.  It was
always occupied.  Our in-house monitors were usually built, IIRC,
for 100 or 120 jobs but we never ran to the max under normal
timesharing.

>The nice thing was, though, you didn't really notice the other users;
>response time was usually very good.

Yup.  That's my point.  For all intents and purposes, one had
the feel of their very own private computer.

>
>> What kinds of things did they do?
>
>   All the same sorts of things people do with computers now, with
>the exception of "fluffy" things like clip-art and cutesy bitmaps.

Well, if we were still in business, I'm sure the fluff would have
been written :-).

<snip>


>> write goofy comments late at night like the tunafish one in
>> the BSD manual?
>
>   Or, "Step into a phone booth"? (for invoking kernel mode)

Until we acquired "professional" editors, there was all kinds
of humor in the programs and the docs.  The early DDT 
documentation had the chemical formula in it.  Later on, the
guys would "hide" some humor in the source code (TOPS-10 did
not review our sources for proper English usage).
>
>> email each other, even when they were sitting at the
>> next terminal?
>
>   The other individual might be at the next terminal 12 hours from
>now.... To be honest, I don't recall e-mail in the -10 world. Did
>it exist?

Yea, it did.  But that was a mess.  The manager "in charge" of
the program was using it to build his empire and his people
never wrote it--he just wanted to sell it for an obscene amount
of money.  Do you remember GRIPE?

>
>> Was there a TOPS-10 culture?
>
>   Absolutely.
>
>   I'll allow others who were more intimately involved with the
>development of the monitor expound on your other questions. There
>are still lots of people around who worked in the lab from whence
>"TOPS-10" sprang who know all of this stuff first hand. Me? I just
>fixed the hardware to pay the rent. :-)
>
Ah, but it's our customers who should be telling their stories, too.

/BAH

Subtract a hundred and four for e-mail.

######

From: dpeschel@u.washington.edu (Derek Peschel)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 23 Apr 1999 09:24:17 GMT
Organization: University of Washington, Seattle
Lines: 20
Message-ID: <7fpe81$7i8$1@nntp1.u.washington.edu>
References: <371be3e4.0@newsfeed.one.net> <7fk9i4$spj$1@ligarius.ultra.net> <slrn7hrjii.dbl.roggblake@unix5.netaxs.com> <37222d81.1058246@news.zetnet.co.uk>
NNTP-Posting-Host: saul5.u.washington.edu
X-Trace: nntp1.u.washington.edu 924859457 7752 (None) 140.142.17.40
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: dpeschel
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.berkeley.edu!news.u.washington.edu!dpeschel

In article <37222d81.1058246@news.zetnet.co.uk>,
Rob Nicholson <rob.nicholson@zetnet.co.uk> wrote:

>I thought the NT kernel was designed by one of the primary VMS designers? Wasn't
>this a good pedigree?

Yes, but [you probably realized this word _had_ to be "but"] a Win32
subsystem run on top of the kernel, entirely translates the kernel's
worldview into its own, adds tons of additional bulk, hides the services the
kernel itself provides, and thereby undoes the kernel designers' work.

Then other Win32 APIs run on the Win32 subsystem, and Win32 programs use
those APIs as well as doing their own dirty tricks.  The final result is
much better than Win95 but not nearly as good as it could be.

There is also a POSIX subsystem but I don't know much about it.  Apparently
it was there to satisfy the POSIX people, but honors POSIX more in the
breach than the observance.

-- Derek

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 23 Apr 99 11:23:14 GMT
Organization: UltraNet Communications, Inc.
Lines: 140
Message-ID: <7fprat$tjd$1@ligarius.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7fk9i4$spj$1@ligarius.ultra.net> <BlpT2.3$MF5.1852@srvr1.engin.umich.edu> <7fmv2b$591$2@antiochus.ultra.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu>
NNTP-Posting-Host: d12.dial-12.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 23 Apr 1999 13:07:41 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!outgoing.news.rcn.net.MISMATCH!feed1.news.rcn.net!rcn!news.ultranet.com!d12

In article <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu>,
   ftit@engin.umich.edu (Sergej Roytman) wrote:
>In article <7fmv2b$591$2@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
>>What kind of stories were you interested in?
>
>Well, since this is a folklore newsgroup, perhaps something about what
>life under TOPS-10 was like?
>
>Since TOPS-10 comes from the time before small computers on everybody's
>desk, I assume that it ran on machines called The Computer (maybe Our
>Computer, by people who liked to hack on it), and supported non-trivial
>numbers of users.  

Actually we called them machines and the room where they resided
was the machine room.  The number of users a system could support
was a variable called JOBMAX settable at MONGEN time.  MONGEN
was the program that took all the parameters and GENned the
appropriate values.  A limitation of TOPS-10 was that a lot of
its data structures were not expandable at boottime.  So a 
customer site had to make a lot of minimum, maximum and 
configuration decisions before the monitor was built.  Since 
our philosophy was to ship sources (the real documentation) and 
documentation which translated the code to human readable text.
About 98% of the code we shipped was in MACRO-10 which was
an assembler; since one could generate code using
macros, it wasn't strictly a pure machine language (from my POV).


>From your stories in this newsgroup, it sounds like
>TOPS-10 was interactive- rather than batch-oriented.

It was very interactive.  We also had software that managed
batch jobs and spoolers.  The batch system allowed a user
to create a control file that would work as if a user had 
logged into the system and used it.  This eliminated the
usual typos and gave a user the option of working while
s/he was asleep or doing other work.
>
>How many users? 

We had sites that built a monitor to support a minimum number of
jobs; their access control file could contain thousands of user
IDs which we called PPNs, short for project, programmer numbers.
In the 80s, the largest system had a maximum number of 500 jobs
that could be logged in at the same time.  Now, this was jobs,
not users.  In my work, I personally would have anywhere from
3-10 jobs logged in at once (I did a lot of stuff when I worked).


> What kinds of things did they do?  High-level
>languages, or was that beneath any programmer who mattered? 

These users could do anything they wanted.  We shipped quite a
number of languages and we had customers who wrote (or used) their
own developments.  Some languages that we supported was COBOL,
BASIC, FORTRAN, DBMS, RMS, BLISS [emoticon throwing up].  I'm 
trying to remember others but can't.  There was that group that
sat over there in the corner who was working on something....
fade to mists of forgetfulness....

> Did people
>accumulate personal collections of shell scripts to make life easier?

Yup.  I developed a whole lot of them as tools to help maintain
sources.  We had to keep track and ensure integrity of many 
versions at once.  Source management was a real headache.  I
keep getting the feeling that lack of source management is one
of the things at the heart of the crap Microsoft ships; it
sure has the feel of it anyway.

>play games? write goofy comments late at night like the tunafish one in
>the BSD manual? email each other, even when they were sitting at the
>next terminal?

Yes.  There was also a SEND TTYnnn command that would send
the following characters terminated by a <CRLF> to that 
device.  But that was real time communication.

>  Was there a TOPS-10 culture?

Yes.  Not only was there a TOPS-10 culture, there was a TOPS-20
culture and underlying all of that was a PDP-10 culture.

>
>Actually, since you seem to disagree with even my multician friend's
>opinion that Multics was the best thing since bit-sliced ALUs, I
>wondered why.

From the few glimmerings I've read (I never met a Multics, much
to my regret), it sounds like the TOPS-10 implementation of disk
I/O was more efficient and robust.  The philosophy of I/O 
implemented in UNIX is distasteful to me.  I preferred the
-10s philosophy.

>  Her justifications were completeness, security,
>reliability and lack of panicking, stuff like that (also, something
>called Xadventure).  You worked on TOPS-10 and like it for that reason;
>any other reasons?

Independence.  Even as a non-privileged user, I could do a lot
of things that other OS implementations wouldn't let me do, and,
more importantly, with JMF's and TW's implementation of SMP,
I usually didn't have to wait for the computer to get them done.
There is nothing more irritating than to have to wait for a system
to finish exiting from an editor.  I also prefer OSes who do
not make decisions for me or make the assumption that it knows
better than I do.  I want an OS to do what I tell it to _when_
I tell it to and immediately respond with an "OK, it's done",
giving me the assurance that I don't have to think about the
task anymore.  When I've finished something, I really, really
do want it in iron, not on a queue.
  
>
>Speaking for myself, my upbringing has been very Unix-centered.
>Interactive model for users; text-centered user interface, pretty much
>assuming a CRT terminal of some kind (though things like e/ed/ex date
>from before those became common); a bit of a minicomputer outlook on
>life as well, I think, with things like out-of-control forking ignored
>because hey, someone can always come over and give the machine a boot
>if the process table gets filled up and it wedges badly. 

We would have considered that a major bug and would have never
shipped a new version if unfixed.

> TOPS-10 was
>different in some ways.  I was curious about how.

Now you're getting into technical areas that I don't feel
comfortable talking about since I didn't do the implementation
work.  Unfortunately, there only a few who do know that
habitually read this newsgroup and there are even fewer who
can take the time to teach it.  It's an aspect of having the
work disappear off the face of the earth [sad emoticon here].

/BAH




Subtract a hundred and four for e-mail.

######

From: roggblake@iname.com ()
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 23 Apr 1999 12:42:34 GMT
Lines: 14
Message-ID: <7fpprq$fps$1@autumn.news.rcn.net>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <slrn7hrjii.dbl.roggblake@unix5.netaxs.com> <37222d81.1058246@news.zetnet.co.uk>
X-Trace: KxmOl0YSAWbjw7hFOC/bgPcJLP5X4yto8HO7OAhgzbc=
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 23 Apr 1999 12:42:34 GMT
X-No-Archive:  Yes
User-Agent:  slrn/0.9.5.4 (Windows)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!master.news.rcn.net!roggblake

On Fri, 23 Apr 1999 08:38:38 GMT, Rob Nicholson <rob.nicholson@zetnet.co.uk> wrote:
>I thought the NT kernel was designed by one of the primary VMS designers? Wasn't
>this a good pedigree?

Yes, Microsoft hired away DEC's chief operating system designer (Dave
Cutler) to design NT. Having seen some of his code from RSX11M and VMS
sources I'm reasonably certain that he is not the one responsible for
NT's problems.  More likely it's the crap Microsoft has layered on top
of his basic designs, such as moving video functions into the kernel so
that a buggy display driver can bugcheck your enterprise server.

-- 
  Roger Blake
  (remove second "g" from address for email)

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 23 Apr 99 12:55:16 GMT
Organization: UltraNet Communications, Inc.
Lines: 77
Message-ID: <7fq0nd$v0l$1@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7fmv2b$591$2@antiochus.ultra.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu> <7fprat$tjd$1@ligarius.ultra.net> <OW_T2.21$MF5.8637@srvr1.engin.umich.edu>
NNTP-Posting-Host: d12.dial-14.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 23 Apr 1999 14:39:41 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newspeer.clara.net!news.clara.net!nntp.news.xara.net!xara.net!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d12

In article <OW_T2.21$MF5.8637@srvr1.engin.umich.edu>,
   ftit@engin.umich.edu (Sergej Roytman) wrote:
>In article <7fprat$tjd$1@ligarius.ultra.net>,  <jmfbahciv@aol.com> wrote:
>>Now you're getting into technical areas that I don't feel
>>comfortable talking about since I didn't do the implementation
>>work.  Unfortunately, there only a few who do know that
>>habitually read this newsgroup and there are even fewer who
>>can take the time to teach it.  It's an aspect of having the
>>work disappear off the face of the earth [sad emoticon here].
>
>Well, thank you for telling about the TOPS-10 world.  

I didn't even reach the tip of the iceburg with my post :-).

>There's a lot of
>questions I could still ask, of course, many of which are in fact
>technical. 

It never hurts to ask.  Usually, in this newsgroup, asking a question
get a whole lot of discussion going with myriads of useful 
thread drift.  The key to understanding is to get somebody to talk
about the philosophy of the implementation; of course, the 
philosophy part usually isn't specified (but can be deduced) 
because few people worked on more than one implementation.
JMF did TOPS-10, some flavor of UNIX, a shade of VMS, and
was the CPU software expert in DEC.  He would have enjoyed
telling people about it all, but he didn't read any newsgroups.
I doubt he even knew about them.

> But learning a little about life on a PDP-10 or other such
>venerable piece of hardware is also a most welcome experience.
>
>We seem to come from subtly different worlds, I think.  In your early
>career, a computer was a lump of iron that needed programming, and the
>interface was an assembly language with occasional dips into the
>hardware level.  

I actually started the higher level languages and mosied my way
into operating systems.

>In mine, the virtual machine with which one normally
>interacts has moved up a few levels; now there's generally an OS and a
>high-level language between the programmer and the hardware.  There is
>now a lot of prior art for a programmer to understand as well, and
>while I've found myself wandering in a large, well-designed subsystem
>and going, "wow!", you may remember when it was _being_ designed.

Something that doesn't seem to get communicated is that none of
this was a one-time design.  It was a matter of many projects
over a long period of time.  Learning was based on experience,
not books or classes.
>
>There is value in even a modern hacker finding out about the old days,
>I think.  Otherwise, well, otherwise you get Microsoft.  On an
>intangible level, though, knowing that it was possible to do something
>on old, limited hardware provides a certain inspiration to go beyond
>complaining about how slow one's box is and actually put in the effort
>to design something worthy of such a past.  

Just remember that it took a lot of work and time to get something
worthy.  And it didn't happen all at once.  The stuff that we
put out evolved.  I wouldn't want to run the software that we
shipped in the early 70s....even the late 70s...because of the
improvements that happened in the 80s.

>Not that I think that the
>golden age of computers is behind us or anything---I'm sure that there
>are truly interesting (i.e., not related to talking paper fasteners)
>things for my generation to do---but a respect for one's predecessors
>and their achievements can not be a bad thing.
>
I'd vote for knowledge rather than respect :-).  And there's a ton
of things that need to be done.  

/BAH

Subtract a hundred and four for e-mail.

######

Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7fmv2b$591$2@antiochus.ultra.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu> <7fprat$tjd$1@ligarius.ultra.net>
Organization: University of Michigan, College of Engineering
From: ftit@engin.umich.edu (Sergej Roytman)
Lines: 36
Message-ID: <OW_T2.21$MF5.8637@srvr1.engin.umich.edu>
Date: Fri, 23 Apr 1999 13:52:46 GMT
NNTP-Posting-Host: 141.212.106.44
X-Trace: srvr1.engin.umich.edu 924875566 141.212.106.44 (Fri, 23 Apr 1999 09:52:46 EDT)
NNTP-Posting-Date: Fri, 23 Apr 1999 09:52:46 EDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!newsxfer.itd.umich.edu.MISMATCH!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!ftit

In article <7fprat$tjd$1@ligarius.ultra.net>,  <jmfbahciv@aol.com> wrote:
>Now you're getting into technical areas that I don't feel
>comfortable talking about since I didn't do the implementation
>work.  Unfortunately, there only a few who do know that
>habitually read this newsgroup and there are even fewer who
>can take the time to teach it.  It's an aspect of having the
>work disappear off the face of the earth [sad emoticon here].

Well, thank you for telling about the TOPS-10 world.  There's a lot of
questions I could still ask, of course, many of which are in fact
technical.  But learning a little about life on a PDP-10 or other such
venerable piece of hardware is also a most welcome experience.

We seem to come from subtly different worlds, I think.  In your early
career, a computer was a lump of iron that needed programming, and the
interface was an assembly language with occasional dips into the
hardware level.  In mine, the virtual machine with which one normally
interacts has moved up a few levels; now there's generally an OS and a
high-level language between the programmer and the hardware.  There is
now a lot of prior art for a programmer to understand as well, and
while I've found myself wandering in a large, well-designed subsystem
and going, "wow!", you may remember when it was _being_ designed.

There is value in even a modern hacker finding out about the old days,
I think.  Otherwise, well, otherwise you get Microsoft.  On an
intangible level, though, knowing that it was possible to do something
on old, limited hardware provides a certain inspiration to go beyond
complaining about how slow one's box is and actually put in the effort
to design something worthy of such a past.  Not that I think that the
golden age of computers is behind us or anything---I'm sure that there
are truly interesting (i.e., not related to talking paper fasteners)
things for my generation to do---but a respect for one's predecessors
and their achievements can not be a bad thing.

-- 
Sergej Roytman

######

From: Robert Billing <unclebob@tnglwood.demon.co.uk>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 23 Apr 1999 14:27:10 +0000
Organization: Tanglewood
Message-ID: <3720833E.253A8CCD@tnglwood.demon.co.uk>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <slrn7hrjii.dbl.roggblake@unix5.netaxs.com> <37222d81.1058246@news.zetnet.co.uk>
NNTP-Posting-Host: localhost.demon.co.uk
X-NNTP-Posting-Host: tnglwood.demon.co.uk:158.152.132.30
X-Trace: news.demon.co.uk 924879262 nnrp-02:1258 NO-IDENT tnglwood.demon.co.uk:158.152.132.30
X-Complaints-To: abuse@demon.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.0.31 i586)
Lines: 18
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!tnglwood.demon.co.uk!not-for-mail

Rob Nicholson wrote:
> 
> >upon in designing stable, efficient multitasking operating systems.
> 
> I thought the NT kernel was designed by one of the primary VMS designers? Wasn't
> this a good pedigree?

 Yes. It's just been getting steadily worse on every subsequent release.

 At some risk of restarting the flamewar that we had last year, I have
now made NT drive this video card correctly. All I have to do is run it
under VMware emulation, and use the X driver.

-- 
I am Robert Billing, Christian, inventor, traveller, cook and animal
lover, I live near 0:46W 51:22N.  http://www.tnglwood.demon.co.uk/
"Bother," said Pooh, "Eeyore, ready two photon torpedoes and lock
phasers on the Heffalump, Piglet, meet me in transporter room three"

######

From: rob.nicholson@zetnet.co.uk (Rob Nicholson)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 23 Apr 1999 14:35:54 GMT
Organization: Rainow Systems Ltd.
Reply-To: rob.nicholson@zetnet.co.uk
Message-ID: <3721821d.6019996@news.zetnet.co.uk>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <slrn7hrjii.dbl.roggblake@unix5.netaxs.com> <37222d81.1058246@news.zetnet.co.uk> <7fpprq$fps$1@autumn.news.rcn.net>
X-Newsreader: Forte Agent 1.5/32.452
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
NNTP-Posting-Host: manr-014.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 924878166 16957 194.247.43.144
Lines: 7
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newspeer.clara.net!news.clara.net!dispose.news.demon.net!demon!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail

>NT's problems.  More likely it's the crap Microsoft has layered on top
>of his basic designs, such as moving video functions into the kernel so
>that a buggy display driver can bugcheck your enterprise server.

I've heard about this. What was the primary reason? Performance?

Rob.

######

From: tsmurphy@students.uiuc.edu (Terry Murphy)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 23 Apr 1999 15:22:39 GMT
Organization: University of Illinois at Urbana-Champaign
Lines: 25
Message-ID: <7fq37v$9dc$1@vixen.cso.uiuc.edu>
References: <371be3e4.0@newsfeed.one.net> <7fprat$tjd$1@ligarius.ultra.net> <OW_T2.21$MF5.8637@srvr1.engin.umich.edu> <7fq0nd$v0l$1@antiochus.ultra.net>
NNTP-Posting-Host: ux5.cso.uiuc.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!vixen.cso.uiuc.edu!tsmurphy

In article <7fq0nd$v0l$1@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:

>It never hurts to ask.  Usually, in this newsgroup, asking a question
>get a whole lot of discussion going with myriads of useful 
>thread drift.  The key to understanding is to get somebody to talk
>about the philosophy of the implementation; of course, the 
>philosophy part usually isn't specified (but can be deduced) 
>because few people worked on more than one implementation.
>JMF did TOPS-10, some flavor of UNIX, a shade of VMS, and
>was the CPU software expert in DEC.  He would have enjoyed
>telling people about it all, but he didn't read any newsgroups.
>I doubt he even knew about them.

What did TOPS-10 look/feel like?  Was it similar to the later DEC
operating systems such as the RSX-11 family and VMS?  (in terms of
commands available, languages used, file system, etc.)

What other pre-VMS OS'es did DEC do?  E.g. what ran on PDP-7 and
PDP-8?

Where does TOPS-10 live today? Is it still in use, and/or has it
influenced other systems?  Do TOPS-10 and -20 have anything to do with
each other?

-- Terry

######

Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu> <371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net>
Organization: University of Michigan, College of Engineering
From: ftit@engin.umich.edu (Sergej Roytman)
Lines: 25
Message-ID: <Ci0U2.24$MF5.9024@srvr1.engin.umich.edu>
Date: Fri, 23 Apr 1999 15:26:26 GMT
NNTP-Posting-Host: 141.212.106.44
X-Trace: srvr1.engin.umich.edu 924881186 141.212.106.44 (Fri, 23 Apr 1999 11:26:26 EDT)
NNTP-Posting-Date: Fri, 23 Apr 1999 11:26:26 EDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!sunqbc.risq.qc.ca!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!ftit

In article <7fpjgn$v7t$4@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
>It depended on how many CPUs were running.  Our 5-CPU site had
>a JOBMAX of 500.  To accomodate their users they hacked LOGIN to
>maintain a "waiting-for-the-next-open-job-slot" queue.  It was
>always occupied.  Our in-house monitors were usually built, IIRC,
>for 100 or 120 jobs but we never ran to the max under normal
>timesharing.

Since BAH herself encouraged me to do so in email, I'll ask about this
in the newsgroup.

The quoted paragraph seems to indicate that TOPS-10's (rough)
equivalents of Unix process table slots were allocated one-per-login.
Was this the case?  Other posts here talk about running and batching
multiple jobs while doing other stuff in the foreground.  What would
one do if his program wanted to start several processes---the TOPS-10
equivalent of a fork/spawn system call?  Did the shell start new
processes when users ran programs?  If not, was there no such thing as
Unix's roll-yer-own-shell phenomenon, which has resulted in so many
different levels of (mutually incompatible (oh well)) levels of support
from the shell, from opening Emacs to edit previous commands (almost)
to not even having a ! to redo the things?

-- 
Sergej Roytman, curious

######

From: roggblake@iname.com ()
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 23 Apr 1999 17:49:28 GMT
Lines: 22
Message-ID: <7fqbr8$74u$1@autumn.news.rcn.net>
References: <371be3e4.0@newsfeed.one.net> <7fprat$tjd$1@ligarius.ultra.net> <OW_T2.21$MF5.8637@srvr1.engin.umich.edu> <7fq0nd$v0l$1@antiochus.ultra.net> <7fq37v$9dc$1@vixen.cso.uiuc.edu>
X-Trace: OO90Kw2zbrRp47HQp9KkcyQtg5Z7P8IICkG7rM9WE9s=
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 23 Apr 1999 17:49:28 GMT
X-No-Archive:  Yes
User-Agent:  slrn/0.9.5.4 (Windows)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!howland.erols.net!master.news.rcn.net!roggblake

On 23 Apr 1999 15:22:39 GMT, Terry Murphy <tsmurphy@students.uiuc.edu> wrote:
>What did TOPS-10 look/feel like?  Was it similar to the later DEC

I never used TOPS-10, but did do some work with TOPS-20 at one time.
Loved the command-line user interface with its command completion and
context-sensitive help.  (A clone of this interface lives on in MS-DOS
Kermit, and was even used in Commodore 64 Kermit!)

>What other pre-VMS OS'es did DEC do?  E.g. what ran on PDP-7 and
>PDP-8?

The PDP-8 had a system called "OS-8."  I still have a copy of my
OS-8 handbook.  Unix was developed on a disused PDP-7 at Bell Labs...

>influenced other systems?  Do TOPS-10 and -20 have anything to do with
>each other?

Despite having worked with TOPS-20 in the past, I have no idea!

-- 
  Roger Blake
  (remove second "g" from address for email)

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
From: alderson@netcom.netcom.com (Richard M. Alderson III)
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
In-Reply-To: ftit@engin.umich.edu's message of Fri, 23 Apr 1999 15:26:26 GMT
Message-ID: <ALDERSON.99Apr23113531@netcom.netcom.com>
Sender: alderson@netcom.netcom.com
Reply-To: alderson@netcom.com
Organization: NETCOM On-line services
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu>
	<371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net>
	<Ci0U2.24$MF5.9024@srvr1.engin.umich.edu>
Date: Fri, 23 Apr 1999 18:35:31 GMT
Lines: 30
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!isdnet!newsfeed.usit.net!newsswitch.lcs.mit.edu!newsfeed.enteract.com!ix.netcom.com!netcom!alderson

In article <Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> ftit@engin.umich.edu
(Sergej Roytman) writes:

>Did the shell start new processes when users ran programs?  If not, was there
>no such thing as Unix's roll-yer-own-shell phenomenon, which has resulted in
>so many different levels of (mutually incompatible (oh well)) levels of
>support from the shell, from opening Emacs to edit previous commands (almost)
>to not even having a ! to redo the things?

First, stop thinking in terms of a shell; the command interpreter was part of
the monitor (= kernel) talking directly to the user's terminal.  (Since Tops-10
system terminals tended to be hard-wired, this was an easy model to follow; the
eventual network support had to deal with this reality.  Same thing was true of
Tops-20.)

The user typed commands to the monitor prompt; the command interpreter would
take appropriate action:  Load a program into the user's address space; start
it if requested to do so; run a program in the monitor's address space on the
user's behalf; and so on.  In later versions of Tops-10, a user could interrupt
a program, issue a monitor command to keep it, run another program, and then
resume the first (up to four forks); before that change, a user could only run
one program at a time.

I'll leave it to real Tops-10 hands to discuss this further; I do Tops-20, but
BAH talks to me anyway. ;-)
-- 
Rich Alderson   You know the sort of thing that you can find in any dictionary
                of a strange language, and which so excites the amateur philo-
                logists, itching to derive one tongue from another that they
                know better: a word that is nearly the same in form and meaning
                as the corresponding word in English, or Latin, or Hebrew, or
                what not.
                                                --J. R. R. Tolkien,
alderson@netcom.com                               _The Notion Club Papers_

######

From: Mark Crispin <mrc@CAC.Washington.EDU>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 23 Apr 1999 20:33:00 -0700
Organization: Networks & Distributed Computing
Lines: 51
Message-ID: <Pine.NXT.4.20.9904232023400.13242-100000@Tomobiki-Cho.CAC.Washington.EDU>
References: <371be3e4.0@newsfeed.one.net> <7fpjgn$v7t$4@antiochus.ultra.net> <Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu>
NNTP-Posting-Host: tomobiki-cho.cac.washington.edu
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Trace: nntp1.u.washington.edu 924924784 21648 (None) 140.142.17.38
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: dak
To: Alexander Viro <viro@weyl.math.psu.edu>
In-Reply-To: <7frcok$3ck@weyl.math.psu.edu>
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!netnews.com!newspeer.monmouth.com!netnews1.nw.verio.net!netnews.nwnet.net!news.verio.net!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc

On 23 Apr 1999, Alexander Viro wrote:
> Wait a bit. Do you mean that there was no scheduling within the same
> login session?

login session == process.

> Does it mean that big compile == sitting and waiting or going
> to other terminal?

Well, not necessarily.  A job could be detached from a terminal, and a new
one logged in.  So, you could get somewhat of the same effect of multiple
processes, by successively attaching and detaching your jobs.

Assuming that the computer center wasn't a bunch of fascist pigs that
didn't let you do this.

> Could the process create a new one, wait for its completion
> and resume when the child finishes (without touching the keyboard, that is)?

There was no such thing as "creating a process".

> From the paper on TOPS-20 I got an impression that it had background
> processes. Was it a difference between TOPS-10 and TOPS-20?

TOPS-20 was a completely different operating system.  It had entirely
different system calls (TOPS-10 system calls were emulated via a user mode
library) and an entirely different job/process mechanism.

In a great many ways, TOPS-20 was more sophisticated than UNIX is today.
TOPS-20 virtual memory is still far ahead of UNIX.  Damn, I miss it!!

TOPS-20 processes were somewhat like a hybrid between UNIX processes and
UNIX threads.  A TOPS-20 fork could run in its own address space or share
portions (or all) of its address space with another process.  However, a
lot of things (e.g. connected directory, access rights) were global to the
entire job tree, nor could a TOPS-20 process be detached from its job
tree.

> BTW, how did ITS behave?

ITS was more like UNIX in terms of its process handling.  Its virtual
memory handling was better than UNIX, but not as good as TOPS-20.  ITS had
no file protections or privileges on system calls; it was a "gentlemen's
timesharing".

-- Mark --

* RCW 19.190 notice: This email address is located in Washington State.	*
* Unsolicited commercial email may be billed $500 per message.		*
Science does not emerge from voting, party politics, or public debate.

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7fpjgn$v7t$4@antiochus.ultra.net> <Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com>
Organization: University of Michigan, College of Engineering
From: ftit@engin.umich.edu (Sergej Roytman)
Lines: 31
Message-ID: <Gs5U2.29$MF5.10448@srvr1.engin.umich.edu>
Date: Fri, 23 Apr 1999 21:18:30 GMT
NNTP-Posting-Host: 141.212.106.44
X-Trace: srvr1.engin.umich.edu 924902310 141.212.106.44 (Fri, 23 Apr 1999 17:18:30 EDT)
NNTP-Posting-Date: Fri, 23 Apr 1999 17:18:30 EDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cliffs.rs.itd.umich.edu!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!ftit

In article <ALDERSON.99Apr23113531@netcom.netcom.com>,
Richard M. Alderson III <alderson@netcom.com> wrote:
>The user typed commands to the monitor prompt; the command interpreter
 would
>take appropriate action:  Load a program into the user's address space
; start
>it if requested to do so; run a program in the monitor's address space
 on the
>user's behalf; and so on.  In later versions of Tops-10, a user could 
interrupt
>a program, issue a monitor command to keep it, run another program, an
d then
>resume the first (up to four forks); before that change, a user could 
only run
>one program at a time.

Hm, interesting.  So what the users saw was a virtual front-panel
machine: load your simulated core with a program, specify an entry
point, and hit run?  Things like (Unixoid---sorry) pipelines would have
been done with intermediate files, then, and peripherals would have a
fake front-end which would really talk to the appropriate queues and
buffers in the kernel.  And, the front-ends being in the kernel
themselves, they could stay more tightly coupled with the underlying
data structures and report more things to the user, hence BAH's stated
dislike for dumping data down a pipe and not knowing what happens to
it?

So, 'dI guess right?  Huh?

-- 
Sergej Roytman

######

From: "Ambrose, Joseph" <joseph.j.ambrose@worldnet.att.net>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 23 Apr 1999 21:53:46 -0400
Organization: AT&T WorldNet Services
Lines: 29
Message-ID: <7fr83e$oi4$1@bgtnsc02.worldnet.att.net>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fj0bj$48e$1@bcrkh13.ca.nortel.com>
NNTP-Posting-Host: 12.79.14.157
X-Trace: bgtnsc02.worldnet.att.net 924918702 25156 12.79.14.157 (24 Apr 1999 01:51:42 GMT)
X-Complaints-To: abuse@worldnet.att.net
NNTP-Posting-Date: 24 Apr 1999 01:51:42 GMT
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 5.00.2314.1300
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!wn4feed!worldnet.att.net!wnmaster1!not-for-mail


Clark <clarkg@xxxnortelnetworks.com> wrote in message
news:7fj0bj$48e$1@bcrkh13.ca.nortel.com...
> In article <Rf5T2.194$W4.62887@srvr1.engin.umich.edu>,
ftit@engin.umich.edu (Sergej Roytman) wrote:
>
> [snip]
>
> >I could also have brought up OS8 on the PDP-8, which I think managed
> >some sort of multi-userness in the standard 4kW of core.  But I've
> >never used this, and don't know about its history.
>
> I don't think OS/8 was multi-user. But schools had the EduSystem
> (BASIC) running multiple users on a PDP-8 with (mumble)kW of core.
>
I used Edusystem 20c in my HS days. I believe that EduSystem 50 WAS OS/8. I
certainly looked multi-user. Had an opportunity to use it as a visitor to
another HS. Definitely was major cool at that time.


--
Joseph Ambrose
NT Network Administrator /
Open VMS System Manager
The Conference Board
ambrose@conference-board.org
ICQ# 13652219

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 23 Apr 1999 23:11:16 -0400
Organization: -ENOENT
Lines: 30
Message-ID: <7frcok$3ck@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7fpjgn$v7t$4@antiochus.ultra.net> <Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!netnews.com!howland.erols.net!newsxfer.itd.umich.edu.MISMATCH!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <ALDERSON.99Apr23113531@netcom.netcom.com>,
Richard M. Alderson III <alderson@netcom.com> wrote:
>First, stop thinking in terms of a shell; the command interpreter was part of
>the monitor (= kernel) talking directly to the user's terminal.  (Since Tops-10
>system terminals tended to be hard-wired, this was an easy model to follow; the
>eventual network support had to deal with this reality.  Same thing was true of
>Tops-20.)
>
>The user typed commands to the monitor prompt; the command interpreter would
>take appropriate action:  Load a program into the user's address space; start
>it if requested to do so; run a program in the monitor's address space on the
>user's behalf; and so on.  In later versions of Tops-10, a user could interrupt
>a program, issue a monitor command to keep it, run another program, and then
>resume the first (up to four forks); before that change, a user could only run
>one program at a time.

	Wait a bit. Do you mean that there was no scheduling within the same
login session? Does it mean that big compile == sitting and waiting or going
to other terminal? Could the process create a new one, wait for its completion
and resume when the child finishes (without touching the keyboard, that is)?
From the paper on TOPS-20 I got an impression that it had background processes.
Was it a difference between TOPS-10 and TOPS-20? BTW, how did ITS behave?
If it was anything close to the model you've described... well, then I can see
why EMACS is such a bloated monolith...
	Life with ulimit -u 4 <shudder> Was there any equivalent of virtual
consoles?

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 24 Apr 1999 00:43:11 -0400
Organization: -ENOENT
Lines: 65
Message-ID: <7fri4v$3eq@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu> <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.axxsys.net!newsfeed.berkeley.edu!cliffs.rs.itd.umich.edu!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net>,
Ric Werme <werme@werme.ne.mediaone.net> wrote:
>viro@weyl.math.psu.edu (Alexander Viro) writes:
>>	Wait a bit. Do you mean that there was no scheduling within the same
>>login session? Does it mean that big compile == sitting and waiting or going
>>to other terminal?
>
>Yeah.  Hey, it sure beat waiting for the batch listing to print and then
>going back to the keypunch room.  Compiles weren't all that big - the
>linker was pretty atrocious (20 minutes to link a monitor on a KA), but
>when people relized the bubble sort was the culprit and fixed that then
	Gaack... I wonder wasn't it the reason why RSX-11 TKB sucked so hard.
Pascal was lighter than that monstrosity.
>it only took a minute or so.

> Could the process create a new one, wait for its completion
>>and resume when the child finishes (without touching the keyboard, that is)?
>
>Sort of.  We had something called CCL, for Concise Command Language, so you
>could say "debug foo" and it would create some command files, switch to the
>assembler, that would switch to the linker, and that would link it with DDT
	It's a bit different, unless you have a convenient way to drop the
process image into file and restore it later... Which has all sorts of nastiness
wrt the opened files, network connections, etc.

>and run it.  We thought that was pretty cool.  Nothing like a Makefile,
>though.  A close relative to the debug command (which we noted _didn't_
>debug your program), was make.  In fact, you could type "MAKE LOVE", and
	Wait a bit. Wasn't it a part of TECO that did document creation
(as opposed to editing)?

>make would sleep a second or so and print "NOT WAR?" and exit.  It _was_
>during the Vietnam war, after all....
>
>>	Life with ulimit -u 4 <shudder> Was there any equivalent of virtual
>>consoles?
>
>Well, you could detach from your terminal, I think with the program running,
>and later attach back to it.  I often worked in my office, detached,
>went to the lab, and attached back.  Can't do that with Unix!  Also, we

	Unless I severely misparsed you - yes, you can do exactly what
you described: man screen for details. It allows to detach the session,
do whatever you want (including logout), go back later (maybe after new
login/telnet from other host/whatever) and reattach it. Is it what you are
talking about?

>did have PTYs, you know, and you could run OPSER to manage several things
	If you are talking about the same thing as UNIX PTYs - yes, it can
make life easier, but allocating a PTY pair for each task... Oh, well...

>at once.  We could also stream reading from magtapes with no special buffering
>or multi-thread dd hacks, something Unix has never figure out, and something
>that could be handy for dealing with files too big for the cache today.
	Are you talking about IO to/from userspace or just about non-buffering
devices? Files too big for the cache are not a problem - caching happens below
the filesystem level. BTW, dd for stream reading from magtape is not needed
since *long* ago - driver does all buffering needed. IO to/from userspace may
be done, but you'll have to mlock() the pages in core - otherwise devices
will be pretty unhappy with the timings (pagein may be too long). How was it
dealt with on TOPS?

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 24 Apr 1999 00:49:37 -0400
Organization: -ENOENT
Lines: 26
Message-ID: <7frih1$3fb@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu> <Pine.NXT.4.20.9904232023400.13242-100000@Tomobiki-Cho.CAC.Washington.EDU>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!newsxfer.itd.umich.edu.MISMATCH!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <Pine.NXT.4.20.9904232023400.13242-100000@Tomobiki-Cho.CAC.Washington.EDU>,
Mark Crispin  <mrc@CAC.Washington.EDU> wrote:
>On 23 Apr 1999, Alexander Viro wrote:
>> From the paper on TOPS-20 I got an impression that it had background
>> processes. Was it a difference between TOPS-10 and TOPS-20?
>
>TOPS-20 was a completely different operating system.  It had entirely
>different system calls (TOPS-10 system calls were emulated via a user mode
>library) and an entirely different job/process mechanism.
>
>In a great many ways, TOPS-20 was more sophisticated than UNIX is today.
>TOPS-20 virtual memory is still far ahead of UNIX.  Damn, I miss it!!

	What about details of TOPS-20 VM? Which UNIX, BTW? VM is different
in different branches...

>TOPS-20 processes were somewhat like a hybrid between UNIX processes and
>UNIX threads.  A TOPS-20 fork could run in its own address space or share
>portions (or all) of its address space with another process.  However, a

	Aha... Was it along the lines of shared mmap() or something totally
different?

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: Will Rose <cwr@cts.com>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 24 Apr 1999 02:17:16 GMT
Organization: CTS Network Services
Message-ID: <7fr9jc$jpm$1@nusku.cts.com>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <slrn7hrjii.dbl.roggblake@unix5.netaxs.com> <37222d81.1058246@news.zetnet.co.uk> <7fpprq$fps$1@autumn.news.rcn.net>
NNTP-Posting-Host: crash-i2.cts.com
X-Trace: nusku.cts.com 924920236 20278 205.163.0.7 (24 Apr 1999 02:17:16 GMT)
X-Complaints-To: usenet@nusku.cts.com
NNTP-Posting-Date: 24 Apr 1999 02:17:16 GMT
User-Agent: tin/pre-1.4-19990413 ("Endemoniada") (UNIX) (crash/3.2 (i386))
Lines: 19
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!news.idt.net!nntp2.cerf.net!mercury.cts.com!nusku.cts.com!not-for-mail

roggblake@iname.com wrote:
: On Fri, 23 Apr 1999 08:38:38 GMT, Rob Nicholson <rob.nicholson@zetnet.co.uk> wrote:
:>I thought the NT kernel was designed by one of the primary VMS designers? Wasn't
:>this a good pedigree?

: Yes, Microsoft hired away DEC's chief operating system designer (Dave
: Cutler) to design NT. Having seen some of his code from RSX11M and VMS
: sources I'm reasonably certain that he is not the one responsible for
: NT's problems.  More likely it's the crap Microsoft has layered on top
: of his basic designs, such as moving video functions into the kernel so
: that a buggy display driver can bugcheck your enterprise server.

Cutler certainly considered a GUI an un-necessary frill for NT.  It was
added pretty late in the day, and I doubt it got his full attention.


Will
cwr@crash.cts.com

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7fpjgn$v7t$4@antiochus.ultra.net>
     <Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu>
X-Newsreader: NN version 6.5.0 CURRENT #119
From: werme@werme.ne.mediaone.net (Ric Werme)
Lines: 45
Message-ID: <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net>
Date: Sat, 24 Apr 1999 03:42:45 GMT
NNTP-Posting-Host: 24.128.109.10
X-Trace: wbnws01.ne.mediaone.net 924925365 24.128.109.10 (Fri, 23 Apr 1999 23:42:45 EDT)
NNTP-Posting-Date: Fri, 23 Apr 1999 23:42:45 EDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!netnews.com!chnws02.mediaone.net!24.128.1.101!chnws05.ne.mediaone.net!24.128.44.7!wbnws01.ne.mediaone.net.POSTED!not-for-mail

viro@weyl.math.psu.edu (Alexander Viro) writes:

>In article <ALDERSON.99Apr23113531@netcom.netcom.com>,
>Richard M. Alderson III <alderson@netcom.com> wrote:
>>First, stop thinking in terms of a shell; the command interpreter was part of
>>the monitor (= kernel) talking directly to the user's terminal.

>	Wait a bit. Do you mean that there was no scheduling within the same
>login session? Does it mean that big compile == sitting and waiting or going
>to other terminal?

Yeah.  Hey, it sure beat waiting for the batch listing to print and then
going back to the keypunch room.  Compiles weren't all that big - the
linker was pretty atrocious (20 minutes to link a monitor on a KA), but
when people relized the bubble sort was the culprit and fixed that then
it only took a minute or so.

 Could the process create a new one, wait for its completion
>and resume when the child finishes (without touching the keyboard, that is)?

Sort of.  We had something called CCL, for Concise Command Language, so you
could say "debug foo" and it would create some command files, switch to the
assembler, that would switch to the linker, and that would link it with DDT
and run it.  We thought that was pretty cool.  Nothing like a Makefile,
though.  A close relative to the debug command (which we noted _didn't_
debug your program), was make.  In fact, you could type "MAKE LOVE", and
make would sleep a second or so and print "NOT WAR?" and exit.  It _was_
during the Vietnam war, after all....

>	Life with ulimit -u 4 <shudder> Was there any equivalent of virtual
>consoles?

Well, you could detach from your terminal, I think with the program running,
and later attach back to it.  I often worked in my office, detached,
went to the lab, and attached back.  Can't do that with Unix!  Also, we
did have PTYs, you know, and you could run OPSER to manage several things
at once.  We could also stream reading from magtapes with no special buffering
or multi-thread dd hacks, something Unix has never figure out, and something
that could be handy for dealing with files too big for the cache today.

	-Ric Werme
--
Ric Werme                  | http://people.ne.mediaone.net/werme
werme@nospam.mediaone.net  | http://www.cyberportal.net/werme
      ^^^^^^^ delete

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7fpjgn$v7t$4@antiochus.ultra.net>
     <Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu>
X-Newsreader: NN version 6.5.0 CURRENT #119
From: werme@werme.ne.mediaone.net (Ric Werme)
Lines: 45
Message-ID: <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net>
Date: Sat, 24 Apr 1999 03:42:45 GMT
NNTP-Posting-Host: 24.128.109.10
X-Trace: wbnws01.ne.mediaone.net 924925365 24.128.109.10 (Fri, 23 Apr 1999 23:42:45 EDT)
NNTP-Posting-Date: Fri, 23 Apr 1999 23:42:45 EDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!netnews.com!chnws02.mediaone.net!24.128.1.101!chnws05.ne.mediaone.net!24.128.44.7!wbnws01.ne.mediaone.net.POSTED!not-for-mail

viro@weyl.math.psu.edu (Alexander Viro) writes:

>In article <ALDERSON.99Apr23113531@netcom.netcom.com>,
>Richard M. Alderson III <alderson@netcom.com> wrote:
>>First, stop thinking in terms of a shell; the command interpreter was part of
>>the monitor (= kernel) talking directly to the user's terminal.

>	Wait a bit. Do you mean that there was no scheduling within the same
>login session? Does it mean that big compile == sitting and waiting or going
>to other terminal?

Yeah.  Hey, it sure beat waiting for the batch listing to print and then
going back to the keypunch room.  Compiles weren't all that big - the
linker was pretty atrocious (20 minutes to link a monitor on a KA), but
when people relized the bubble sort was the culprit and fixed that then
it only took a minute or so.

 Could the process create a new one, wait for its completion
>and resume when the child finishes (without touching the keyboard, that is)?

Sort of.  We had something called CCL, for Concise Command Language, so you
could say "debug foo" and it would create some command files, switch to the
assembler, that would switch to the linker, and that would link it with DDT
and run it.  We thought that was pretty cool.  Nothing like a Makefile,
though.  A close relative to the debug command (which we noted _didn't_
debug your program), was make.  In fact, you could type "MAKE LOVE", and
make would sleep a second or so and print "NOT WAR?" and exit.  It _was_
during the Vietnam war, after all....

>	Life with ulimit -u 4 <shudder> Was there any equivalent of virtual
>consoles?

Well, you could detach from your terminal, I think with the program running,
and later attach back to it.  I often worked in my office, detached,
went to the lab, and attached back.  Can't do that with Unix!  Also, we
did have PTYs, you know, and you could run OPSER to manage several things
at once.  We could also stream reading from magtapes with no special buffering
or multi-thread dd hacks, something Unix has never figure out, and something
that could be handy for dealing with files too big for the cache today.

	-Ric Werme
--
Ric Werme                  | http://people.ne.mediaone.net/werme
werme@nospam.mediaone.net  | http://www.cyberportal.net/werme
      ^^^^^^^ delete

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sat, 24 Apr 99 09:15:30 GMT
Organization: UltraNet Communications, Inc.
Lines: 30
Message-ID: <7fs87j$hii$2@ligarius.ultra.net>
References: <7fq0nd$v0l$1@antiochus.ultra.net> <654.782T231T6563612@sky.bus.com>
NNTP-Posting-Host: d11.dial-14.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 24 Apr 1999 11:00:03 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d11

In article <654.782T231T6563612@sky.bus.com>,
   "Charlie Gibbs" <cgibbs@sky.bus.com> wrote:
>In article <7fprat$tjd$1@ligarius.ultra.net>, <jmfbahciv@aol.com> wrote:
>
>>Now you're getting into technical areas that I don't feel
>>comfortable talking about since I didn't do the implementation
>>work.  Unfortunately, there only a few who do know that
>>habitually read this newsgroup and there are even fewer who
>>can take the time to teach it.  It's an aspect of having the
>>work disappear off the face of the earth [sad emoticon here].
>
>One of the most depressing things about getting older is when
>you find you've been around long enough to watch knowledge die.
>
Especially when one gets to watch the people (who developed that
knowledge) die.  I knew very early on that there was a game afoot
to quash the PDP-10 line so I spent a lot of time ensuring that
everything we had in bits and on paper got preserved.  Everything
went into the shitcan when I got sick and had to stop working.
Even the "official" archives that had been guaranteed to preserve
everything submitted to it, had a personnel turnover.  So the
original agreements were also forgotten and even that stuff
has been shitcanned.  This includes source tapes of stuff that
never went out to the field.  We had huge fights with
product management to get sources out to the field.  After
the advent of VMS-lobotomyism, we generally lost those fights.

/BAH

Subtract a hundred and four for e-mail.

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 24 Apr 1999 09:29:06 -0500
Organization: none
Lines: 30
Message-ID: <7fskfi$527@bonkers.taronga.com>
References: <371be3e4.0@newsfeed.one.net> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu> <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net>
NNTP-Posting-Host: bonkers.in.taronga.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!logbridge.uoregon.edu!newsfeed.stanford.edu!news.ems.psu.edu!bonkers!not-for-mail

In article <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net>,
Ric Werme <werme@werme.ne.mediaone.net> wrote:
>Well, you could detach from your terminal, I think with the program running,
>and later attach back to it.  I often worked in my office, detached,
>went to the lab, and attached back.  Can't do that with Unix!

Funny, I do it all the time.

<misty-eyed>

Damn, I wrote my first "screen-like" program in 1979 or 1980. It didn't use
ptys...  V7 didn't have them... so you couldn't run raw-mode programs in the
sessions.  But then the 11/70 was so overloaded that running "vi" during peak
usage was severely frowned upon... imagine 70 users on something about
equivalent to an AT&T UNIX-PC with 2MB.

It was actually an attempt to emulate a Honeywell level-6 console, where you
could start multiple jobs and get the output line-by-line with a job-ID
prepended to it. I was an early adopter of the "nested ~ problem"... I noticed
that "mailx" used a "~" for a command escape, so I used that myself. Then
Ken Arnold bitched at me for using a broken version of "write", because
my first cut at using the double-tilde escape passed both tildes on.

</misty-eyed>

-- 
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
"Twenty-two points, plus triple-word-score, plus fifty points for using all
 my letters.  Game's over.  I'm outta here." -- Melissa ... by Kwyjibo.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sat, 24 Apr 99 11:06:50 GMT
Organization: UltraNet Communications, Inc.
Lines: 123
Message-ID: <7fseoc$1rh$1@ligarius.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7fprat$tjd$1@ligarius.ultra.net> <OW_T2.21$MF5.8637@srvr1.engin.umich.edu> <7fq0nd$v0l$1@antiochus.ultra.net> <7fq37v$9dc$1@vixen.cso.uiuc.edu>
NNTP-Posting-Host: d15.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 24 Apr 1999 12:51:24 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!uunet!ams.uu.net!ffx.uu.net!in5.uu.net!cyclone.swbell.net!news.idt.net!feed1.news.rcn.net!rcn!news.ultranet.com!d15

In article <7fq37v$9dc$1@vixen.cso.uiuc.edu>,
   tsmurphy@students.uiuc.edu (Terry Murphy) wrote:
>In article <7fq0nd$v0l$1@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
>
>>It never hurts to ask.  Usually, in this newsgroup, asking a question
>>get a whole lot of discussion going with myriads of useful 
>>thread drift.  The key to understanding is to get somebody to talk
>>about the philosophy of the implementation; of course, the 
>>philosophy part usually isn't specified (but can be deduced) 
>>because few people worked on more than one implementation.
>>JMF did TOPS-10, some flavor of UNIX, a shade of VMS, and
>>was the CPU software expert in DEC.  He would have enjoyed
>>telling people about it all, but he didn't read any newsgroups.
>>I doubt he even knew about them.
>
>What did TOPS-10 look/feel like?  Was it similar to the later DEC
>operating systems such as the RSX-11 family and VMS?  (in terms of
>commands available, languages used, file system, etc.)

The command prompt of TOPS-10 was dot (.).  Jim was very proud
of that dot.  Whenever he was showing somebody around the plant,
he would bring the guest to a terminal, point at the dot and
say "I write the program that types that dot."  Very few
people appreciated the humor and the implication.

TOPS-10 did not have a "user-friendly" command interface.  It
had a very terse standard for commands.  I preferred that myself
since I considered that a lot of my time was saved by an OS
that was terse.  

In the beginnings of the TOPS-10 implementations, the way
a user got a file copied was to do the following commands
(the first character is the user prompt but I won't underline
it as we would have in the docs--oh yea, that underline used
to be a backarrow--and I'm going to have to preface all those
lines with spaces because this bloody software is using a
dot as a post delimiter grump, grump, grump):

        .R PIP
        *dev:file02.exe_dev:file01.ext
        *^C  ;this is really one character and would have
        .    ;circled in our docs.  It's a <ctrl-c>

This would copy file01.ext to file02.ext...that backarrow really
gave the direction of the flow of bits (unlike those minis that
tended to do things bassackwards :-)).

TOPS-10 allowed ^C-intercept.  So one could type ahead all
the work one had to do, go to lunch, and come back with all
the work done (provided one didn't make a typo, the system
didn't crash, there wasn't an error in the work, and somebody 
didn't mess with one's terminal).

Later on, a command decoder was implemented in the monitor.
So if I typed:

        .COPY FOO1.BAR=FOO1.BAZ

The program PIP.EXE would be run by the monitor in behalf of
the user and the rest of the command would be passed to the 
program.  TOPS-10 had a whole bunch of system programs that
we supported, developed and shipped to the customers.  They
were programs that were generally useful.  Our group did not
do special code for single customers (those tasks belonged
to a different group).  However, if the software developed
by the special systems group turned out to be generally 
useful and in demand by a good number of our customers, the
special systems group would start up a joint project with
our monitor group and merge sources.  That way the special
systems group no longer had to support the individual 
customers that they had serviced and all the hard/software
suppport mechanisms would be handled by the people
who did the general work.

As far as what kind of work was done on TOPS-10 systems--
Anything would be my answer.  Jim's first job, when hired
by DEC, was to work on a project that allowed various
flavors of computers to talk to each other.  IIRC, those
computers were PDP-12, PDP-10, some kinds of IBM, and there
had to be -11s and -08s.  It was the first "network" that
I knew about.  This site was a scientific site that was
funded by the government; Jim always called it the ORNL
project.

>
>Where does TOPS-10 live today? 

I don't know.  I don't think it does other than those
few who have glommed onto a KS10 and keep it running
just as a hobby.  TOPS-20, on the other hand, is actively
being developed and sold (I hope) by a company that had
the sense to apply today's technology to PDP-10 architecture.


>Is it still in use, and/or has it influenced other systems?

That's a real difficult question to answer.  I suspect that
anybody who used a -10 got the feel of having his/her own
personal computer.  And there were thousands and thousands 
of kiddies going to college who had that exposure.  TOPS-10
did networking very, very well (although it took a long time
to learn what not to do).  

One of the reasons VMS is such a pig is because the architect
was bound and determined not to do anything like TOPS-10; so
that was a negative influence [wry emoticon here].

>  Do TOPS-10 and -20 have anything to do with each other?

The short answer is yes.  But this is a story for another
day.  There was a lot of in-fighting between the two
monitor groups because of budget constraints, prima
donnaism, programming styles, and a corporate effort
to put the PDP-10 out of business.  All the language
developement groups did their development for both
operating systems which usually meant maintaining
two sets of sources.  Source management was not
a trivial task; reams and reams of discussions could
be spent on just this subject alone.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sat, 24 Apr 99 11:10:58 GMT
Organization: UltraNet Communications, Inc.
Lines: 25
Message-ID: <7fsf02$1rh$2@ligarius.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7fprat$tjd$1@ligarius.ultra.net> <OW_T2.21$MF5.8637@srvr1.engin.umich.edu> <7fq0nd$v0l$1@antiochus.ultra.net> <7fq37v$9dc$1@vixen.cso.uiuc.edu> <7fqbr8$74u$1@autumn.news.rcn.net>
NNTP-Posting-Host: d15.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 24 Apr 1999 12:55:30 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!netnews.com!newsswitch.lcs.mit.edu!news.ultranet.com!d15

In article <7fqbr8$74u$1@autumn.news.rcn.net>,
   roggblake@iname.com () wrote:
>On 23 Apr 1999 15:22:39 GMT, Terry Murphy <tsmurphy@students.uiuc.edu> 
wrote:
>>What did TOPS-10 look/feel like?  Was it similar to the later DEC
>
>I never used TOPS-10, but did do some work with TOPS-20 at one time.
>Loved the command-line user interface with its command completion and
>context-sensitive help.  (A clone of this interface lives on in MS-DOS
>Kermit, and was even used in Commodore 64 Kermit!)

<snip>

TOPS-10 did not have this kind of user help interface.  There
were budgetary and technical reasons for this.  We did try
to address the help problems later on.  The main reason we
didn't do very well with this is because anything reasonable
required a rewrite of SCNSER.MAC which was the module of the
TOPS-10 monitor that handled user commands.  There was one
thing that both Jim and Tony never, ever wanted to do and
that was deal with anything that had to do with SCNSER.

/BAH

Subtract a hundred and four for e-mail.

######

From: smj@fohnix.metronet.com (Stephen Jones)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 24 Apr 1999 11:37:22 -0500
Organization: Texas Metronet, Inc  (login info (214/488-2590 - 817/571-0400))
Message-ID: <7fss02$sj7@fohnix.metronet.com>
References: <371be3e4.0@newsfeed.one.net> <7frcok$3ck@weyl.math.psu.edu> <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net> <7fri4v$3eq@weyl.math.psu.edu>
NNTP-Posting-Host: 207.170.106.4
Lines: 14
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!feed1.news.rcn.net!rcn!korova.insync.net!diablo.metronet.com!news.metronet.com!fohnix.metronet.com!not-for-mail

>	Unless I severely misparsed you - yes, you can do exactly what
>you described: man screen for details. It allows to detach the session,
>do whatever you want (including logout), go back later (maybe after new
>login/telnet from other host/whatever) and reattach it. Is it what you are
>talking about?

What sort of parser are you using?  No, UNIX job control is eternally
broken (LINUX never fixed this either).  Don't point a TOPS-20 hacker 
to the bogus man page for screen.  That is a programme, we are talking
about MONITORS and real job control.  DETACH and ATTACH are excellent.
In UNIX you have to be running screen before your process goes "zombie"
(and even then, a zombied process is lost).  Even linux has this 
problem of zombies that don't go away until you reboot the machine.

######

Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
From: lisard@zetnet.co.uk
X-Everything: Net-Tamer V 1.08X
Date: 24 Apr 1999 14:53:54 GMT
Message-ID: <3721db02$0$493@news.zetnet.co.uk>
NNTP-Posting-Host: man-199.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 924965634 493 194.247.40.252
Lines: 15
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.icl.net!dispose.news.demon.net!demon!btnet-peer!btnet-feed1!btnet!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail



On 1999-04-23 jmfbahciv@aol.com said:
   :rob.nicholson@zetnet.co.uk (Rob Nicholson) wrote:
   :>I thought the NT kernel was designed by one of the primary
   :>VMS designers? Wasn't this a good pedigree?

   :Exactly.

Miaow. :>

So, for those of use who have never used and will never use a PDP-10...
care to say a few words about TOPS-10 from the user's perspective?
--
Communa (lisard@zetnet.co.uk) -- you know soft spoken changes nothing

######

From: rob.nicholson@zetnet.co.uk (Rob Nicholson)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sat, 24 Apr 1999 15:07:42 GMT
Organization: Rainow Systems Ltd.
Reply-To: rob.nicholson@zetnet.co.uk
Message-ID: <3721de15.608786@news.zetnet.co.uk>
References: <371be3e4.0@newsfeed.one.net> <7fprat$tjd$1@ligarius.ultra.net> <OW_T2.21$MF5.8637@srvr1.engin.umich.edu> <7fq0nd$v0l$1@antiochus.ultra.net> <7fq37v$9dc$1@vixen.cso.uiuc.edu> <7fseoc$1rh$1@ligarius.ultra.net>
X-Newsreader: Forte Agent 1.5/32.452
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
NNTP-Posting-Host: manr-027.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 924966465 490 194.247.43.157
Lines: 5
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!dispose.news.demon.net!demon!peernews!peernews.manap.net!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail

>PIP.EXE

Ah ha - so that's where PIP on CP/M came from?

Rob.l

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu> <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net> <7fri4v$3eq@weyl.math.psu.edu> <8LoU2.21349$tY1.13078@wbnws01.ne.mediaone.net>
X-P-Meow: Meow
X-Attribution: Mw
Mail-Copies-To: never
X-Face: OuVjX/xTC||oPCROag,@&DMNx`dB{O~n#TV*1Sit(!\i|e@#e%Sp=d"/%XEpDa[mLBY1_v8
 EJIi0IQqSPwkFBsrnq/RCk>\kgh9-[S?_!4bI{SUW)4&$cMfp3e0RIFh*ciA?\o@HL1gwj-t{$rVZJ
 6]UFLxlCMxsiq=hjU-zMgC+"QA(2jzJi;DdQP?JW]:kFX|R(5:n%lSLOj&&'2W>c<
X-P-Virus: Hi! I'm a header virus! Copy me into yours and join the fun!
From: greg andruk <meowing@banet.net>
Date: 24 Apr 1999 15:36:07 -0400
Message-ID: <87hfq5dd5j.fsf@banet.net>
Organization: Meow Industries CSSRI
X-Newsreader: Gnus v5.7/XEmacs 20.4 - "Emerald"
Cancel-Lock: sha1:V0az2QCvU1fhPHBmLfEX9sAp/w8=
NNTP-Posting-Host: 32.100.55.247
X-Trace: 24 Apr 1999 19:36:24 GMT, 32.100.55.247
X-Notice: Items posted that violate the IBM.NET Acceptable Use Policy
X-Notice: should be reported to postmaster@ibm.net
X-Complaints-To: postmaster@ibm.net
Lines: 18
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!news.maxwell.syr.edu!newsfeed.us.ibm.net!ibm.net!news3.us.ibm.net!32.100.55.247

werme@werme.ne.mediaone.net (Ric Werme) writes:

> On RH Linux:
>    $ man screen
>    No manual entry for screen

That's odd.  I can see where stuff like virtual consoles and X make
screen less useful if you're using linux on a workstation, but it's
still nifty for remote work.

Whellll, <URL:ftp://ftp.gnu.org/pub/gnu/screen/> if you don't have it
and want a copy.

-- 
            Misc. Meowing: <URL:http://members.xoom.com/meowing/>
                                  **meow**
  "There is nothing worse than having a spare couple of hours and you can't
                 find an open server to abuse." - Tim Thorne

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 24 Apr 1999 15:40:39 -0500
Organization: none
Lines: 25
Message-ID: <7fta87$ccc@bonkers.taronga.com>
References: <371be3e4.0@newsfeed.one.net> <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net> <7fri4v$3eq@weyl.math.psu.edu> <7fss02$sj7@fohnix.metronet.com>
NNTP-Posting-Host: bonkers.in.taronga.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!howland.erols.net!math.ohio-state.edu!news.cis.ohio-state.edu!news.ems.psu.edu!bonkers!not-for-mail

In article <7fss02$sj7@fohnix.metronet.com>,
Stephen Jones <smj@fohnix.metronet.com> wrote:
>What sort of parser are you using?  No, UNIX job control is eternally
>broken (LINUX never fixed this either).  Don't point a TOPS-20 hacker 
>to the bogus man page for screen.  That is a programme, we are talking
>about MONITORS and real job control.

That's an implementation detail. UNIX does a lot of stuff with separate
programs that older designed like TOPS and VMS build into the OS. It's a
different point along the continuum from raw microkernels like QNX and
AmigaOS to pure monitors.

>DETACH and ATTACH are excellent.
>In UNIX you have to be running screen before your process goes "zombie"
>(and even then, a zombied process is lost).

Zombies have nothing to do with job control. A Zombie isn't a stopped process,
it's just a placeholder in the process table created when a program terminates
that's removed when the parent process picks up the exit code.

-- 
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
"Twenty-two points, plus triple-word-score, plus fifty points for using all
 my letters.  Game's over.  I'm outta here." -- Melissa ... by Kwyjibo.

######

From: "Peter Hendén" <phenden@algonet.se>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sat, 24 Apr 1999 16:10:28 +0200
Organization: Teknisk Dokumentation AB
Lines: 30
Message-ID: <7fsjbg$91o$1@zingo.tninet.se>
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu><371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net><Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7fs7mv$hii$1@ligarius.ultra.net>
NNTP-Posting-Host: du145-146.ppp.algonet.se
X-Trace: zingo.tninet.se 924962992 9272 195.100.146.145 (24 Apr 1999 14:09:52 GMT)
X-Complaints-To: abuse@algo.net
NNTP-Posting-Date: 24 Apr 1999 14:09:52 GMT
X-Newsreader: Microsoft Outlook Express 4.72.3155.0
X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3155.0
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!news.algonet.se!algonet!pepsi.tninet.se!not-for-mail

jmfbahciv@aol.com wrote:
>In article <ALDERSON.99Apr23113531@netcom.netcom.com>,
>   alderson@netcom.netcom.com (Richard M. Alderson III) wrote:
<snip>
>>I'll leave it to real Tops-10 hands to discuss this further;
>>I do Tops-20, but BAH talks to me anyway. ;-)
>Of course I do :-)).  You still use a PDP-10.  TOPS-20 was
>the best for doing user mode programming in machine language.
>The reason I didn't like to use the -20 is because I couldn't
>get anything done as soon as somebody else logged in; it was
>too slow for my taste (remember I was always doing about
>5-10 jobs at once).  I always expected a compile, load, and
>save to only take a minute or two :-)).

Wasn't this dependent on whether your programs were
TOPS-10 (used PA1050) or TOPS-20 native. As I recall
(1977 - 80) we had a lot of users on our machine running
Interlisp-20 and a bunch of other native programs, and they
didn't get sluggish until long after the TOPS-10 progs did.

Regards,
Peter

--
Peter Hendén           http://www.algonet.se/~phenden
ICQ: 14672398
Teknisk Dokumentation AB          http://www.tdab.com

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net> <7fri4v$3eq@weyl.math.psu.edu> <7fss02$sj7@fohnix.metronet.com>
Organization: Plethora . Net - More net, less spam!
X-Newsreader: trn 4.0-test70 (17 January 1999)
From: seebs@plethora.net (Peter Seebach)
Lines: 20
Message-ID: <IUmU2.671$WA4.140554@ptah.visi.com>
Date: Sat, 24 Apr 1999 17:08:56 GMT
NNTP-Posting-Host: 205.166.146.8
X-Complaints-To: abuse@visi.com
X-Trace: ptah.visi.com 924973736 205.166.146.8 (Sat, 24 Apr 1999 12:08:56 CDT)
NNTP-Posting-Date: Sat, 24 Apr 1999 12:08:56 CDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newshub.northeast.verio.net!scrappy.visi.com!news-out.mcia.org!news-out.visi.com!ptah.visi.com!not-for-mail

In article <7fss02$sj7@fohnix.metronet.com>,
Stephen Jones <smj@fohnix.metronet.com> wrote:
>What sort of parser are you using?  No, UNIX job control is eternally
>broken (LINUX never fixed this either).  Don't point a TOPS-20 hacker 
>to the bogus man page for screen.  That is a programme, we are talking
>about MONITORS and real job control.  DETACH and ATTACH are excellent.
>In UNIX you have to be running screen before your process goes "zombie"
>(and even then, a zombied process is lost).  Even linux has this 
>problem of zombies that don't go away until you reboot the machine.

I have never seen a zombie that doesn't go away until you reboot the
machine.  I have seen zombies that don't go away until their parent dies
too, that's it.

-s
-- 
Copyright 1999, All rights reserved.  Peter Seebach / seebs@plethora.net
C/Unix wizard, Pro-commerce radical, Spam fighter.  Boycott Spamazon!
Will work for interesting hardware.  http://www.plethora.net/~seebs/
Visit my new ISP <URL:http://www.plethora.net/> --- More Net, Less Spam!

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 24 Apr 1999 17:29:08 -0400
Organization: -ENOENT
Lines: 26
Message-ID: <7ftd34$3pb@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net> <7fri4v$3eq@weyl.math.psu.edu> <7fss02$sj7@fohnix.metronet.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!newspump.monmouth.com!newspeer.monmouth.com!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7fss02$sj7@fohnix.metronet.com>,
Stephen Jones <smj@fohnix.metronet.com> wrote:
>What sort of parser are you using?  No, UNIX job control is eternally
>broken (LINUX never fixed this either).  Don't point a TOPS-20 hacker 
	What part of job control are you talking about? I had a distinct
impression that BSD scheme (process groups + ability to stop a group / separate
process + ability to resume a group / process + signal delivery to groups +
PTYs as possible control terminals + group leaders, etc. - all in the kernel)
was copied from TOPS-20. Could you comment on how the TOPS-20 scheme works?

>to the bogus man page for screen.  That is a programme, we are talking
>about MONITORS and real job control.  DETACH and ATTACH are excellent.
	D'oh. If you think that screen pokes in kernel tables... It uses a
real job control. Which sits in the kernel.
>In UNIX you have to be running screen before your process goes "zombie"
>(and even then, a zombied process is lost).  Even linux has this 
>problem of zombies that don't go away until you reboot the machine.
	What? How did it come that I've missed your PR? <looking through
archives> Nope, looks like I didn't. Oh, well... If that happens you've got
init stuck in something. Which Should Not Happen (tm). How to reproduce that?
BTW, in that respect Linux is not different from BSD, so BSD folks would
probably also like to hear about the thing. Details, way to reproduce?

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <ALDERSON.99Apr23113531@netcom.netcom.com>
     <7frcok$3ck@weyl.math.psu.edu> <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net> <7fri4v$3eq@weyl.math.psu.edu>
X-Newsreader: NN version 6.5.0 CURRENT #119
From: werme@werme.ne.mediaone.net (Ric Werme)
Lines: 36
Message-ID: <8LoU2.21349$tY1.13078@wbnws01.ne.mediaone.net>
Date: Sat, 24 Apr 1999 19:15:16 GMT
NNTP-Posting-Host: 24.128.109.10
X-Trace: wbnws01.ne.mediaone.net 924981316 24.128.109.10 (Sat, 24 Apr 1999 15:15:16 EDT)
NNTP-Posting-Date: Sat, 24 Apr 1999 15:15:16 EDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!netnews.com!chnws02.mediaone.net!24.128.1.101!chnws05.ne.mediaone.net!24.128.44.7!wbnws01.ne.mediaone.net.POSTED!not-for-mail

viro@weyl.math.psu.edu (Alexander Viro) writes:

>In article <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net>,
>Ric Werme <werme@werme.ne.mediaone.net> wrote:

>>  A close relative to the debug command (which we noted _didn't_
>>debug your program), was make.  In fact, you could type "MAKE LOVE", and
>	Wait a bit. Wasn't it a part of TECO that did document creation
>(as opposed to editing)?

Indeed, but the CCL stuff (I forget all the command it handled), ran first.
Glitch - .MAKE LOVE probably _did_ let you create a file named love.
Basically, MAKE created a command file that was simply EWfile.ext$$ and passed
it to teco.  The EDIT? TECO? command passed EBfile.ext$Y$ instead.  Y for
"yank", which read the first page of the file into the edit buffer.

>>Well, you could detach from your terminal, I think with the program running,
>>and later attach back to it.  I often worked in my office, detached,
>>went to the lab, and attached back.  Can't do that with Unix!  Also, we

>	Unless I severely misparsed you - yes, you can do exactly what
>you described: man screen for details. It allows to detach the session,
>do whatever you want (including logout), go back later (maybe after new
>login/telnet from other host/whatever) and reattach it. Is it what you are
>talking about?

On RH Linux:

   $ man screen
   No manual entry for screen

I'll check it on Tru64 Unix next week.
--
Ric Werme                  | http://people.ne.mediaone.net/werme
werme@nospam.mediaone.net  | http://www.cyberportal.net/werme
      ^^^^^^^ delete

######

Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7fprat$tjd$1@ligarius.ultra.net>
     <OW_T2.21$MF5.8637@srvr1.engin.umich.edu> <7fq0nd$v0l$1@antiochus.ultra.net>
     <7fq37v$9dc$1@vixen.cso.uiuc.edu> <7fseoc$1rh$1@ligarius.ultra.net>
X-Newsreader: NN version 6.5.0 CURRENT #119
From: werme@werme.ne.mediaone.net (Ric Werme)
Lines: 24
Message-ID: <MVoU2.21351$tY1.13191@wbnws01.ne.mediaone.net>
Date: Sat, 24 Apr 1999 19:26:36 GMT
NNTP-Posting-Host: 24.128.109.10
X-Trace: wbnws01.ne.mediaone.net 924981996 24.128.109.10 (Sat, 24 Apr 1999 15:26:36 EDT)
NNTP-Posting-Date: Sat, 24 Apr 1999 15:26:36 EDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!netnews.com!chnws02.mediaone.net!24.128.1.101!chnws05.ne.mediaone.net!24.128.44.7!wbnws01.ne.mediaone.net.POSTED!not-for-mail

jmfbahciv@aol.com writes:

>In article <7fq37v$9dc$1@vixen.cso.uiuc.edu>,
>   tsmurphy@students.uiuc.edu (Terry Murphy) wrote:

>>Is it still in use, and/or has it influenced other systems?

TOPS-10 influenced many systems.  TS-8 and RT-11 had very similar command
processors.  I accidentally dialed into a TS-8 system once and was really
confused by its SYSTAT output.  (SYSTAT and Eliza/Lisp's "MAKNAM UNHAPPY"
made it onto the Firesign Theatre's "We're All bozos on this Bus.")
CP/M was greatly influenced by RT-11.  Microsoft started out with a PDP-10,
their Z80 assembler MACRO-80 had many directives straight from MACRO-10.
E.g. IRP and IRPC.

MS-DOS was greatly influenced by CP/M, and once Microsoft got control, I
suppose the PDP-10 too.  With version 2? 3? Unix started having more influence.
The forward slash was already taken by command line options (ala PDP-10), so
the hierarchical directory structure of MS-DOS used backslash as the pathname
component separator.  Sigh. 
--
Ric Werme                  | http://people.ne.mediaone.net/werme
werme@nospam.mediaone.net  | http://www.cyberportal.net/werme
      ^^^^^^^ delete

######

Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
From: lisard@zetnet.co.uk
X-Everything: Net-Tamer V 1.08X
Date: 24 Apr 1999 19:49:07 GMT
Lines: 8
Message-ID: <37222033$0$496@news.zetnet.co.uk>
NNTP-Posting-Host: man-197.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 924983347 496 194.247.40.250
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newspeer.clara.net!news.clara.net!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail



On 1999-04-24 jmfbahciv@aol.com said:
   :...a corporate effort to put the PDP-10 out of business.

For heaven's sake - why???
--
Communa (lisard@zetnet.co.uk) -- you know soft spoken changes nothing

######

From: cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 24 Apr 1999 21:36:41 GMT
Organization: Honest Chris' Sysadmin Emporium
Message-ID: <7ftdh9$oi4$1@teabag.demon.co.uk>
References: <37222033$0$496@news.zetnet.co.uk>
NNTP-Posting-Host: localhost
X-NNTP-Posting-Host: teabag.demon.co.uk:193.237.4.110
X-Trace: news.demon.co.uk 924991119 nnrp-08:5012 NO-IDENT teabag.demon.co.uk:193.237.4.110
X-Complaints-To: abuse@demon.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Newsreader: knews 1.0b.0
Lines: 15
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!dispose.news.demon.net!demon!news.demon.co.uk!demon!teabag.demon.co.uk!not-for-mail

In article <37222033$0$496@news.zetnet.co.uk>,
	lisard@zetnet.co.uk writes:
> On 1999-04-24 jmfbahciv@aol.com said:
>   :...a corporate effort to put the PDP-10 out of business.
> 
> For heaven's sake - why???

Because the VAX was DEC's new baby, and it was embarrassing that
the PDP-10 could outperform it.  I get the same impression about
the Alpha and VAX situation, history repeating itself...

[I believe that some of the PDP-10 hardware design was recycled
for use in the VAX-8600 series, but I'm not certain]

Chris.

######

From: Eivind <db96ek@oliven.hib.no>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sat, 24 Apr 1999 23:24:06 +0200
Organization: UNINETT news service
Lines: 28
Sender: db96ek@solo.hib.no
Message-ID: <Pine.LNX.4.10.9904242321320.23119-100000@solo.hib.no>
References: <371be3e4.0@newsfeed.one.net> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu> <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net> <7fri4v$3eq@weyl.math.psu.edu> <8LoU2.21349$tY1.13078@wbnws01.ne.mediaone.net> <87hfq5dd5j.fsf@banet.net>
Reply-To: ekj@vestdata.no
NNTP-Posting-Host: solo.hib.no
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Trace: snipp.uninett.no 924988897 10739 158.37.86.20 (24 Apr 1999 21:21:37 GMT)
X-Complaints-To: news-abuse@uninett.no
In-Reply-To: <87hfq5dd5j.fsf@banet.net>
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!uni-erlangen.de!newsfeed.nacamar.de!newsfeed.nacamar.de!uninett.no!solo.hib.no!db96ek

On 24 Apr 1999, greg andruk wrote:
>werme@werme.ne.mediaone.net (Ric Werme) writes:

>> On RH Linux:
>>    $ man screen
>>    No manual entry for screen
>That's odd.  I can see where stuff like virtual consoles and X make
>screen less useful if you're using linux on a workstation, but it's
>still nifty for remote work.

Screen is optional with RedHat Linux. It's surely still a useful thing
to have around. there's a limit to how many xterms you can have around,
and it's very neat to be able to detach something and reattach it
somewhere else.

On my system:

$ rpm -q screen
screen-3.7.4-3

Does anyone happen to know the origin of this nifty little program ? It
occurs to me that a terminal multiplexer must've been a really useful
thing to have around for a rather long time.


	Eivind

######

From: blarsen@infostream.no (Bjørn Hell Larsen)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 25 Apr 1999 08:45:27 +0200
Lines: 46
Message-ID: <u4sm5tczc.fsf@infostream.no>
References: <371be3e4.0@newsfeed.one.net> <7fpjgn$v7t$4@antiochus.ultra.net>      <Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu> <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net>
NNTP-Posting-Host: aubert.svg.infostream.no
X-Newsreader: Gnus v5.5/Emacs 20.3
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news.algonet.se!algonet!uninett.no!Norway.EU.net!not-for-mail


werme@werme.ne.mediaone.net (Ric Werme) writes:

> Sort of.  We had something called CCL, for Concise Command Language, so you
> could say "debug foo" and it would create some command files, switch to the
> assembler, that would switch to the linker, and that would link it with DDT
> and run it.  We thought that was pretty cool.  Nothing like a Makefile,
> though. 

Most of the TOPS-10 sites did, however, have MIC - Macro Interpreted
Commands - created at Hatfield Polytechnical in the UK.  This allowed
you to create command files with internal logic that could be run
interactively with the command DO. It had more in common with shell
scripts than with make(1), but at least it allowed you to automate
sequences of commands for things like compiles, without having to
submit such jobs to the batch system.

At the UiO TOPS-10 site, one of the system programmers created a
software maintenance tool called SOUPE with in retrospect worked like
a combination of RCS and make - you entered all the source files that
was part of a product (I used it for compilers and assemblers) into a
SOUPE source deck, which provided version control and a command
language that allowed you to specify dependencies etc.

SOUPE was able to create MIC files that performed a complete compile
of a given version of the product.


> A close relative to the debug command (which we noted _didn't_
> debug your program), was make.  In fact, you could type "MAKE LOVE", and
> make would sleep a second or so and print "NOT WAR?" and exit.  It _was_
> during the Vietnam war, after all....

Nah. "MAKE" was one of the commands that invoked the TECO
editor. "MAKE FOO.FOR" created the file FOO.FOR, and invoked TECO on
it. If I recall correctly, "MAKE LOVE" indeed did create the file
"LOVE" and let you edit it, it just said "NOT WAR?" before invoking
TECO.

MIC had it's own little hack - it implemented a number of MIC-specific
commands in the monitor, and one of these was GOTO, to allow loops in
.MIC files. If you entered "GOTO HELL" interactively to the monitor,
the response was "get stuffed".


Bjørn

######

From: blarsen@infostream.no (Bjørn Hell Larsen)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 25 Apr 1999 08:47:51 +0200
Lines: 25
Message-ID: <u3e1ptcvc.fsf@infostream.no>
References: <371be3e4.0@newsfeed.one.net> <7fpjgn$v7t$4@antiochus.ultra.net> <Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu> <Pine.NXT.4.20.9904232023400.13242-100000@Tomobiki-Cho.CAC.Washington.EDU>
NNTP-Posting-Host: aubert.svg.infostream.no
X-Newsreader: Gnus v5.5/Emacs 20.3
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!news.algonet.se!algonet!newsfeed1.online.no!newsfeed.online.no!Norway.EU.net!not-for-mail


Mark Crispin <mrc@CAC.Washington.EDU> writes:

> On 23 Apr 1999, Alexander Viro wrote:
> > Wait a bit. Do you mean that there was no scheduling within the same
> > login session?
> 
> login session == process.
> 
> > Does it mean that big compile == sitting and waiting or going
> > to other terminal?
> 
> Well, not necessarily.  A job could be detached from a terminal, and a new
> one logged in.  So, you could get somewhat of the same effect of multiple
> processes, by successively attaching and detaching your jobs.

With the limitation that a detached job didn't run. At least, it
blocked once it tried to perform any terminal IO.

Which is why I was lucky enough to end up with three terminals, two
VT100's and a LA36 for running all-night unsupervised compiles.


Bjørn

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 25 Apr 99 09:38:27 GMT
Organization: UltraNet Communications, Inc.
Lines: 24
Message-ID: <7futut$um6$3@antiochus.ultra.net>
References: <3721db02$0$493@news.zetnet.co.uk>
NNTP-Posting-Host: d16.dial-12.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 25 Apr 1999 11:23:09 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d16

In article <3721db02$0$493@news.zetnet.co.uk>, lisard@zetnet.co.uk wrote:
>
>
>On 1999-04-23 jmfbahciv@aol.com said:
>   :rob.nicholson@zetnet.co.uk (Rob Nicholson) wrote:
>   :>I thought the NT kernel was designed by one of the primary
>   :>VMS designers? Wasn't this a good pedigree?
>
>   :Exactly.
>
>Miaow. :>

You've never tried to get something done with that guy in charge.
Until then...I reserve my right to express my cats.
>
>So, for those of use who have never used and will never use a PDP-10...
>care to say a few words about TOPS-10 from the user's perspective?

I'm trying :-).

/BAH


Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 25 Apr 99 09:39:28 GMT
Organization: UltraNet Communications, Inc.
Lines: 14
Message-ID: <7fuu0r$um6$4@antiochus.ultra.net>
References: <37222033$0$496@news.zetnet.co.uk>
NNTP-Posting-Host: d16.dial-12.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 25 Apr 1999 11:24:11 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!howland.erols.net!newsswitch.lcs.mit.edu!news.ultranet.com!d16

In article <37222033$0$496@news.zetnet.co.uk>, lisard@zetnet.co.uk wrote:
>
>
>On 1999-04-24 jmfbahciv@aol.com said:
>   :...a corporate effort to put the PDP-10 out of business.
>
>For heaven's sake - why???

I have lots of theories.  Small computer thinking may have been
one of them.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 25 Apr 99 10:13:03 GMT
Organization: UltraNet Communications, Inc.
Lines: 12
Message-ID: <7fuvvp$8v5$1@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7fprat$tjd$1@ligarius.ultra.net> <OW_T2.21$MF5.8637@srvr1.engin.umich.edu> <7fq0nd$v0l$1@antiochus.ultra.net> <7fq37v$9dc$1@vixen.cso.uiuc.edu> <7fseoc$1rh$1@ligarius.ultra.net> <3721de15.608786@news.zetnet.co.uk>
NNTP-Posting-Host: d16.dial-12.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 25 Apr 1999 11:57:45 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!netnews.com!newsswitch.lcs.mit.edu!news.ultranet.com!d16

In article <3721de15.608786@news.zetnet.co.uk>,
   rob.nicholson@zetnet.co.uk (Rob Nicholson) wrote:
>>PIP.EXE
>
>Ah ha - so that's where PIP on CP/M came from?

And, if I had my Black Packs, I'd open up PIP.MAC and
the edit history would tell me where we got it from :-).

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 25 Apr 99 10:24:26 GMT
Organization: UltraNet Communications, Inc.
Lines: 30
Message-ID: <7fv0l4$8v5$2@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net> <7fri4v$3eq@weyl.math.psu.edu> <7fss02$sj7@fohnix.metronet.com> <7fta87$ccc@bonkers.taronga.com>
NNTP-Posting-Host: d16.dial-12.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 25 Apr 1999 12:09:08 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newspump.monmouth.com!newspeer.monmouth.com!news.ultranet.com!d16

In article <7fta87$ccc@bonkers.taronga.com>,
   peter@taronga.com (Peter da Silva) wrote:
>In article <7fss02$sj7@fohnix.metronet.com>,
>Stephen Jones <smj@fohnix.metronet.com> wrote:

>>What sort of parser are you using?  No, UNIX job control is eternally
>>broken (LINUX never fixed this either).  Don't point a TOPS-20 hacker 
>>to the bogus man page for screen.  That is a programme, we are talking
>>about MONITORS and real job control.
>
>That's an implementation detail. ...<snip>...

Actually, it's not.  It has everything to do with a philosophy
of where and who should control what functions.  A host of
security problems, bugs, etc. occur when the philosophy gets
blurred or misimplemented.  Unfortunately, this aspect of
an operating system really doesn't get documented.  The only
way that I know people learned about our philosophy was 
mentorship or working with the code.  Quite a few of our
customers wanted to have us take their particular implementations
and distribute and support it.  Quite a few were rejected
because of a different philosophy basis.  

I know this is rather vague but I really don't know how
to talk about it.  JMF always called it style.  I think it's
even deeper that that.

/BAH

Subtract a hundred and four for e-mail.

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 25 Apr 1999 12:01:43 -0500
Organization: none
Lines: 70
Message-ID: <7fvhpn$241@bonkers.taronga.com>
References: <371be3e4.0@newsfeed.one.net> <7fss02$sj7@fohnix.metronet.com> <7fta87$ccc@bonkers.taronga.com> <7fv0l4$8v5$2@antiochus.ultra.net>
NNTP-Posting-Host: bonkers.in.taronga.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.berkeley.edu!newsfeed.stanford.edu!news.ems.psu.edu!bonkers!not-for-mail

In article <7fv0l4$8v5$2@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
>In article <7fta87$ccc@bonkers.taronga.com>,
>   peter@taronga.com (Peter da Silva) wrote:
>>In article <7fss02$sj7@fohnix.metronet.com>,
>>Stephen Jones <smj@fohnix.metronet.com> wrote:
>>>What sort of parser are you using?  No, UNIX job control is eternally
>>>broken (LINUX never fixed this either).  Don't point a TOPS-20 hacker 
>>>to the bogus man page for screen.  That is a programme, we are talking
>>>about MONITORS and real job control.

>>That's an implementation detail. ...<snip>...

>Actually, it's not.  It has everything to do with a philosophy
>of where and who should control what functions.  A host of
>security problems, bugs, etc. occur when the philosophy gets
>blurred or misimplemented.

Indeed.

I think you're misinterpreting what I was getting at, though. There was
a bunch of context in the stuff you snipped, and I definitely agree that
you do need to understand the deep structure of an OS.

What I'm getting at is that UNIX doesn't use "job control" to provide
that sort of functionality, and the "job control" layered on top of UNIX
is an alien artifact that doesn't really mesh well into the rest of the
system. Trying to map a "monitor" style OS's job control into UNIX's
process model is just going to lead to confusion.

Stephen's comments about zombie processes, for example, are a perfect example
of that confusion. Zombies have nothing to do with "job control" in UNIX. They
are an accidentally exposed implementation detail [1]. 

And the fact that the implementation of detaching and reattaching uses a
different mechanism on different systems has nothing to do with whether
one's "better" than another. Screen provides functionality you couldn't get
on the -11... I can detach a screen and connect to it later without having
the detached process hang on I/O, for example.

>Unfortunately, this aspect of
>an operating system really doesn't get documented.  The only
>way that I know people learned about our philosophy was 
>mentorship or working with the code.

Indeed.

There's a lot of stuff that's been layered on top of UNIX because people were
working with it without understanding the deep structures bhind the system's
design. Berkeley sockets, job control, System V IPC (the old Xenix IPC which
was bound to special files was a lot better).

--

[1] the kernel structures could be organized so that return values from
dead processes were propogated to some structure in the parent process
table, for example... so you wouldn't have zombies. But a program that
spawned processes without collecting their return values would still cause
resource leaks, just in a different structure.

And in a sytsem where process creation was expensive or limited the whole
problem couldn't occur. Instead you get a situation where a single user's
interactive environment (login, monitor, job, what have you) becomes basically
single-threaded. These are all tradeoffs you make when you pick a point along
the continuum from microkernel to monitor.

-- 
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
"Twenty-two points, plus triple-word-score, plus fifty points for using all
 my letters.  Game's over.  I'm outta here." -- Melissa ... by Kwyjibo.

######

From: Tim Shoppa <shoppa@trailing-edge.com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 25 Apr 1999 14:03:19 -0400
Organization: Trailing Edge Technology
Lines: 14
Message-ID: <372320A7.2895C5B0@trailing-edge.com>
References: <371be3e4.0@newsfeed.one.net> <7fss02$sj7@fohnix.metronet.com> <7fta87$ccc@bonkers.taronga.com> <7fv0l4$8v5$2@antiochus.ultra.net> <7fvhpn$241@bonkers.taronga.com>
NNTP-Posting-Host: trailing-edge.wdn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.0     DEC 3000 Model 300L)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!europa.clark.net!europa.netcrusader.net!207.106.0.20!news-xfer.newsread.com!netaxs.com!newsread.com!netaxs.newsread.com!usenet

Peter da Silva wrote:
> And the fact that the implementation of detaching and reattaching uses a
> different mechanism on different systems has nothing to do with whether
> one's "better" than another. Screen provides functionality you couldn't get
> on the -11... I can detach a screen and connect to it later without having
> the detached process hang on I/O, for example.

Non-Unix OS's on an -11 do provide "screen" functionality (TSX+,
in particular, does it quite nicely.)  TSX+ is probably better
described as a monitor than as a kernel, though, and certainly
the interface to the windowing features is done in a more
monitor-like way (via EMT's).

Tim.

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 25 Apr 1999 15:04:25 -0500
Organization: none
Lines: 18
Message-ID: <7fvsg9$4lg@bonkers.taronga.com>
References: <371be3e4.0@newsfeed.one.net> <7fv0l4$8v5$2@antiochus.ultra.net> <7fvhpn$241@bonkers.taronga.com> <372320A7.2895C5B0@trailing-edge.com>
NNTP-Posting-Host: bonkers.in.taronga.com
X-No-Archive: yes
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!netnews.com!feed1.news.rcn.net!rcn!newsfeed.stanford.edu!news.ems.psu.edu!bonkers!not-for-mail

In article <372320A7.2895C5B0@trailing-edge.com>,
Tim Shoppa  <shoppa@trailing-edge.com> wrote:
>Peter da Silva wrote:
>> And the fact that the implementation of detaching and reattaching uses a
>> different mechanism on different systems has nothing to do with whether
>> one's "better" than another. Screen provides functionality you couldn't get
>> on the -11... I can detach a screen and connect to it later without having
>> the detached process hang on I/O, for example.

>Non-Unix OS's on an -11 do provide "screen" functionality

Sorry, that was a typo. I meant to type -10.

-- 
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
"Twenty-two points, plus triple-word-score, plus fifty points for using all
 my letters.  Game's over.  I'm outta here." -- Melissa ... by Kwyjibo.

######

From: roggblake@iname.com (Roger Blake)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 25 Apr 1999 15:11:43 GMT
Organization: Ministry of Silly Walks
Lines: 10
Message-ID: <slrn7i6c5e.43b.roggblake@unix3.netaxs.com>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <slrn7hrjii.dbl.roggblake@unix5.netaxs.com> <37222d81.1058246@news.zetnet.co.uk> <7fpprq$fps$1@autumn.news.rcn.net> <3721821d.6019996@news.zetnet.co.uk>
NNTP-Posting-Host: unix3.netaxs.com
X-No-Archive: Yes
X-Newsreader: slrn (0.9.4.3 UNIX)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.eurocyber.net!newspeer1.nac.net!news-xfer.newsread.com!netaxs.com!newsread.com!netaxs.newsread.com!roggblake

On Fri, 23 Apr 1999 14:35:54 GMT, Rob Nicholson <rob.nicholson@zetnet.co.uk> wrote:
>I've heard about this. What was the primary reason? Performance?

Yes, specifically video performance, so that NT could become a viable game
platform for the kiddies when MS consolidates everything into "Windows 2000."
Just the kind of "feature" your business wants in its servers.. :-)

-- 
  Roger Blake
  (Remove second "g" in name to send e-mail.)

######

From: "Carl R. Friend" <carl.friend@stoneweb.com>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 25 Apr 1999 17:19:50 -0400
Organization: as little as possible!
Lines: 14
Message-ID: <372386F6.E0EA04A8@stoneweb.com>
References: <37237197$0$496@news.zetnet.co.uk>
NNTP-Posting-Host: zephyr.ultranet.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 25 Apr 1999 21:19:51 GMT
X-Mailer: Mozilla 4.5 [en] (X11; I; Linux 2.0.29 i586)
X-Accept-Language: en
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!not-for-mail

lisard@zetnet.co.uk wrote:
> 
> There's a PDP-6 schematic online somewhere; can someone point me in
> its direction?

   http://www.ai.mit.edu/people/tk/pdp6/pdp6.html

-- 
+------------------------------------------------+---------------------+
| Carl Richard Friend (UNIX Sysadmin)            | West Boylston       |
| Minicomputer Collector / Enthusiast            | Massachusetts, USA  |
| mailto:crfriend@ma.ultranet.com                |                     |
| http://www.ultranet.com/~crfriend/museum       | ICBM: N42:22 W71:47 |
+------------------------------------------------+---------------------+

######

From: smj@fohnix.metronet.com (Stephen Jones)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 25 Apr 1999 19:41:59 -0500
Organization: Texas Metronet, Inc  (login info (214/488-2590 - 817/571-0400))
Lines: 34
Message-ID: <7g0con$68m@fohnix.metronet.com>
References: <371be3e4.0@newsfeed.one.net> <7fta87$ccc@bonkers.taronga.com> <7fv0l4$8v5$2@antiochus.ultra.net> <7fvhpn$241@bonkers.taronga.com>
NNTP-Posting-Host: 207.170.106.4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!korova.insync.net!diablo.metronet.com!news.metronet.com!fohnix.metronet.com!not-for-mail

Peter da Silva <peter@taronga.com> wrote:
>
>Stephen's comments about zombie processes, for example, are a perfect example
>of that confusion. Zombies have nothing to do with "job control" in UNIX. They
>are an accidentally exposed implementation detail [1]. 

Peter - I did not mean to link zombies with job control.  I know they have
nothing to do with that as UNIX has had the problem of "the process won't
die" and "the child process with no parent" and "the run away process that
lost its TTY and is now eating up CPU and memory".  They are all crocking
UNIXism. Just reboot and everything is fine.  That is the philosophy.  There
is not concept of attach a debugger to the job/process and fix it (unless
you are running HURD?)  I'm sorry, with my understanding of UNIX internals
I should be vomiting in a toilet rather than writing garbage here :)

>And the fact that the implementation of detaching and reattaching uses a
>different mechanism on different systems has nothing to do with whether
>one's "better" than another. Screen provides functionality you couldn't get
>on the -11... I can detach a screen and connect to it later without having
>the detached process hang on I/O, for example.

Its a philosophy.  And the UNIX philosophy is, if you shoot yourself
in the foot, you die.  If we shoot you in the head, you die too.  /BAH has
it right that we're talking about philosophy here.  And we should step
back and take a lesson from the many Hindu religions and philosophies 
argued over, critiqued and revised for thousands of years.  We know that
the best ideas ALWAYS fall short of dung (ie, TOPS-20 and Buddhism (but
you can blame the khmer rouge, chinese army and others for that too)).

We can look at this like UNIX and Christianity.  Both are eternally
broken, cause wars, plague nations, promise you something but give you
nothing, bleed and suck you dry.  Yet, people flock to them.  I guess
the promise is so attractive.

######

Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
From: lisard@zetnet.co.uk
X-Everything: Net-Tamer V 1.08X
Date: 25 Apr 1999 19:48:39 GMT
Lines: 25
Message-ID: <37237197$0$496@news.zetnet.co.uk>
NNTP-Posting-Host: man-087.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 925069719 496 194.247.41.108
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.cwix.com!194.6.79.69!peernews!peernews.manap.net!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail



On 1999-04-25 jmfbahciv@aol.com said:
   :In article <37222033$0$496@news.zetnet.co.uk>, lisard@zetnet.co.uk
   :wrote: >
   :>On 1999-04-24 jmfbahciv@aol.com said:
   :>   :...a corporate effort to put the PDP-10 out of business.

   :>For heaven's sake - why???

   :I have lots of theories.  Small computer thinking may have been
   :one of them.

What do you mean by "small computer thinking"? To my mind, it just
doesn't make much sense to discontinue a proven and well-loved
architecture just because it's not quite the latest thing. For example,
if the -10 became underpowered, why not simply recast it in silicon and
move it down the range? The core processor would appear to be simple
enough that this presents no great challenge to the hardware. And then
we might have had workstations with PDP-10 chips in them, for example...

Which reminds me. There's a PDP-6 schematic online somewhere; can
someone point me in its direction?
--
Communa (lisard@zetnet.co.uk) -- you know soft spoken changes nothing

######

Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
From: lisard@zetnet.co.uk
X-Everything: Net-Tamer V 1.08X
Date: 25 Apr 1999 19:48:41 GMT
Message-ID: <37237199$0$496@news.zetnet.co.uk>
NNTP-Posting-Host: man-087.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 925069721 496 194.247.41.108
Lines: 16
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!dispose.news.demon.net!demon!colt.net!newspeer.clara.net!news.clara.net!peernews!peernews.manap.net!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail



On 1999-04-24 werme@werme.ne.mediaone.net(RicWerme) said:
   :MS-DOS was greatly influenced by CP/M, and once Microsoft got
   :control, I suppose the PDP-10 too.  With version 2? 3? Unix started
   :having more influence. The forward slash was already taken by
   :command line options (ala PDP-10), so the hierarchical directory
   :structure of MS-DOS used backslash as the pathname component
   :separator.  Sigh.

Only the command processor or the client programs care, though; the file
calls themselves are fine with forward-slash-delimited names. But if you
use them a whole slew of programs are going to complain about duff
options...
--
Communa (lisard@zetnet.co.uk) -- you know soft spoken changes nothing

######

Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
From: lisard@zetnet.co.uk
X-Everything: Net-Tamer V 1.08X
Date: 25 Apr 1999 19:48:42 GMT
Message-ID: <3723719a$0$496@news.zetnet.co.uk>
NNTP-Posting-Host: man-087.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 925069722 496 194.247.41.108
Lines: 15
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!dispose.news.demon.net!demon!colt.net!newspeer.clara.net!news.clara.net!peernews!peernews.manap.net!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail



On 1999-04-25 jmfbahciv@aol.com said:
   :In article <3721db02$0$493@news.zetnet.co.uk>, lisard@zetnet.co.uk
   :wrote: >
   :>Miaow. :>

   :You've never tried to get something done with that guy in charge.
   :Until then...I reserve my right to express my cats.

Tell us more. This is exactly the kind of perspective we could use for
working out at least some of the reasons for NT being a steaming turd of
an operating system...
--
Communa (lisard@zetnet.co.uk) -- you know soft spoken changes nothing

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 25 Apr 1999 21:18:24 -0500
Organization: none
Lines: 45
Message-ID: <7g0idg$avi@bonkers.taronga.com>
References: <371be3e4.0@newsfeed.one.net> <7fv0l4$8v5$2@antiochus.ultra.net> <7fvhpn$241@bonkers.taronga.com> <7g0con$68m@fohnix.metronet.com>
NNTP-Posting-Host: bonkers.in.taronga.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.cwix.com!128.32.206.55!newsfeed.berkeley.edu!newsfeed.stanford.edu!news.ems.psu.edu!bonkers!not-for-mail

In article <7g0con$68m@fohnix.metronet.com>,
Stephen Jones <smj@fohnix.metronet.com> wrote:
>Peter - I did not mean to link zombies with job control.  I know they have
>nothing to do with that

You just found it more convenient to toss that in than something that made
sense?

>They are all crocking
>UNIXism. Just reboot and everything is fine.  That is the philosophy.

No, that's not the philosophy. The only time I have ever been in a situation
where you couldn't kill a process has been where the hardware had failed or
there was a bug in a device driver... and apart from a busted Xenix device
driver hanging terminals in the 1984 release of Xenix 286 every time I've
tracked down a busted device driver it's been some PC hardware that didn't
perform as it was documented to perform.

>There
>is not concept of attach a debugger to the job/process and fix it (unless
>you are running HURD?)

% man gdb
[...]

       You can, instead, specify a process ID as a  second  argu-
       ment, if you want to debug a running process:

       gdb program 1234

[...]
% uname -a
FreeBSD mongrel.in.taronga.com 3.1-RELEASE FreeBSD 3.1-RELEASE #8: Wed Mar 17 22:26:40 CST 1999     root@mongrel.in.taronga.com:/work/src/sys/compile/woof  i386

>I'm sorry, with my understanding of UNIX internals
>I should be vomiting in a toilet rather than writing garbage here :)

If your understanding of the UNIX "deep structures" is that shallow, you
probably ought to purge yourself of something, yes.

-- 
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
"Twenty-two points, plus triple-word-score, plus fifty points for using all
 my letters.  Game's over.  I'm outta here." -- Melissa ... by Kwyjibo.

######

From: "Alan H. Martin" <AMartin@MA.UltraNet.Com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 25 Apr 1999 23:21:09 -0400
Organization: UltraNet Communications , an RCN Company http://www.ultranet.com/
Lines: 23
Message-ID: <3723DBA5.E5B2FF0@MA.UltraNet.Com>
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu><371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net><Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7fs7mv$hii$1@ligarius.ultra.net> <7fsjbg$91o$1@zingo.tninet.se>
NNTP-Posting-Host: 207-172-216-194.s194.tnt1.sbo.ma.dialup.rcn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 26 Apr 1999 03:21:40 GMT
X-Mailer: Mozilla 4.5 [en] (Win95; U)
X-Accept-Language: en,en-US,en-GB,es
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!uunet!ams.uu.net!uunet!zur.uu.net!nyc.uu.net!newsfeed.mathworks.com!newsswitch.lcs.mit.edu!news.ultranet.com!not-for-mail

"Peter Hendén" wrote:
> 
> jmfbahciv@aol.com wrote:
...
> >The reason I didn't like to use the -20 is because ...; it was
> >too slow for my taste ...
> 
> Wasn't this dependent on whether your programs were
> TOPS-10 (used PA1050) or TOPS-20 native. As I recall
> (1977 - 80) we had a lot of users on our machine running
> Interlisp-20 and a bunch of other native programs, and they
> didn't get sluggish until long after the TOPS-10 progs did.

Could be.  However, here's something to consider about -10 vs. -20 I/O
performance:

On the one hand, I don't know how fast PA1050-simulated buffered I/O went
compared to PMAP.  However, I definitely recall hearing that the DUMPI and
DUMPO JSY' that were intended for simulation of unbuffered I/O were *the*
fastest form of I/O on Tops-20.
				/AHM
-- 
Alan Howard Martin				AMartin@MA.UltraNet.Com

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 26 Apr 1999 01:41:40 -0700
Organization: Chez Inwap
Message-ID: <7g18s4$75o$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <Gs5U2.29$MF5.10448@srvr1.engin.umich.edu>
Lines: 43
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925116104 219 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <Gs5U2.29$MF5.10448@srvr1.engin.umich.edu>,
Sergej Roytman <ftit@engin.umich.edu> wrote:
>Hm, interesting.  So what the users saw was a virtual front-panel
>machine: load your simulated core with a program, specify an entry
>point, and hit run?

Commands included:

  GET DTA0:ADVENT.SAV		;Get program from DECtape
  GET MTA0:BACKUP.EXE 40K	;From 9-track tape, assign 40Kwords of core
  E 137				;Examine version number in location 137
  D 1 2 74			;Deposit 1,,000002 in location 74
  START 141			;Start with PC=000141 (default was 000140)
  REENTER			;Return to TECO after typing Control-C
  SAVE				;Save current core image to disk or tape

>Things like pipelines would have been done with intermediate files

COMPIL determined whether an assembler or compiler needed to be run to
create an up-to-date REL file.  For job 12, it would create DSK:012MAC.TMP
to have commands for MACRO and DSK:012LNK.TMP for LINK, and then start
SYS:MACRO at its starting address plus one.  That told MACRO to look
for commands in the TMP file.  After assembling the code, MACRO would
run SYS:LINK at its CCL entry point, and LINK would load the REL files
into a core image that was saved to disk, left in core, or executed.

>, then, and peripherals would have a
>fake front-end which would really talk to the appropriate queues and
>buffers in the kernel.  And, the front-ends being in the kernel
>themselves, they could stay more tightly coupled with the underlying
>data structures and report more things to the user, hence BAH's stated
>dislike for dumping data down a pipe and not knowing what happens to
>it?
>
>So, 'dI guess right?  Huh?

Ooo.  That's too accurate.  Must be a ringer, then.
	-Joe

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 26 Apr 1999 01:51:11 -0400
Organization: -ENOENT
Lines: 25
Message-ID: <7g0usf$4lg@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7fv0l4$8v5$2@antiochus.ultra.net> <7fvhpn$241@bonkers.taronga.com> <7g0con$68m@fohnix.metronet.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newspump.monmouth.com!newspeer.monmouth.com!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7g0con$68m@fohnix.metronet.com>,
Stephen Jones <smj@fohnix.metronet.com> wrote:
>Peter da Silva <peter@taronga.com> wrote:
>>
>>Stephen's comments about zombie processes, for example, are a perfect example
>>of that confusion. Zombies have nothing to do with "job control" in UNIX. They
>>are an accidentally exposed implementation detail [1]. 
>
>Peter - I did not mean to link zombies with job control.  I know they have
>nothing to do with that as UNIX has had the problem of "the process won't
>die" and "the child process with no parent" and "the run away process that
		... becomes child of init, which reaps it ASAP.
>lost its TTY and is now eating up CPU and memory".  They are all crocking
	Huh?
>UNIXism. Just reboot and everything is fine.  That is the philosophy.  There
	Huh?
>is not concept of attach a debugger to the job/process and fix it (unless
	Huh? man gdb, please.
>you are running HURD?)  I'm sorry, with my understanding of UNIX internals
>I should be vomiting in a toilet rather than writing garbage here :)
	s/writing garabage//. Sorry.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 26 Apr 1999 02:19:13 -0700
Organization: Chez Inwap
Message-ID: <7g1b2h$ifj$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu>
Lines: 67
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925118358 212 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7frcok$3ck@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>	Wait a bit. Do you mean that there was no scheduling within the same
>login session?

Yes.

> Does it mean that big compile == waiting or going to other terminal?

Yes.

> Could the process create a new one ...

No.  Processes were job numbers.  Job numbers were assigned when a command was
typed at a terminal that was not logged in.  The LOGIN command created a
long-term job number.  Other commands that could be run when not logged in
(such as SYSTAT and QUEUE) released their job number when they exited.

> wait for its completion and resume when the child finishes (without
> touching the keyboard, that is)?

The answer to that was posted in two other articles in this thread.

1) Using CCL (Concise COmmand Language), a chain of programs could be
   run.  Each program in the chain was told to run the next one upon
   successful termination.  If you wanted to, you could type ahead on the
   terminal.  The characters you typed would not be looked at until
   either a program went to read from the terminal, or the terminal
   went back into monitor mode (in which case the Monitor would read
   the typed ahead command).

2) Using MIC, a third-party program that DEC included wiith the Monitor
   sources.  There was a bit in the TTY status word that indicated that
   the terminal was under MIC's control.  Whenever the program blocked
   waitting for TTY input, or the Monitor was waiting for the next
   command, the master MIC process would be woken up.  It would read
   the next line of input from the current MIC file, and then simulate
   terminal input.  (This was done without PTYs; the characters were
   stuffed into the device-driver buffer.)

>From the paper on TOPS-20 I got an impression that it had background processes.

TOPS-10 and TOPS-20 both had background processes.  TOPS-10 prior to GALAXY
relied on OPSER to start all the background processes.  It would allocate
a PTY, send "LOGIN 1,2" to it, then "GET SYS:PROGRAM", "CSTART" and
"DETACH".  The program would have a job number, but no TTY, making it a
a background process.

>	Life with ulimit -u 4 <shudder> Was there any equivalent of virtual
>consoles?

Nope.  But if you had a key to the Computer Center (back when the Terminal
Room was not open 24 hours) and came in after hours, you could log in separate
jobs on the 12 ADM3A terminals and use a chair with wheels to do multitasking
between the 12 keyboards.

But if it was a big compile, it made more sense to put the commands into a
*.CTL file and submit it to the batch queue.  BATCON would allocate a PTY,
send your commands to the PTY one at a time, and send the results to
a *.LOG file.  (There was limited error-recovery in the batch control
language.)  Batch jobs could submit themselves back into the queue to
run again in 24 hours; there was no 'cron' facility.
	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: cdu@jawa. (chris ulrich)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 26 Apr 1999 02:42:39 GMT
Organization: UCR Computing and Communications
Lines: 21
Message-ID: <7g0jqv$fbb$1@pravda.ucr.edu>
References: <371be3e4.0@newsfeed.one.net> <7fri4v$3eq@weyl.math.psu.edu> <8LoU2.21349$tY1.13078@wbnws01.ne.mediaone.net> <87hfq5dd5j.fsf@banet.net>
NNTP-Posting-Host: jawa.ucr.edu
X-Newsreader: trn 4.0-test69 (20 September 1998)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!masternews.telia.net!news-feed.inet.tele.dk!bofh.vszbr.cz!newsfeed.stanford.edu!newsfeed.berkeley.edu!news.ucdavis.edu!pravda.ucr.edu!not-for-mail

In article <87hfq5dd5j.fsf@banet.net>, greg andruk  <meowing@banet.net> wrote:
>That's odd.  I can see where stuff like virtual consoles and X make
>screen less useful if you're using linux on a workstation, but it's
>still nifty for remote work.
>            Misc. Meowing: <URL:http://members.xoom.com/meowing/>

  I've always been surprised that screen for X was 99% written but
never completely finished.  I mean, if you are going to the trouble of
making an X server as an X client, why not include the extra couple
primitives of "suspend this session's connection to this X server" and
"resume this session's connection to that X server."  There are obvious
problems of color depth and key mapping and other sorts of things, but
really, I can't see it as a huge modification to the existing X nest, 
or a *hard* problem to add the remaining functionality necessary.

  Okay, to make this post slightly on-topic:
  how hard would it be to write/run X clients on a pdp-10 OS? 
What is the oldest still running system that has X clients (even if
they are not in active use)?  My guess would be an early vax running
a slightly modern version of VMS, anyone have something older?
chris

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 26 Apr 1999 06:26:16 -0500
Organization: none
Lines: 23
Message-ID: <7g1igo$luj@bonkers.taronga.com>
References: <371be3e4.0@newsfeed.one.net> <8LoU2.21349$tY1.13078@wbnws01.ne.mediaone.net> <87hfq5dd5j.fsf@banet.net> <7g0jqv$fbb$1@pravda.ucr.edu>
NNTP-Posting-Host: bonkers.in.taronga.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!netnews.com!newsfeed.berkeley.edu!newsfeed.stanford.edu!news.ems.psu.edu!bonkers!not-for-mail

In article <7g0jqv$fbb$1@pravda.ucr.edu>, chris ulrich <cdu@jawa.> wrote:
>  I've always been surprised that screen for X was 99% written but
>never completely finished.  I mean, if you are going to the trouble of
>making an X server as an X client, why not include the extra couple
>primitives of "suspend this session's connection to this X server" and
>"resume this session's connection to that X server."

Are you talking about Xnest here? That would be a logical step, but as it
turns out there is a proxy server with attach and detach called "xmove".

ftp://ftp.cs.columbia.edu/pub/xmove/

Support for non-8-bit visuals is only in the beta. Remember, an X display has
a lot more state for the proxy to save and restore than a terminal does...

As for the topic, it began because of the golden rule: he who has the gold
makes the rules.

-- 
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
"Twenty-two points, plus triple-word-score, plus fifty points for using all
 my letters.  Game's over.  I'm outta here." -- Melissa ... by Kwyjibo.

######

From: Robert Billing <unclebob@tnglwood.demon.co.uk>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 26 Apr 1999 07:18:41 +0000
Organization: Tanglewood
Message-ID: <37241351.F947285A@tnglwood.demon.co.uk>
References: <371be3e4.0@newsfeed.one.net> <7fta87$ccc@bonkers.taronga.com> <7fv0l4$8v5$2@antiochus.ultra.net> <7fvhpn$241@bonkers.taronga.com> <7g0con$68m@fohnix.metronet.com>
NNTP-Posting-Host: localhost.demon.co.uk
X-NNTP-Posting-Host: tnglwood.demon.co.uk:158.152.132.30
X-Trace: news.demon.co.uk 925111717 nnrp-02:24877 NO-IDENT tnglwood.demon.co.uk:158.152.132.30
X-Complaints-To: abuse@demon.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.0.31 i586)
Lines: 12
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!dispose.news.demon.net!demon!news.demon.co.uk!demon!tnglwood.demon.co.uk!not-for-mail

Stephen Jones wrote:

> We can look at this like UNIX and Christianity.  Both are eternally
> broken, cause wars, plague nations, promise you something but give you

 ...and attract groundless criticism.

-- 
I am Robert Billing, Christian, inventor, traveller, cook and animal
lover, I live near 0:46W 51:22N.  http://www.tnglwood.demon.co.uk/
"Bother," said Pooh, "Eeyore, ready two photon torpedoes and lock
phasers on the Heffalump, Piglet, meet me in transporter room three"

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 26 Apr 99 09:05:06 GMT
Organization: UltraNet Communications, Inc.
Lines: 66
Message-ID: <7g1gcn$86c$1@ligarius.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7fta87$ccc@bonkers.taronga.com> <7fv0l4$8v5$2@antiochus.ultra.net> <7fvhpn$241@bonkers.taronga.com> <7g0con$68m@fohnix.metronet.com>
NNTP-Posting-Host: d2.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 26 Apr 1999 10:49:59 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!cam-news-hub1.bbnplanet.com!news.gtei.net!newsfeed.mathworks.com!newsswitch.lcs.mit.edu!news.ultranet.com!d2

In article <7g0con$68m@fohnix.metronet.com>,
   smj@fohnix.metronet.com (Stephen Jones) wrote:
>Peter da Silva <peter@taronga.com> wrote:
>>
>>Stephen's comments about zombie processes, for example, 
>>are a perfect example
>>of that confusion. Zombies have nothing to do with 
>>"job control" in UNIX. They
>>are an accidentally exposed implementation detail [1]. 
>
>Peter - I did not mean to link zombies with job control.  I know they have
>nothing to do with that as UNIX has had the problem of "the process won't
>die" and "the child process with no parent" and "the run away process that
>lost its TTY and is now eating up CPU and memory".  They are all crocking
>UNIXism. Just reboot and everything is fine.  That is the 
>philosophy.  There
>is not concept of attach a debugger to the job/process and fix it (unless
>you are running HURD?)  I'm sorry, with my understanding of UNIX internals
>I should be vomiting in a toilet rather than writing garbage here :)
>
>>And the fact that the implementation of detaching and reattaching uses a
>>different mechanism on different systems has nothing to do with whether
>>one's "better" than another. Screen provides functionality you 
>>couldn't get
>>on the -11... I can detach a screen and connect to it later 
>>without having
>>the detached process hang on I/O, for example.
>
>Its a philosophy.  And the UNIX philosophy is, if you shoot yourself
>in the foot, you die.  If we shoot you in the head, you die too.  /BAH has
>it right that we're talking about philosophy here.

But you're not when you say the following:

>  And we should step
>back and take a lesson from the many Hindu religions and philosophies 
>argued over, critiqued and revised for thousands of years.  We know that
>the best ideas ALWAYS fall short of dung (ie, TOPS-20 and Buddhism (but
>you can blame the khmer rouge, chinese army and others for that too)).
>
>We can look at this like UNIX and Christianity.  Both are eternally
>broken, cause wars, plague nations, promise you something but give you
>nothing, bleed and suck you dry.  Yet, people flock to them.  I guess
>the promise is so attractive.
> 
Look, just because I have a preference to another operating system
over UNIX (I prefer UNIX to VMS, by the way) does not imply that
I think UNIX sucks.  It's an operating system that was developed
for a different use than TOPS10.  It would be a very useful thing
to discuss the differences between OSes based on their philosophies.
Designs are done to solve particular problems.  There is no perfect
operating system and, I sincerely hope, there never will be.
The reason I talk about TOPS10 is because I know it best.  
Part of my intent with this thread is for the people who have
dealt with operating systems at the coding level to discuss
trade-offs when designing.  In reading these newsgroups, I am
getting the distinct impression that the kiddies don't even
know what "trade-offs" mean.

[harumphing emoticon]

/BAH



Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 26 Apr 99 09:14:32 GMT
Organization: UltraNet Communications, Inc.
Lines: 31
Message-ID: <7g1gvb$bid$1@ligarius.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu><371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net><Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7fs7mv$hii$1@ligarius.ultra.net> <7fsjbg$91o$1@zingo.tninet.se> <3723DBA5.E5B2FF0@MA.UltraNet.Com>
NNTP-Posting-Host: d17.dial-16.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 26 Apr 1999 10:59:55 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!uunet!ams.uu.net!newsfeed.amsterdam.nl.net!sun4nl!newspump.monmouth.com!newspeer.monmouth.com!news.ultranet.com!d17

In article <3723DBA5.E5B2FF0@MA.UltraNet.Com>,
   "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
>"Peter Hendén" wrote:
>> 
>> jmfbahciv@aol.com wrote:
>....
>> >The reason I didn't like to use the -20 is because ...; it was
>> >too slow for my taste ...
>> 
>> Wasn't this dependent on whether your programs were
>> TOPS-10 (used PA1050) or TOPS-20 native. As I recall
>> (1977 - 80) we had a lot of users on our machine running
>> Interlisp-20 and a bunch of other native programs, and they
>> didn't get sluggish until long after the TOPS-10 progs did.
>
>Could be.  However, here's something to consider about -10 vs. -20 I/O
>performance:
>
>On the one hand, I don't know how fast PA1050-simulated buffered I/O went
>compared to PMAP.  However, I definitely recall hearing that the DUMPI and
>DUMPO JSY' that were intended for simulation of unbuffered I/O were *the*
>fastest form of I/O on Tops-20.

You completely missed my point.  Just because one user used
something that did UUOs, should NOT have affected another user's
performance who is using JSYSes.  The -20 was a really good
stand-alone system, unless one wanted to do COMPILes.

/BAH

Subtract a hundred and four for e-mail.

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 26 Apr 1999 10:15:17 -0400
Organization: -ENOENT
Lines: 31
Message-ID: <7g1sdl$4vs@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu> <7g1b2h$ifj$1@shell3.ba.best.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cliffs.rs.itd.umich.edu!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7g1b2h$ifj$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
[snip]
>>	Life with ulimit -u 4 <shudder> Was there any equivalent of virtual
>>consoles?
>
>Nope.  But if you had a key to the Computer Center (back when the Terminal
>Room was not open 24 hours) and came in after hours, you could log in separate
>jobs on the 12 ADM3A terminals and use a chair with wheels to do multitasking
>between the 12 keyboards.

	<chuckle> Could the equivalent be done by allocating 12 PTYs, putting
TTY into the raw mode, reading from it, watching for switch characters and
feeding the input to appropriate PTY? (with corresponding stuff for output,
indeed). BTW, here's a serious design question - AFAICS large chunk of the
functionality of shell was put into the terminal driver, right? I.e. together
with the equivalent of cooked mode. Moreover, equivalent of process control
primitives was done via feeding the commands into that driver (BTW, was it
available via separate syscalls?) How much of that was duplicated over the
TTY/PTY/console drivers and how much was really independent from the type
of terminal? IOW, was there a separation between line discipline and terminal
driver? With a very little hackery the stuff you are describing can be
implemented as LDISC on UNIX. Hmm... That leads to the next question: how
monolitic the monitor was? That is, were there standard interfaces (within
the monitor) that could be used for writing a new LDISC/device driver/fs/
executable format? I know very well what a PITA is to pick the corresponding
pieces from existing parts of kernel if originally they were scattered all
over the place. To what extent it was done in case of TOPS-10?

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <ALDERSON.99Apr23113531@netcom.netcom.com> <Gs5U2.29$MF5.10448@srvr1.engin.umich.edu> <7g18s4$75o$1@shell3.ba.best.com>
Organization: University of Michigan, College of Engineering
From: ftit@engin.umich.edu (Sergej Roytman)
Lines: 11
Message-ID: <uiXU2.42$MF5.14297@srvr1.engin.umich.edu>
Date: Mon, 26 Apr 1999 10:34:02 GMT
NNTP-Posting-Host: 141.212.106.44
X-Trace: srvr1.engin.umich.edu 925122842 141.212.106.44 (Mon, 26 Apr 1999 06:34:02 EDT)
NNTP-Posting-Date: Mon, 26 Apr 1999 06:34:02 EDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cliffs.rs.itd.umich.edu!srvr1.engin.umich.edu!ftit

In article <7g18s4$75o$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>In article <Gs5U2.29$MF5.10448@srvr1.engin.umich.edu>,
>Sergej Roytman <ftit@engin.umich.edu> wrote:
>  GET MTA0:BACKUP.EXE 40K	;From 9-track tape, assign 40Kwords of core

And those 40K (virtual? (swapped? paged?)) memory?  Were they protected
from other users and their sloppy processes by software or specialized
hardware in the PDP-10?

-- 
Sergej Roytman

######

From: "Alan T. Bowler" <atbowler@thinkage.on.ca>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 26 Apr 1999 11:39:29 -0400
Organization: UUNET Canada News Transport
Lines: 11
Message-ID: <372488B1.8C38DC2C@thinkage.on.ca>
References: <371be3e4.0@newsfeed.one.net> <7fk9i4$spj$1@ligarius.ultra.net> <BlpT2.3$MF5.1852@srvr1.engin.umich.edu> <7fmv2b$591$2@antiochus.ultra.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu> <371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net>
NNTP-Posting-Host: 192.102.11.4
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.5 [en] (WinNT; U)
X-Accept-Language: en
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!howland.erols.net!newsswitch.lcs.mit.edu!cyclone.bc.net!news.uunet.ca!not-for-mail

jmfbahciv@aol.com wrote:
> 

> >The nice thing was, though, you didn't really notice the other users;
> >response time was usually very good.
> 
> Yup.  That's my point.  For all intents and purposes, one had
> the feel of their very own private computer.

In some respects it was a litle better.  The files on those timeshared
systems were reliably backed up.

######

From: Mark Crispin <mrc@CAC.Washington.EDU>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 26 Apr 1999 11:59:44 -0700
Organization: Networks & Distributed Computing
Lines: 25
Message-ID: <Pine.NXT.4.20.9904261154270.15191-100000@Tomobiki-Cho.CAC.Washington.EDU>
References: <371be3e4.0@newsfeed.one.net> <ALDERSON.99Apr23113531@netcom.netcom.com> <Gs5U2.29$MF5.10448@srvr1.engin.umich.edu> <7g18s4$75o$1@shell3.ba.best.com> <uiXU2.42$MF5.14297@srvr1.engin.umich.edu>
NNTP-Posting-Host: tomobiki-cho.cac.washington.edu
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Trace: nntp1.u.washington.edu 925153187 7896 (None) 140.142.17.37
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: tonyb
To: Sergej Roytman <ftit@engin.umich.edu>
In-Reply-To: <uiXU2.42$MF5.14297@srvr1.engin.umich.edu>
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!logbridge.uoregon.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc

On Mon, 26 Apr 1999, Sergej Roytman wrote:
> >  GET MTA0:BACKUP.EXE 40K	;From 9-track tape, assign 40Kwords of core
> 
> And those 40K (virtual? (swapped? paged?)) memory?  Were they protected
> from other users and their sloppy processes by software or specialized
> hardware in the PDP-10?

Of course.  Even the earliest PDP-6 CPUs and operating systems had that
feature from day 1.

One of the least desireable aspects of 1980s computing was the massive
creation of toy computers such as PCs and Macs which didn't do this.
Proving that a sucker is born ever minute, people still buy Macintoys and
Piece'o'Craps in which a single false address calculation can corrupt the
entire system.

Lest UNIX nerds start getting smug, there are toy UNIX platforms too.
Plus, any UNIX system can be made that way with a single "chmod".

-- Mark --

* RCW 19.190 notice: This email address is located in Washington State.	*
* Unsolicited commercial email may be billed $500 per message.		*
Science does not emerge from voting, party politics, or public debate.

######

From: Mark Crispin <mrc@CAC.Washington.EDU>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 26 Apr 1999 12:02:01 -0700
Organization: Networks & Distributed Computing
Lines: 18
Message-ID: <Pine.NXT.4.20.9904261200000.15191-100000@Tomobiki-Cho.CAC.Washington.EDU>
References: <371be3e4.0@newsfeed.one.net> <7fta87$ccc@bonkers.taronga.com> <7fv0l4$8v5$2@antiochus.ultra.net> <7fvhpn$241@bonkers.taronga.com> <7g0con$68m@fohnix.metronet.com> <37241351.F947285A@tnglwood.demon.co.uk>
NNTP-Posting-Host: tomobiki-cho.cac.washington.edu
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Trace: nntp1.u.washington.edu 925153324 33098 (None) 140.142.17.39
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: whlam
To: Robert Billing <unclebob@tnglwood.demon.co.uk>
In-Reply-To: <37241351.F947285A@tnglwood.demon.co.uk>
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!logbridge.uoregon.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc

On Mon, 26 Apr 1999, Robert Billing wrote:
> Stephen Jones wrote:
> > We can look at this like UNIX and Christianity.  Both are eternally
> > broken, cause wars, plague nations, promise you something but give you
>  ...and attract groundless criticism.

Be careful.  There are those of us who know quite a bit about UNIX and
Christianity, and can say with some authority that the main thing wrong
with criticism of either is that it is understated.

This is not the forum to continue this discussion.

-- Mark --

* RCW 19.190 notice: This email address is located in Washington State.	*
* Unsolicited commercial email may be billed $500 per message.		*
Science does not emerge from voting, party politics, or public debate.

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 26 Apr 1999 12:20:53 -0400
Organization: -ENOENT
Lines: 23
Message-ID: <7g23p5$559@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7g0con$68m@fohnix.metronet.com> <7g0usf$4lg@weyl.math.psu.edu> <H20V2.1086$WA4.216881@ptah.visi.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!newsxfer.itd.umich.edu.MISMATCH!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <H20V2.1086$WA4.216881@ptah.visi.com>,
Peter Seebach <seebs@plethora.net> wrote:
>In article <7g0usf$4lg@weyl.math.psu.edu>,
>Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>>lost its TTY and is now eating up CPU and memory".  They are all crocking
>>	Huh?
>
>There's a common problem (mostly this used to bite us with PINE) with crufty
>software that doesn't notice EOF on stdin, and sits around spinning on some
>read loop forever.  Many applications will be vulnerable if they're denied
>SIGHUP, but some just do it all the time anyway.  There's nothing the OS can
>plausibly do to distinguish this from a program which is merely badly written,
>in general.  And you can, of course, always kill 'em.

	Been there, seen that. Program that ignores the return value of
function it called *is* badly written. AFAIK no OS is able to stop program
from loop: goto loop;. Fresh example: nethack. Function reads a character,
checks if it belongs to <set> and either returns the index in set or rereads.
Ignores EOF.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: p98mccabe@alltel.net (Micheal H. McCabe)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 26 Apr 1999 12:26:58 -0400
Organization: Micheal McCabe Data Systems
Lines: 31
Message-ID: <p98mccabe-2604991226580001@mepa2pp48.alltel.net>
References: <371be3e4.0@newsfeed.one.net> <7fvhpn$241@bonkers.taronga.com> <7g0con$68m@fohnix.metronet.com> <7g0usf$4lg@weyl.math.psu.edu> <H20V2.1086$WA4.216881@ptah.visi.com>
NNTP-Posting-Host: mepa2pp48.alltel.net
X-Trace: news1.alltel.net 925143845 18346 166.102.103.49 (26 Apr 1999 16:24:05 GMT)
X-Complaints-To: abuse@client.alltel.net
NNTP-Posting-Date: 26 Apr 1999 16:24:05 GMT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!remarQ-easT!supernews.com!remarQ.com!feeder.qis.net!peerfeed.ncal.verio.net!news.central.agis.net!agis!news1.alltel.net!mepa2pp48.alltel.net!user


OK folks, here's where my ignorance really starts to show...

To begin with, I know nothing about UNIX internals.  My experiences with
time-sharing are limited to TS/OS on the Univac 90/60 (An IBM 360 clone),
VAX/VMS (Somewhere around version 2.? - Later upgraded to 4.?), and RT-11
on the pdp-11.

ISTR, when you submitted a DO file under VMS or RT-11 (hazy recall of an
@filename command), it appeared as if commands were entered from your
terminal and executed directly.  Output was displayed (or printed) on your
terminal unless redirected to a file.  During execution of this command
file, the only actions possible from the terminal keyboard were CTRL-S to
stop scrolling, CTRL-Q to resume, CTRL-C to interrupt execution of the
current command, or CTRL-Z to abort the whole thing.

Likewise, when you used SUBMIT to enter a batch job, an independant
process was created that would execute until it completed its task, ran
out of CPU time, or was killed by the user (or system operator.)  Once the
job was submitted, your terminal was free to do additional work.

Did I miss some fine distinction here?

How did you do this sort of thing under TOPS-10 or TOPS-20?  Also, my
understanding of UNIX was that each invocation of a command generally
created an independant process.  Is this an incorrect assumption?

Thanks in advance.

Micheal H. McCabe
p98mccabe@alltel.net

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 26 Apr 1999 13:13:03 -0400
Organization: -ENOENT
Lines: 71
Message-ID: <7g26qv$55r@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7g0usf$4lg@weyl.math.psu.edu> <H20V2.1086$WA4.216881@ptah.visi.com> <p98mccabe-2604991226580001@mepa2pp48.alltel.net>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!cam-news-hub1.bbnplanet.com!cpk-news-hub1.bbnplanet.com!news.gtei.net!news-peer.gip.net!news.gsl.net!gip.net!cliffs.rs.itd.umich.edu!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <p98mccabe-2604991226580001@mepa2pp48.alltel.net>,
Micheal H. McCabe <p98mccabe@alltel.net> wrote:
>
>OK folks, here's where my ignorance really starts to show...
>
>To begin with, I know nothing about UNIX internals.  My experiences with
>time-sharing are limited to TS/OS on the Univac 90/60 (An IBM 360 clone),
>VAX/VMS (Somewhere around version 2.? - Later upgraded to 4.?), and RT-11
>on the pdp-11.
>
>ISTR, when you submitted a DO file under VMS or RT-11 (hazy recall of an
>@filename command), it appeared as if commands were entered from your
>terminal and executed directly.  Output was displayed (or printed) on your
>terminal unless redirected to a file.  During execution of this command
>file, the only actions possible from the terminal keyboard were CTRL-S to
>stop scrolling, CTRL-Q to resume, CTRL-C to interrupt execution of the
>current command, or CTRL-Z to abort the whole thing.
>
>Likewise, when you used SUBMIT to enter a batch job, an independant
>process was created that would execute until it completed its task, ran
>out of CPU time, or was killed by the user (or system operator.)  Once the
>job was submitted, your terminal was free to do additional work.
>
>Did I miss some fine distinction here?
>
>How did you do this sort of thing under TOPS-10 or TOPS-20?  Also, my
	<AOL></AOL>
>understanding of UNIX was that each invocation of a command generally
>created an independant process.  Is this an incorrect assumption?
	In case of UNIX you have this functionality divided between several
layers.
	First of all, creation of new process is *fast*. There are two
primitives - one of them (fork()) creates a copy of the virtual machine (==
process) and lets it run in parallel with everything else. It's lazy, i.e.
pages are marked as copy-on-write rather than being copied. If a process
tries to modify such page it gets a copy (transparently) and c-o-w bit on
that copy is reset. fork() doesn't load a new program - it's done by
exec() (as far as I could understand TOPS-10 chain execution is done via
something pretty similar). exec() drops all memory of the virtual machine
that had called it and maps it on the executable you've asked for.
	Now, since the creation of processes is cheap it's offloaded from
the kernel. Any program can fork a copy, exec whatever you want in said
copy and either wait for the child's completion or simply go on with its
business. It doesn't have to bother with dumping its state, loading the
child instead of itself and restoring the state later. I.e. invocation
of child process is easy. Shell is a regular program that reads from input
(either terminal or file with a script) and invokes appropriate commands.
How much will be implemented by internal functions and how much will be
in utilities depends on the authors' tastes. Many other programs do the
same - e.g. mailer can invoke one of the standard editors instead of
implementing an internal one.
	Processing of ^U, ^C and friends is a different story - you can
either put the terminal into raw mode and make all processing yourself or
you can use a line discipline to process this stuff in the kernel. In the
latter case it's transparent for you. You can make a LDISC driver to interpret
a non-trivial language if you want (and are willing to write it). One can
implement a shell-alike as a LDISC and get a situation similar to (as far
as I could understand) one in TOPS-10. Generally nobody does it (e.g. since
debugging the user program is much safer and easier than debugging the
kernel code ;-/), but it is doable - I did something similar when I hacked
on EXT2 and needed a twisted equivalent of kdb. It didn't launch new
processes, but upon Esc-!-A it gave a prompt and allowed to watch/modify
VFS data, set/reset semaphores and call inode methods. Ugly, but it saved my
ass quite a few times. Attempt to do it from userspace was, to put it
mildly, unwise - half of operations would be impossible and it would have
very good chances to cause additional harm. So it can be done, but unless you
are doing pretty hairy kernel hacking it's simply not needed.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7fvhpn$241@bonkers.taronga.com> <7g0con$68m@fohnix.metronet.com> <7g0usf$4lg@weyl.math.psu.edu>
Organization: Plethora . Net - More net, less spam!
X-Newsreader: trn 4.0-test70 (17 January 1999)
From: seebs@plethora.net (Peter Seebach)
Lines: 31
Message-ID: <H20V2.1086$WA4.216881@ptah.visi.com>
Date: Mon, 26 Apr 1999 15:58:31 GMT
NNTP-Posting-Host: 205.166.146.8
X-Complaints-To: abuse@visi.com
X-Trace: ptah.visi.com 925142311 205.166.146.8 (Mon, 26 Apr 1999 10:58:31 CDT)
NNTP-Posting-Date: Mon, 26 Apr 1999 10:58:31 CDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!nntp.primenet.com!chippy.visi.com!news-out.visi.com!ptah.visi.com!not-for-mail

In article <7g0usf$4lg@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>Peter - I did not mean to link zombies with job control.  I know they have
>>nothing to do with that as UNIX has had the problem of "the process won't
>>die" and "the child process with no parent" and "the run away process that
>		... becomes child of init, which reaps it ASAP.
>>lost its TTY and is now eating up CPU and memory".  They are all crocking
>	Huh?

There's a common problem (mostly this used to bite us with PINE) with crufty
software that doesn't notice EOF on stdin, and sits around spinning on some
read loop forever.  Many applications will be vulnerable if they're denied
SIGHUP, but some just do it all the time anyway.  There's nothing the OS can
plausibly do to distinguish this from a program which is merely badly written,
in general.  And you can, of course, always kill 'em.

>>is not concept of attach a debugger to the job/process and fix it (unless
>	Huh? man gdb, please.
>>you are running HURD?)  I'm sorry, with my understanding of UNIX internals
>>I should be vomiting in a toilet rather than writing garbage here :)
>	s/writing garabage//. Sorry.

Yeah, well, it's pretty clear he doesn't know much about Unix internals, I'll
say that much.

-s
-- 
Copyright 1999, All rights reserved.  Peter Seebach / seebs@plethora.net
C/Unix wizard, Pro-commerce radical, Spam fighter.  Boycott Spamazon!
Will work for interesting hardware.  http://www.plethora.net/~seebs/
Visit my new ISP <URL:http://www.plethora.net/> --- More Net, Less Spam!

######

From: alex*@*rockvax.rockefeller.edu (Alexandre Pechtchanski)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Organization: Rockefeller University Hospital (GCRC), New York
Message-ID: <3728c7aa.452058724@Rockyd>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fj0bj$48e$1@bcrkh13.ca.nortel.com> <7fr83e$oi4$1@bgtnsc02.worldnet.att.net>
X-Newsreader: Forte Agent 1.5/32.451
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Lines: 30
Date: Mon, 26 Apr 1999 20:13:04 GMT
NNTP-Posting-Host: 129.85.24.56
X-Trace: rockyd.rockefeller.edu 925157589 129.85.24.56 (Mon, 26 Apr 1999 16:13:09 EDT)
NNTP-Posting-Date: Mon, 26 Apr 1999 16:13:09 EDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.nyu.edu!rockyd.rockefeller.edu!not-for-mail

On Fri, 23 Apr 1999 21:53:46 -0400, "Ambrose, Joseph"
<joseph.j.ambrose@worldnet.att.net> wrote:

>Clark <clarkg@xxxnortelnetworks.com> wrote in message
>news:7fj0bj$48e$1@bcrkh13.ca.nortel.com...
>> In article <Rf5T2.194$W4.62887@srvr1.engin.umich.edu>,
>ftit@engin.umich.edu (Sergej Roytman) wrote:
>>
>> [snip]
>>
>> >I could also have brought up OS8 on the PDP-8, which I think managed
>> >some sort of multi-userness in the standard 4kW of core.  But I've
>> >never used this, and don't know about its history.
>>
>> I don't think OS/8 was multi-user. But schools had the EduSystem
>> (BASIC) running multiple users on a PDP-8 with (mumble)kW of core.
>>
>I used Edusystem 20c in my HS days. I believe that EduSystem 50 WAS OS/8. I
>certainly looked multi-user. Had an opportunity to use it as a visitor to
>another HS. Definitely was major cool at that time.

No, EduSystem 50 was a PDP-11 version (IIRC, of course).  OS/8 didn't figure
into EduSystems at all.  OS/8 wasn't multi-user, but Fortran IV under it allowed
one to create multi-user applications (only one needed 12 kW of memory for this,
it would compile in 8 kW, but needed 12 kW for execution of the resulting task).
AFAIK, EduSystems were built around multi-user BASIC with some rudiments of
batch processing and file system, pretty similar to the OS/8 one.

[ When replying, remove *'s from address ]
Alexandre Pechtchanski, Systems Manager, RUH, NY

######

From: "Carl R. Friend" <carl.friend@stoneweb.com>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 26 Apr 1999 20:40:55 -0400
Organization: as little as possible!
Lines: 29
Message-ID: <37250797.6543C69A@stoneweb.com>
References: <37237197$0$496@news.zetnet.co.uk> <ALDERSON.99Apr26165228@netcom19.netcom.com>
NNTP-Posting-Host: zephyr.ultranet.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 27 Apr 1999 00:41:00 GMT
X-Mailer: Mozilla 4.5 [en] (X11; I; Linux 2.0.29 i586)
X-Accept-Language: en
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newspeer.clara.net!news.clara.net!news-lond.gip.net!news-peer.gip.net!news.gsl.net!gip.net!newsfeed.cwix.com!18.24.4.11!newsswitch.lcs.mit.edu!news.ultranet.com!not-for-mail

"Richard M. Alderson III" wrote:
> 
> The founder of cisco Systems had this great idea when he worked at
> DEC in the 70s, a Ten On A Desk.

   What with the advances in VLSI technology, and the fairly simple
and elegant nature of the -10 architecture, one would think that now
it'd be possible to stick a quad-10 on a chip with all the memory
it could possibly support. Add more memory for a "solid-state disk",
and you're in business. The die size would probably be manageable,
too, and the speed could be outrageous.

   However, I think that the folks from Redmond would have no clue
what to do with an 18-bit address space. 256k? How can _anyone_
program in that? "We talk in MEGs (GIGs, soon), not ks!"

> It's a bit large to stick on a desktop, but it eventually got done.

  That was the "Minnow" project, wasn't it? Were any ever produced?
I'm given to understand a few prototypes were built. What sort of chip
technology did they use?

-- 
+------------------------------------------------+---------------------+
| Carl Richard Friend (UNIX Sysadmin)            | West Boylston       |
| Minicomputer Collector / Enthusiast            | Massachusetts, USA  |
| mailto:carl.friend@stoneweb.com                |                     |
| http://www.ultranet.com/~crfriend/museum       | ICBM: N42:22 W71:47 |
+------------------------------------------------+---------------------+

######

From: Stefan Skoglund <stetson@ebox.tninet.se>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 26 Apr 1999 22:27:17 +0200
Organization: Telenordia
Lines: 16
Message-ID: <3724CC25.464B2C68@ebox.tninet.se>
References: <wk7lr7z16r.fsf@erols.com> <7fju8g$9ba$3@roch.zetnet.co.uk>
NNTP-Posting-Host: du209-241.ppp.algonet.se
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Trace: zingo.tninet.se 925161310 782 195.100.241.209 (26 Apr 1999 21:15:10 GMT)
X-Complaints-To: abuse@algo.net
NNTP-Posting-Date: 26 Apr 1999 21:15:10 GMT
X-Mailer: Mozilla 4.06 [en] (Win95; I)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!news.algonet.se!algonet!pepsi.tninet.se!not-for-mail

lisard@zetnet.co.uk wrote:
> 
> I'm prepared to bet that this is down to lousy memory management. NT's
> services are based on COM, and COM's memory management is based on
> self-managed reference counting. Nobody's going to convince me that this
> is year-at-a-time stable.

Hmm, i know of a sure way of detecting that.
Install linux and  then have an egcs install fest ie
get source, unpack, use the distribs gcc for a compile
and then compile egcs with the just compiled version.

Enough data so that a bad RAM should cause intermittent core-dumps.

Emacs is also sensitive to this.

######

From: dpeschel@u.washington.edu (Derek Peschel)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 26 Apr 1999 22:39:01 GMT
Organization: University of Washington, Seattle
Lines: 60
Message-ID: <7g2pu5$peq$1@nntp1.u.washington.edu>
References: <371be3e4.0@newsfeed.one.net> <7g18s4$75o$1@shell3.ba.best.com> <uiXU2.42$MF5.14297@srvr1.engin.umich.edu> <Pine.NXT.4.20.9904261154270.15191-100000@Tomobiki-Cho.CAC.Washington.EDU>
NNTP-Posting-Host: saul6.u.washington.edu
X-Trace: nntp1.u.washington.edu 925166341 26074 (None) 140.142.17.37
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: dpeschel
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!howland.erols.net!newspeer.monmouth.com!netnews1.nw.verio.net!netnews.nwnet.net!news.verio.net!news.u.washington.edu!dpeschel

In article <Pine.NXT.4.20.9904261154270.15191-100000@Tomobiki-Cho.CAC.Washington.EDU>,
Mark Crispin  <mrc@CAC.Washington.EDU> wrote:

>One of the least desireable aspects of 1980s computing was the massive
>creation of toy computers such as PCs and Macs which didn't do this.
>Proving that a sucker is born ever minute, people still buy Macintoys and
>Piece'o'Craps in which a single false address calculation can corrupt the
>entire system.
>
>Lest UNIX nerds start getting smug, there are toy UNIX platforms too.
>Plus, any UNIX system can be made that way with a single "chmod".

Note, when I write "TOPS" I mean TOPS-10 (since that's what people have been
talking about) or possibly TOPS-20 if applicable.

I've become much more vigilant lately about spotting claims without evidence. 
Please clarify:

	- What are the toy UNIX platforms?

	- Why are they toys?

I thought you might be thinking of Linux, in which case I question your claim.
it seems to me that the CPUs in current PCs and Macs are perfectly capable
of memory protection; the main reason Macintoys and Piece'o'Craps are toys
(or pieces o' crap) is that their OSs don't make enough use of the hardware
features.  I understand that Linux makes better use of the hardware.

If you're not thinking of Linux but another toy UNIX system, then just erase
the paragraph above.  If you _are_ thinking of Linux, then the burden is on
you to explain why Linux is deficient (and why the PDP-10 systems are
better).  You could also explain why PC/Mac hardware is deficient (and why
PDP-10 hardware is better).

As for the problem with "chmod" -- It's fair to criticize UNIX's security
model (and explain why).  You could certainly say that security which
depends on a large number of cooperating programs (all of which must be
"chmod"ed correctly) is weak. But you then have to explain how TOPS solves
that problem (and if it creates any new problems).

But any OS will fail if you set its security loose enough, and I expect that
TOPS has that problem in some form.  What about the batch system?  What
about the various GEN processes?  I bet there's some part of TOPS that needs
privileges and can cause havoc if those privileges aren't set right.

Look... you told me yourself that a lot of the info about the -10 is gone.
Here we have a bunch of smart curious people wanting to know more.  For
those two reasons, your opinion is very important!  But you do everyone a
disservice by writing such sweeping statements.  As well as being smart and
curious, we're pretty ignorant of the -10, so we can't fill in the gaps of
your claims to make a fair judgement.  You'll just have to deal with that.

And the UNIX fans have the same faults, BTW.  I saved some messages from
Peter da Silva; I'm going to e-mail him as soon as I've figured out exactly
what I want to say.

Sorry if I sound too harsh... should we just say that Monday's to blame all
around?

-- Derek

######

From: rob.nicholson@zetnet.co.uk (Rob Nicholson)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 26 Apr 1999 22:51:54 GMT
Organization: Rainow Systems Ltd.
Reply-To: rob.nicholson@zetnet.co.uk
Message-ID: <3727306c.2212246@news.zetnet.co.uk>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <slrn7hrjii.dbl.roggblake@unix5.netaxs.com> <37222d81.1058246@news.zetnet.co.uk> <7fpprq$fps$1@autumn.news.rcn.net> <3721821d.6019996@news.zetnet.co.uk> <slrn7i6c5e.43b.roggblake@unix3.netaxs.com>
X-Newsreader: Forte Agent 1.5/32.452
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
NNTP-Posting-Host: manr-037.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 925167125 498 194.247.43.167
Lines: 6
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!128.230.129.112.MISMATCH!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!dispose.news.demon.net!demon!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail

>Just the kind of "feature" your business wants in its servers.. :-)

I remember reading an article about a 3D user interface. Maybe that's what it's
really for :-)

Rob.

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
From: alderson@netcom19.netcom.com (Richard M. Alderson III)
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
In-Reply-To: jmfbahciv@aol.com's message of Sat, 24 Apr 99 09:06:38 GMT
Message-ID: <ALDERSON.99Apr26162045@netcom19.netcom.com>
Sender: alderson@netcom19.netcom.com
Reply-To: alderson@netcom.com
Organization: NETCOM On-line services
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu>
	<371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net>
	<Ci0U2.24$MF5.9024@srvr1.engin.umich.edu>
	<ALDERSON.99Apr23113531@netcom.netcom.com>
	<7fs7mv$hii$1@ligarius.ultra.net>
Date: Mon, 26 Apr 1999 23:20:45 GMT
Lines: 17
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.axxsys.net!newsfeed.enteract.com!ix.netcom.com!netcom19!alderson

In article <7fs7mv$hii$1@ligarius.ultra.net> jmfbahciv@aol.com writes:

>In article <ALDERSON.99Apr23113531@netcom.netcom.com>,
>   alderson@netcom.netcom.com (Richard M. Alderson III) wrote:

>>In later versions of Tops-10

>Did you really mean TOPS-10 here, Rich?  Typos, when talking about monitors
>can get terribly misleading :-).  Especially when youngsters are really trying
>to learn about designs that have been proven to work. :-)

Hmm.  I was taught "Tops-10" (to match DECsystem-10) and "TOPS-20" (to match
DECSYSTEM-20).  Sorry, Barb.
-- 
Rich Alderson			Last LOTS Tops-20 Systems Programmer, 1984-1991
				Current maintainer, MIT TECO EMACS (v. 170)
last name @ XKL dot COM		Chief systems administrator, XKL LLC, 1998-now

######

Newsgroups: alt.folklore.computers
From: alderson@netcom19.netcom.com (Richard M. Alderson III)
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
In-Reply-To: lisard@zetnet.co.uk's message of 25 Apr 1999 19:48:39 GMT
Message-ID: <ALDERSON.99Apr26165228@netcom19.netcom.com>
Sender: alderson@netcom19.netcom.com
Reply-To: alderson@netcom.com
Organization: NETCOM On-line services
References: <37237197$0$496@news.zetnet.co.uk>
Date: Mon, 26 Apr 1999 23:52:28 GMT
Lines: 18
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newspump.monmouth.com!newspeer.monmouth.com!newsfeed.enteract.com!ix.netcom.com!netcom19!alderson

In article <37237197$0$496@news.zetnet.co.uk> lisard@zetnet.co.uk writes:

>What do you mean by "small computer thinking"? To my mind, it just doesn't
>make much sense to discontinue a proven and well-loved architecture just
>because it's not quite the latest thing. For example, if the -10 became
>underpowered, why not simply recast it in silicon and move it down the range?
>The core processor would appear to be simple enough that this presents no
>great challenge to the hardware. And then we might have had workstations with
>PDP-10 chips in them, for example...

The founder of cisco Systems had this great idea when he worked at DEC in the
70s, a Ten On A Desk.  That was the original plan for cisco, too.  It's a bit
large to stick on a desktop, but it eventually got done.  (Should I mention
that he started another company after leaving cisco?)
-- 
Rich Alderson			Last LOTS Tops-20 Systems Programmer, 1984-1991
				Current maintainer, MIT TECO EMACS (v. 170)
last name @ XKL dot COM		Chief systems administrator, XKL LLC, 1998-now

######

From: korpela@islay.ssl.berkeley.edu (Eric J. Korpela)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 27 Apr 1999 00:13:41 GMT
Organization: Cal Berkeley-- Space Sciences Lab
Lines: 29
Message-ID: <7g2vfl$akj$1@agate.berkeley.edu>
References: <371be3e4.0@newsfeed.one.net> <7fv0l4$8v5$2@antiochus.ultra.net> <7fvhpn$241@bonkers.taronga.com> <7g0con$68m@fohnix.metronet.com>
NNTP-Posting-Host: islay.ssl.berkeley.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!sunqbc.risq.qc.ca!newsfeed.berkeley.edu!agate!islay.ssl.berkeley.edu!korpela

In article <7g0con$68m@fohnix.metronet.com>,
Stephen Jones <smj@fohnix.metronet.com> wrote:
>Peter - I did not mean to link zombies with job control.  I know they have
>nothing to do with that as UNIX has had the problem of "the process won't
>die" and "the child process with no parent" and "the run away process that
>lost its TTY and is now eating up CPU and memory".  They are all crocking
>UNIXism. 

When's the last time you used a UNIX system.  As far as the above "examples,"
I've never run into one that couldn't be solved by "kill".

>          Just reboot and everything is fine.  That is the philosophy.  There
>is not concept of attach a debugger to the job/process and fix it (unless
>you are running HURD?)  

I've spent much of the day connecting and disconnecting gdb from running
processes.  I'm certainly not running HURD.


>                       I'm sorry, with my understanding of UNIX internals
>I should be vomiting in a toilet rather than writing garbage here :)

Maybe you should be eating from one.

Eric
-- 
Eric Korpela                        |  An object at rest can never be
korpela@ssl.berkeley.edu            |  stopped.
<a href="http://sag-www.ssl.berkeley.edu/~korpela">Click for home page.</a>

######

From: michael.wojcik@merant.com (Michael Wojcik)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 27 Apr 1999 01:30:56 GMT
Organization: MERANT Inc.
Lines: 28
Message-ID: <7g340g$1407@news1.newsguy.com>
References: <371be3e4.0@newsfeed.one.net> <7g0con$68m@fohnix.metronet.com> <7g0usf$4lg@weyl.math.psu.edu> <H20V2.1086$WA4.216881@ptah.visi.com> <7g23p5$559@weyl.math.psu.edu>
NNTP-Posting-Host: p-838.newsdawg.com
X-Newsreader: xrn 9.00
Originator: mww@raederle.microfocus.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.berkeley.edu!pln-w!spln!extra.newsguy.com!newsp.newsguy.com!mww


In article <7g23p5$559@weyl.math.psu.edu>, viro@weyl.math.psu.edu (Alexander Viro) writes:
> 
> 	Been there, seen that. Program that ignores the return value of
> function it called *is* badly written. AFAIK no OS is able to stop program
> from loop: goto loop;.

Except that looping programs consume CPU time, and many OSes kill off
processes that consume more than a set amount of CPU time.  Like most
Unix variants, for example.

Sometimes this can be a real hassle.  I ran into a case on CICS,
which limits how many CPU instructions a transaction can perform
between issuing requests to CICS - a peculiar feature somewhere in
the middle ground between "cooperative" and preemptive multitasking.
It would kill my cryptographic-hash routine...  Since the hash
didn't need to be very strong anyway (it wasn't used to secure
anything important, just the software license key), I just reduced
the number of rounds.


Michael Wojcik          	        michael.wojcik@merant.com
AAI Development, MERANT                 (block capitals are a company mandate)
Department of English, Miami University

Shakespeare writes bombast and knows it; Mr Thomas writes bombast and
doesn't.  That is the difference.
-- Geoffrey Johnson

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 27 Apr 99 07:43:24 GMT
Organization: UltraNet Communications, Inc.
Lines: 27
Message-ID: <7g3vvp$qnq$1@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu>
	<371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net>
	<Ci0U2.24$MF5.9024@srvr1.engin.umich.edu>
	<ALDERSON.99Apr23113531@netcom.netcom.com>
	<7fs7mv$hii$1@ligarius.ultra.net> <ALDERSON.99Apr26162045@netcom19.netcom.com>
NNTP-Posting-Host: d14.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 27 Apr 1999 09:28:25 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!195.211.211.32.MISMATCH!newscore.gigabell.net!newscore.ipf.de!news.ndh.net!newsfeed.tli.de!newsfeed.enteract.com!newsswitch.lcs.mit.edu!news.ultranet.com!d14

In article <ALDERSON.99Apr26162045@netcom19.netcom.com>,
   alderson@netcom19.netcom.com (Richard M. Alderson III) wrote:
>In article <7fs7mv$hii$1@ligarius.ultra.net> jmfbahciv@aol.com writes:
>
>>In article <ALDERSON.99Apr23113531@netcom.netcom.com>,
>>   alderson@netcom.netcom.com (Richard M. Alderson III) wrote:
>
>>>In later versions of Tops-10
>
>>Did you really mean TOPS-10 here, Rich?  Typos, when talking about 
monitors
>>can get terribly misleading :-).  Especially when youngsters are really 
trying
>>to learn about designs that have been proven to work. :-)
>
>Hmm.  I was taught "Tops-10" (to match DECsystem-10) and "TOPS-20" (to 
match
>DECSYSTEM-20).  Sorry, Barb.

Oh, no.  I wasn't quibbling about caps.  I was asking about 
"TOPS-10" vs. "TOPS-20".  If we're discussing the differences
between the two products, a typo of 1 or 2 confuses everything...
just as it did here [wry emoticon here]./

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 27 Apr 99 07:57:55 GMT
Organization: UltraNet Communications, Inc.
Lines: 37
Message-ID: <7g40r0$qnq$4@antiochus.ultra.net>
References: <37237197$0$496@news.zetnet.co.uk> <ALDERSON.99Apr26165228@netcom19.netcom.com>
NNTP-Posting-Host: d14.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 27 Apr 1999 09:42:56 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.ecrc.net!newsfeed.usit.net!newsswitch.lcs.mit.edu!news.ultranet.com!d14

In article <ALDERSON.99Apr26165228@netcom19.netcom.com>,
   alderson@netcom19.netcom.com (Richard M. Alderson III) wrote:
>In article <37237197$0$496@news.zetnet.co.uk> lisard@zetnet.co.uk writes:
>
>>What do you mean by "small computer thinking"? To my mind, 
>>it just doesn't
>>make much sense to discontinue a proven and well-loved architecture just
>>because it's not quite the latest thing. For example, if the -10 became
>>underpowered, why not simply recast it in silicon and move it 
>>down the range?
>>The core processor would appear to be simple enough that this presents no
>>great challenge to the hardware. And then we might have had 
>>workstations with
>>PDP-10 chips in them, for example...
>
>The founder of cisco Systems had this great idea when he 
>worked at DEC in the 70s, a Ten On A Desk.  That was the 
>original plan for cisco, too.  It's a bit
>large to stick on a desktop, but it eventually got done.  
>(Should I mention
>that he started another company after leaving cisco?)

It doesn't hurt :-).  That PDP-10 on a desk was only one of
a lot of hardware projects cancelled that would have 
enhanced the product.  The first serious blow that I remember
is the cancellation of the (now, here we go again with my
inability to remember the cybercrud) MG20..I think that's
what it was called.   Anyway, it was a memory box that had
many ports, but I don't recall the specs.  With the advent
if implementing SMP in TOPS-10, we would have been able to
sell our own hardware rather than giving AMPEX all the memory
business.

/BAH


Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 27 Apr 99 08:09:12 GMT
Organization: UltraNet Communications, Inc.
Lines: 50
Message-ID: <7g41g4$qnq$5@antiochus.ultra.net>
References: <3723719a$0$496@news.zetnet.co.uk>
NNTP-Posting-Host: d14.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 27 Apr 1999 09:54:12 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!newsswitch.lcs.mit.edu!news.ultranet.com!d14

In article <3723719a$0$496@news.zetnet.co.uk>, lisard@zetnet.co.uk wrote:
>
>
>On 1999-04-25 jmfbahciv@aol.com said:
>   :In article <3721db02$0$493@news.zetnet.co.uk>, lisard@zetnet.co.uk
>   :wrote: >
>   :>Miaow. :>
>
>   :You've never tried to get something done with that guy in charge.
>   :Until then...I reserve my right to express my cats.
>
>Tell us more. This is exactly the kind of perspective we could use for
>working out at least some of the reasons for NT being a steaming turd of
>an operating system...

I told you.  Let me parse the phrase for you...small [computer thinking].
The following are examples of small computer thinking and not to be
construed as the opinions of any individual, group, etc.

For instance, once upon a time, there were arugments about sticking
a disk drive on a mini.  The reasoning was that nobody would
need that much storage space.  Later on, after customers proved that
they did want disk space, another statement was made, there's no
need for a multi-pack disk structure, besides, it's hard to do.
There also existed another attitude which dismissed software
as a necessity of selling computer hardware.

Now, to talk about this particular individual....
The project that I was working on should have been the sexiest
and first typesetting programming system at DEC.  This guy
was the manager of the project.  Everybody was afraid of him
and spent more time devising coping plans to either not piss
him off or to placate him rather than actually work building
the project.  I was not allowed in the same room as him.  My
supervisor thought that, if I was in the same room with him,
blood would be let, and she didn't know his blood type.  Not
a line of code was written, but there were lots and lots of
meetings talking about it.  The guy also had a very weird
sense of what a computer should do and what a user should do.
He wanted to hogtie the user so that the only thing the user
would be able to do is type only with one's little toe (that's
an exaggeration, by the way).  I don't recall the specifics
but I had never been so disgusted in my life.

/BAH


/BAH

Subtract a hundred and four for e-mail.

######

From: cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 27 Apr 1999 08:11:07 GMT
Organization: Honest Chris' Sysadmin Emporium
Message-ID: <7g3rer$s1d$1@teabag.demon.co.uk>
References: <371be3e4.0@newsfeed.one.net> <7fta87$ccc@bonkers.taronga.com>
    <7fv0l4$8v5$2@antiochus.ultra.net> <7fvhpn$241@bonkers.taronga.com>
    <7g0con$68m@fohnix.metronet.com>
    <37241351.F947285A@tnglwood.demon.co.uk>
    <Pine.NXT.4.20.9904261200000.15191-100000@Tomobiki-Cho.CAC.Washington.EDU>
NNTP-Posting-Host: localhost
X-NNTP-Posting-Host: teabag.demon.co.uk:193.237.4.110
X-Trace: news.demon.co.uk 925203509 nnrp-11:5123 NO-IDENT teabag.demon.co.uk:193.237.4.110
X-Complaints-To: abuse@demon.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Newsreader: knews 1.0b.0
Lines: 19
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!newsfeed1.uni2.dk!newsfeed.tli.de!newsfeed.nacamar.de!dispose.news.demon.net!demon!news.demon.co.uk!demon!teabag.demon.co.uk!not-for-mail

In article <Pine.NXT.4.20.9904261200000.15191-100000@tomobiki-cho.cac.washington.edu>,
	Mark Crispin <mrc@CAC.Washington.EDU> writes:
> Be careful.  There are those of us who know quite a bit about UNIX and
> Christianity, and can say with some authority that the main thing wrong
> with criticism of either is that it is understated.

For whom do you speak with said authority?  Certainly not me; there's a
place for constructive criticism about either, sure, but I think the
point being made was that both attract unwarranted, uninformed criticism
with tiresome regularity; examples are in this very thread.

> This is not the forum to continue this discussion.

I agree, but why try to round it off with such a contentious assertion?
People will see it as simply trying to have the last word by playing the
"this is inappropriate discussion" card, which doesn't work as you've
already contravened that claim in your own post by continuing it yourself.

Chris.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 27 Apr 99 08:17:21 GMT
Organization: UltraNet Communications, Inc.
Lines: 37
Message-ID: <7g41ve$qnq$6@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <ALDERSON.99Apr23113531@netcom.netcom.com> <Gs5U2.29$MF5.10448@srvr1.engin.umich.edu> <7g18s4$75o$1@shell3.ba.best.com> <uiXU2.42$MF5.14297@srvr1.engin.umich.edu>
NNTP-Posting-Host: d14.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 27 Apr 1999 10:02:22 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!news.ultranet.com!d14

In article <uiXU2.42$MF5.14297@srvr1.engin.umich.edu>,
   ftit@engin.umich.edu (Sergej Roytman) wrote:
>In article <7g18s4$75o$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> 
wrote:
>>In article <Gs5U2.29$MF5.10448@srvr1.engin.umich.edu>,
>>Sergej Roytman <ftit@engin.umich.edu> wrote:
>>  GET MTA0:BACKUP.EXE 40K	;From 9-track tape, assign 40Kwords of core
>
>And those 40K (virtual? (swapped? paged?)) memory?  Were they protected
>from other users and their sloppy processes by software or specialized
>hardware in the PDP-10?
>
I've been hoping that people who knew the internals would
answer your questions.  But they seem to have digressed in
pissing mode.  First of all, there were no sloppy processes in
both the -10 and -20 operating systems.  If there was any slop,
it was considered a serious bug and fixed.  Virtual and paged 
memory were implementations that JMF did.  He wrote VMSER.MAC 
and the paging flavors were implemented based on the flavor
of the CPU.  We had KI-paging and KL-paging which (I think)
was different types of hardware technology.  I should warn
the readers that anything I say about the hardware should be
taken with a ton of salt because I usually ignored that part
of our biz; there were all those other guys to take care of it.

Swapping was a part of memory management and a bane in JMF's
and TW's side; I think SWPSER.MAC was one of the
implemetations that they would have liked to have done over.
Any modifications to SWPSER.MAC was always negotiated between
the two.  I think they final decided to just take turns.

/BAH
 

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 27 Apr 99 08:21:04 GMT
Organization: UltraNet Communications, Inc.
Lines: 27
Message-ID: <7g426c$qnq$7@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7fk9i4$spj$1@ligarius.ultra.net> <BlpT2.3$MF5.1852@srvr1.engin.umich.edu> <7fmv2b$591$2@antiochus.ultra.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu> <371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net> <372488B1.8C38DC2C@thinkage.on.ca>
NNTP-Posting-Host: d14.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 27 Apr 1999 10:06:04 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!howland.erols.net!newsswitch.lcs.mit.edu!news.ultranet.com!d14

In article <372488B1.8C38DC2C@thinkage.on.ca>,
   "Alan T. Bowler" <atbowler@thinkage.on.ca> wrote:
>jmfbahciv@aol.com wrote:
>> 
>
>> >The nice thing was, though, you didn't really notice the other users;
>> >response time was usually very good.
>> 
>> Yup.  That's my point.  For all intents and purposes, one had
>> the feel of their very own private computer.
>
>In some respects it was a litle better.  The files on those timeshared
>systems were reliably backed up.

That's a really important point.  The advantage (and, boy, do I
miss it now) was that the operations and hardware support service
was an unseen benefit.  One didn't have to be one's own field
service guy when a piece of hardware breaks.  All I had to do
is call loudly, "FIELD SERVICE", make sure they heard me, write
down the problem in the log, and go do something else (like
design the next release at the local bar).

The same thing with backups. 

/BAH

Subtract a hundred and four for e-mail.

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 27 Apr 1999 08:43:34 -0500
Organization: none
Lines: 43
Message-ID: <7g4eu6$is7@bonkers.taronga.com>
References: <371be3e4.0@newsfeed.one.net> <7g0usf$4lg@weyl.math.psu.edu> <H20V2.1086$WA4.216881@ptah.visi.com> <p98mccabe-2604991226580001@mepa2pp48.alltel.net>
NNTP-Posting-Host: bonkers.in.taronga.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!logbridge.uoregon.edu!newsfeed.stanford.edu!news.ems.psu.edu!bonkers!not-for-mail

In article <p98mccabe-2604991226580001@mepa2pp48.alltel.net>,
Micheal H. McCabe <p98mccabe@alltel.net> wrote:
>How did you do this sort of thing under TOPS-10 or TOPS-20?  Also, my
>understanding of UNIX was that each invocation of a command generally
>created an independant process.  Is this an incorrect assumption?

That's correct. UNIX processes are relatively lightweight, and in fact most
traditional UNIX programs are intended to run as a part of a pipeline of
cooperating processes, each of which does a part of the job: formatting
a file or generating input, snipping out portions of that stream, sorting
or reformatting it for display or further processing.

The heavy duty interactive programs that people are complaining about (like
your mail reader, or editor) are an exception. Traditionally, you spend most
of your time either editing or running these short pipelines. Apart from
the editor, interactive programs were rare.

Large interactive programs that serve a lot of different purposes were
generally ports from other operating systems (emacs, for example) or written
to emulate programs used in other environments (I believe that PINE falls
into this category). As these have become more popular this non-UNIXy mode
of use has spawned a lot of problems: buggy system calls (the looping
process problem) and security problems (caused usually by having a large
complex program that also requires special privileges to run).

It's a different approach to a system design from the traditional monitors,
and you have different tradeoffs, some things are easier, some things are
harder... I've used all sorts of operating systems from pure batch systems
through to microkernels, and all of them have features I like and dislike.

And that includes UNIX and TOPS-20. I found UNIX a relative relief after TOPS:
it was (back in Version 6 days, anyway) much simpler and easier to understand.
I missed the rich terminal driver in TOPS, though... the V6 driver didn't even
properly handle kill-line (it just printed the kill character and a newline).

Hell, there's even things about Windows I like. But I better stop here before
I'm lynched by BOTH sides.

-- 
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
"Twenty-two points, plus triple-word-score, plus fifty points for using all
 my letters.  Game's over.  I'm outta here." -- Melissa ... by Kwyjibo.

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 27 Apr 1999 08:55:34 -0500
Organization: none
Lines: 36
Message-ID: <7g4fkm$j34@bonkers.taronga.com>
References: <371be3e4.0@newsfeed.one.net> <7g1b2h$ifj$1@shell3.ba.best.com> <7g1sdl$4vs@weyl.math.psu.edu> <7g47sv$rg5$2@antiochus.ultra.net>
NNTP-Posting-Host: bonkers.in.taronga.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!logbridge.uoregon.edu!newsfeed.stanford.edu!news.ems.psu.edu!bonkers!not-for-mail

In article <7g47sv$rg5$2@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
>On the other hand, the advantage of having the device drivers in
>the monitor was that the code for that device driver did not
>have to be included in the user's address space.  I have
>no idea how UNIX deals with this.

UNIX device drivers are typically relatively lightweight: they provide just
enough functionality to expose a uniform interface to the layer above (the
filesystem layer, for disks, or the stream I/O layer for devices like serial
ports or tapes, or ptys). With System V a uniform "terminal input editing"
layer was created so that ptys and serial ports could just expose a stream
to that. You could also slide network stacks (PPP, for example) or specialized
I/O drivers (for real-time protocols, MIDI, and the like) in at the streams
layer.

The number of these layers has grown as different kinds of devices become
available (network interfaces or bitmal displays), and shrunk as common
features allowed multiple kinds of devices to use a common interface (ptys,
for example, or the System V streams model).

Some things have never been moved below the kernel layer. Cursor addressing,
for example. Others have been pulled out and moved into user space (the
Version 6 and version 7 printer drivers knew a lot about line printers, but
these days all parallel ports look the same and filters are used to deal
with eccentric hardware).

But in general the traditional UNIX model has had the device drivers compiled
into the kernel, and you needed a system build to change them. Loadable device
drivers are a common elaboration, now, though of course you need to at least
have all the drivers for boot compiled in.

-- 
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
"Twenty-two points, plus triple-word-score, plus fifty points for using all
 my letters.  Game's over.  I'm outta here." -- Melissa ... by Kwyjibo.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 27 Apr 99 09:31:32 GMT
Organization: UltraNet Communications, Inc.
Lines: 38
Message-ID: <7g46ag$rg5$1@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7fvhpn$241@bonkers.taronga.com> <7g0con$68m@fohnix.metronet.com> <7g0usf$4lg@weyl.math.psu.edu> <H20V2.1086$WA4.216881@ptah.visi.com> <p98mccabe-2604991226580001@mepa2pp48.alltel.net>
NNTP-Posting-Host: d9.dial-14.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 27 Apr 1999 11:16:32 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news.tvd.be!remarQ-easT!supernews.com!remarQ.com!newsswitch.lcs.mit.edu!news.ultranet.com!d9

In article <p98mccabe-2604991226580001@mepa2pp48.alltel.net>,
   p98mccabe@alltel.net (Micheal H. McCabe) wrote:
>
>OK folks, here's where my ignorance really starts to show...
>
>To begin with, I know nothing about UNIX internals.  My experiences with
>time-sharing are limited to TS/OS on the Univac 90/60 (An IBM 360 clone),
>VAX/VMS (Somewhere around version 2.? - Later upgraded to 4.?), and RT-11
>on the pdp-11.
>
>ISTR, when you submitted a DO file under VMS or RT-11 (hazy recall of an
>@filename command), it appeared as if commands were entered from your
>terminal and executed directly.  Output was displayed (or printed) on your
>terminal unless redirected to a file.  During execution of this command
>file, the only actions possible from the terminal keyboard were CTRL-S to
>stop scrolling, CTRL-Q to resume, CTRL-C to interrupt execution of the
>current command, or CTRL-Z to abort the whole thing.
>
>Likewise, when you used SUBMIT to enter a batch job, an independant
>process was created that would execute until it completed its task, ran
>out of CPU time, or was killed by the user (or system operator.)  Once the
>job was submitted, your terminal was free to do additional work.
>
>Did I miss some fine distinction here?
>
>How did you do this sort of thing under TOPS-10 or TOPS-20?

Similarly, except for the use of the ^Z character.

>  Also, my
>understanding of UNIX was that each invocation of a command generally
>created an independant process.  Is this an incorrect assumption?

I don't know.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 27 Apr 99 09:58:25 GMT
Organization: UltraNet Communications, Inc.
Lines: 109
Message-ID: <7g47sv$rg5$2@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu> <7g1b2h$ifj$1@shell3.ba.best.com> <7g1sdl$4vs@weyl.math.psu.edu>
NNTP-Posting-Host: d9.dial-14.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 27 Apr 1999 11:43:27 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!news.shore.net!news.ultranet.com!d9

In article <7g1sdl$4vs@weyl.math.psu.edu>,
   viro@weyl.math.psu.edu (Alexander Viro) wrote:
>In article <7g1b2h$ifj$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> 
wrote:
>[snip]
>>>	Life with ulimit -u 4 <shudder> Was there any equivalent of virtual
>>>consoles?
>>
>>Nope.  But if you had a key to the Computer Center (back 
>>when the Terminal
>>Room was not open 24 hours) and came in after hours, you 
>>could log in separate
>>jobs on the 12 ADM3A terminals and use a chair with wheels 
>>to do multitasking between the 12 keyboards.
>
>	<chuckle> Could the equivalent be done by allocating 
>12 PTYs, 

Yes.

>putting TTY into the raw mode, 

I don't know what you mean by raw mode.

>reading from it, watching for switch characters and
>feeding the input to appropriate PTY? (with corresponding 
>stuff for output, indeed). 

The device PTY was used usually via the program OPSER.  One
could login a number of jobs (there were admistrative 
settable limits that had to do with timesharing access).
Each job's terminal input and output was identified with
prefixes.  So if PTY0 was the job in question.  One 
communicated with it by prefixing all lines with the two
characters 0-.  But that was a functional spec detail of
the CUSP OPSER.MAC.  In the monitor, PTYs and TTYs were
handled slightly differently but I don't know how.

>BTW, here's a serious design question - AFAICS large chunk of the
>functionality of shell was put into the terminal driver, right? 

If I understand your question, the answer is no.  Most of the
"shell" part was in the stand alone program we called OPSER.
The address space of the user didn't mix with the address space
of the monitor.

>I.e. together with the equivalent of cooked mode. 

I don't know what "cooked mode" means.

>Moreover, equivalent of process control
>primitives was done via feeding the commands into that driver (BTW, was it
>available via separate syscalls?)

The monitor decided when and where the job was going to run, i.e.,
execute an instruction or set of instructions in behalf of the user.  
The instructions to be executed were in the control of the user 
via the user mode program he ran.  If the user mode program 
required more input or desired to output from/to the terminal, 
the request to do that was handled via a syscall (we called 
them UUOs on the -10 and JSYSs on the -20); on the -10 that 
was an extended UUO called TTCALL.  When a program wanted to 
do I/O to the disk, it would ask the monitor for service.  
That way a user mode program did not have to know the 
hardware architecture.  That was a knowledge
left to the monitor.  This design allowed user mode programs to
run on any system that ran TOPS-10, no matter what the hardware
was.

> How much of that was duplicated over the
>TTY/PTY/console drivers and how much was really independent from the type
>of terminal? IOW, was there a separation between line discipline and 
>terminal driver?

TOPS10 didn't have a terminal driver for every flavor
of terminal.  The screen addressing was settable by the job
via SETUUOs.  

> With a very little hackery the stuff you are describing can be
>implemented as LDISC on UNIX. Hmm... That leads to the next question: how
>monolitic the monitor was? That is, were there standard interfaces (within
>the monitor) that could be used for writing a new LDISC/device driver/fs/
>executable format? 

Not really.  One of limitations of TOPS10 was that any additional
new hardware required a rebuild of the monitor if the parameters
or device drivers hadn't been included in the last build.  There
wasn't the notion of plug-n-play in the same vein as one sees today
with PCs.  There was very little extensibility during uptime.  SMP
implementation addressed quite a bit of that but we couldn't solve
those problems because of the data base TOPS10 had designed into
it.  

On the other hand, the advantage of having the device drivers in
the monitor was that the code for that device driver did not
have to be included in the user's address space.  I have
no idea how UNIX deals with this.

>I know very well what a PITA is to pick the corresponding
>pieces from existing parts of kernel if originally they were scattered all
>over the place. To what extent it was done in case of TOPS-10?

I'm not sure what you are talking about here.  Are you talking
about source management or are you talking about code organization?

/BAH


Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 27 Apr 99 10:03:07 GMT
Organization: UltraNet Communications, Inc.
Lines: 39
Message-ID: <7g485m$rg5$3@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7g0con$68m@fohnix.metronet.com> <7g0usf$4lg@weyl.math.psu.edu> <H20V2.1086$WA4.216881@ptah.visi.com> <7g23p5$559@weyl.math.psu.edu>
NNTP-Posting-Host: d9.dial-14.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 27 Apr 1999 11:48:06 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newspump.monmouth.com!newspeer.monmouth.com!news.ultranet.com!d9

In article <7g23p5$559@weyl.math.psu.edu>,
   viro@weyl.math.psu.edu (Alexander Viro) wrote:
>In article <H20V2.1086$WA4.216881@ptah.visi.com>,
>Peter Seebach <seebs@plethora.net> wrote:
>>In article <7g0usf$4lg@weyl.math.psu.edu>,
>>Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>>>lost its TTY and is now eating up CPU and memory".  
>>>>They are all crocking
>>>	Huh?
>>
>>There's a common problem (mostly this used to bite us 
>>with PINE) with crufty
>>software that doesn't notice EOF on stdin, and sits around 
>>spinning on some
>>read loop forever.  Many applications will be vulnerable 
>>if they're denied
>>SIGHUP, but some just do it all the time anyway.  There's 
>>nothing the OS can
>>plausibly do to distinguish this from a program which is 
>>merely badly written,
>>in general.  And you can, of course, always kill 'em.
>
>	Been there, seen that. Program that ignores the return value of
>function it called *is* badly written. AFAIK no OS is able to stop program
>from loop: goto loop;. Fresh example: nethack. Function reads a character,
>checks if it belongs to <set> and either returns the index 
>in set or rereads.  Ignores EOF.

Why should an OS stop a user mode program from looping?  If that's
what the user wants, the OS should give it him.  It's just another
job, albeit probably doing useless work.  It is the job of the
OS to ensure that a looping job doesn't hog the system precluding
any other job from running.  If runnable jobs don't get a slice
of execution time, then it's a very bad bug.

/BAH


Subtract a hundred and four for e-mail.

######

Sender: eric@ruckus.brouhaha.com
From: Eric Smith <eric-no-spam-for-me@brouhaha.com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7fv0l4$8v5$2@antiochus.ultra.net> <7fvhpn$241@bonkers.taronga.com> <7g0con$68m@fohnix.metronet.com> <7g2vfl$akj$1@agate.berkeley.edu> <3728e2b4.8169296@news.ford.com>
X-Disclaimer: Everything I write is false.
Organization: Eric Conspiracy Secret Labs
X-Eric-Conspiracy: There is no conspiracy.
Date: 27 Apr 1999 14:59:17 -0700
Message-ID: <qhu2u1d8sq.fsf@ruckus.brouhaha.com>
X-Newsreader: Gnus v5.5/Emacs 20.2
NNTP-Posting-Host: ruckus.brouhaha.com
X-Trace: 27 Apr 1999 14:04:37 -0700, ruckus.brouhaha.com
Lines: 41
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!newsfeed.berkeley.edu!enews.sgi.com!news.sgi.com!news.spies.com!ruckus.brouhaha.com

cpierce1@ford.com (Clinton Pierce) writes:
> It's quite possible in UNIX to create a process that will never completely
> go away.

As someone else pointed out, that is due to buggy device drivers, not due
to "UNIX".

> A process read-blocking on a hard-mounted NFS filesystem that's "gone
> away" will remain forever.  You can't get rid of this process.  You can
> kill it, but it won't go away completely.

That's because NFS is seriously brain-damaged in both design and
implementation.  Nothing about UNIX mandated that NFS be designed or
implemented in such a way as to cause this problem.

In fact, this "problem" is nothing more than the deliberately designed
semantics of NFS hard mounts.  If this causes a problem for you, you should
ask yourself *why* you are using hard mounts.  And also, perhaps, why you are
using NFS.

> A process that was waiting for a slow device (like a tape drive) and the
> device gets wedged--that process never goes away.

Only if the device driver is badly written.  On my systems, I routinely ^C
out of lengthy I/O operations, *especially* if I know that external conditions
are such that the operation will never complete.

Again, this is not a fundamental problem with UNIX.  Any monolithic
operating system is unable to protect against broken device drivers.  A
broken driver in TOPS-10 can just as easily cause problems.

AFAIK, though, the TOPS-10 monitor and device drivers were usually pretty well
debugged and tested by the time they shipped to customers.  This was a much
different situation from UNIX, in that TOPS-10 was only supplied by a single
vendor, and only had to support that vendor's hardware.  When you say "UNIX",
you aren't referring to a specific operating system, much less a specific
vendor.

######

From: cpierce1@ford.com (Clinton Pierce)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 27 Apr 1999 16:20:58 GMT
Organization: Hardly
Lines: 24
Message-ID: <3728e2b4.8169296@news.ford.com>
References: <371be3e4.0@newsfeed.one.net> <7fv0l4$8v5$2@antiochus.ultra.net> <7fvhpn$241@bonkers.taronga.com> <7g0con$68m@fohnix.metronet.com> <7g2vfl$akj$1@agate.berkeley.edu>
NNTP-Posting-Host: 19.88.81.187
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Newsreader: Forte Agent 1.5/32.452
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!cam-news-hub1.bbnplanet.com!cpk-news-hub1.bbnplanet.com!news.gtei.net!cliffs.rs.itd.umich.edu!jobone!dailyplanet.srl.ford.com!eccws1.dearborn.ford.com!news

On 27 Apr 1999 00:13:41 GMT, korpela@islay.ssl.berkeley.edu (Eric J.
Korpela) wrote:

>In article <7g0con$68m@fohnix.metronet.com>,
>Stephen Jones <smj@fohnix.metronet.com> wrote:
>>[...] "the process won't die" 
>
>When's the last time you used a UNIX system.  As far as the above "examples,"
>I've never run into one that couldn't be solved by "kill".

It's quite possible in UNIX to create a process that will never completely
go away.

A process read-blocking on a hard-mounted NFS filesystem that's "gone
away" will remain forever.  You can't get rid of this process.  You can
kill it, but it won't go away completely.

A process that was waiting for a slow device (like a tape drive) and the
device gets wedged--that process never goes away.

-- 
Clinton A. Pierce       "If you rush a Miracle Man, you get rotten
clintp@geeksalad.org        Miracles."  -- Miracle Max, The Princess Bride
http://www.geeksalad.org

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 27 Apr 1999 18:05:56 -0400
Organization: -ENOENT
Lines: 23
Message-ID: <7g5cc4$6lf@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <H20V2.1086$WA4.216881@ptah.visi.com> <7g23p5$559@weyl.math.psu.edu> <7g485m$rg5$3@antiochus.ultra.net>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!sunqbc.risq.qc.ca!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7g485m$rg5$3@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
>In article <7g23p5$559@weyl.math.psu.edu>,
>   viro@weyl.math.psu.edu (Alexander Viro) wrote:
>>	Been there, seen that. Program that ignores the return value of
>>function it called *is* badly written. AFAIK no OS is able to stop program
>>from loop: goto loop;. Fresh example: nethack. Function reads a character,
>>checks if it belongs to <set> and either returns the index 
>>in set or rereads.  Ignores EOF.
>
>Why should an OS stop a user mode program from looping?  If that's
	Exactly.
>what the user wants, the OS should give it him.  It's just another
>job, albeit probably doing useless work.  It is the job of the
	See above - 99:1 that it's a bug in the program and OS has no business
stopping it.
>OS to ensure that a looping job doesn't hog the system precluding
>any other job from running.  If runnable jobs don't get a slice
>of execution time, then it's a very bad bug.
	It gets a slice, indeed.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: "Gareth Alun Evans" <gareth@cemetery.demon.co.uk>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 27 Apr 1999 18:25:07 +0100
Message-ID: <925234210.12743.0.nnrp-03.9e98250c@news.demon.co.uk>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fj0bj$48e$1@bcrkh13.ca.nortel.com> <7fr83e$oi4$1@bgtnsc02.worldnet.att.net> <3728c7aa.452058724@Rockyd>
NNTP-Posting-Host: cemetery.demon.co.uk
X-NNTP-Posting-Host: cemetery.demon.co.uk:158.152.37.12
X-Trace: news.demon.co.uk 925234210 nnrp-03:12743 NO-IDENT cemetery.demon.co.uk:158.152.37.12
X-Complaints-To: abuse@demon.net
X-Newsreader: Microsoft Outlook Express 4.72.2106.4
X-Mimeole: Produced By Microsoft MimeOLE V4.72.2106.4
Lines: 21
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.tli.de!newsfeed.nacamar.de!dispose.news.demon.net!demon!news.demon.co.uk!demon!cemetery.demon.co.uk!not-for-mail

Alexandre Pechtchanski wrote in message <3728c7aa.452058724@Rockyd>...
>On Fri, 23 Apr 1999 21:53:46 -0400, "Ambrose, Joseph"
><joseph.j.ambrose@worldnet.att.net> wrote:
>>I used Edusystem 20c in my HS days. I believe that EduSystem 50 WAS OS/8.
>>certainly looked multi-user. Had an opportunity to use it as a visitor to
>>another HS. Definitely was major cool at that time.
>
>No, EduSystem 50 was a PDP-11 version (IIRC, of course

I have here a DEC catalogue from 1971, "Digital Products And
Applications", which comments, "Edusystems
60 through 90 will be based  on the powerful and flexible PDP-11.".

All references to lower numbered Edusystems are accompanied by
terms such as "Focal", "PALIII" and "PAL-D" which would put them
firmly in the PDP-8 domain. In addition, Edusystems are mentioned
in the PDP-8 Chapter, but not in the PDP-11 Chapter.

######

From: Walter Scherer <ws@snafu.de>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 27 Apr 1999 22:40:23 +0200
Organization: Walter Scherer
Lines: 31
Message-ID: <372620B7.A36411BB@snafu.de>
References: <3723719a$0$496@news.zetnet.co.uk> <7g41g4$qnq$5@antiochus.ultra.net>
NNTP-Posting-Host: n242-116.berlin.snafu.de
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.51 [en] (X11; I; Linux 2.0.33 i586)
X-Accept-Language: en, de-DE
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!newscore.gigabell.net!newscore.ipf.de!unlisys!news.snafu.de!not-for-mail

Moin!

jmfbahciv@aol.com wrote:
> >Tell us more. This is exactly the kind of perspective we could use for
> >working out at least some of the reasons for NT being a steaming turd of
> >an operating system...
>
> I told you.  Let me parse the phrase for you...small [computer thinking].
> The following are examples of small computer thinking and not to be
> construed as the opinions of any individual, group, etc.
> Now, to talk about this particular individual....

Seems you are still afraid of *Him* ?

Are you talking about Dave Cutler?

I have heard a story about such a guy at DEC. He was the chief
designer of VMS. Someone enhanced the shell to do filename completion
and after it was working and ready to ship they showed *Him*. Bad
move! He turned it down and it never made it into the next release.
His reasoning was that doing this would cost too much performance.
Of course it would slow down the machine but improve productivity
of the humans.

Maybe I got it all wrong and it wasn't Dave Cutler or worse this 
nonsense never happened?

-- 
Tschau
------
Walter

######

From: mol@bartlet.df.lth.se (Magnus Olsson)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 27 Apr 1999 23:52:41 +0200
Organization: The Computer Society at Lund University and Lund Institute of Technology
Lines: 21
Message-ID: <7g5bj9$b87$1@bartlet.df.lth.se>
References: <371be3e4.0@newsfeed.one.net> <7g0con$68m@fohnix.metronet.com> <37241351.F947285A@tnglwood.demon.co.uk> <Pine.NXT.4.20.9904261200000.15191-100000@Tomobiki-Cho.CAC.Washington.EDU>
NNTP-Posting-Host: bartlet.df.lth.se
NNTP-Posting-User: mol
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news.lth.se!not-for-mail

In article <Pine.NXT.4.20.9904261200000.15191-100000@Tomobiki-Cho.CAC.Washington.EDU>,
Mark Crispin  <mrc@CAC.Washington.EDU> wrote:
>On Mon, 26 Apr 1999, Robert Billing wrote:
>> Stephen Jones wrote:
>> > We can look at this like UNIX and Christianity.  Both are eternally
>> > broken, cause wars, plague nations, promise you something but give you
>>  ...and attract groundless criticism.
>
>Be careful.  There are those of us who know quite a bit about UNIX and
>Christianity, and can say with some authority that the main thing wrong
>with criticism of either is that it is understated.

Hmm. I've seen quite a few "religious wars" on operating system issues,
but this is the first time one has turned into a real one.

>This is not the forum to continue this discussion.

Agreed. 
-- 
Magnus Olsson (mol@df.lth.se, zebulon@pobox.com)
------    http://www.pobox.com/~zebulon   ------

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
From: alderson@netcom19.netcom.com (Richard M. Alderson III)
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
In-Reply-To: jmfbahciv@aol.com's message of Sat, 24 Apr 99 09:06:38 GMT
Message-ID: <ALDERSON.99Apr27170136@netcom19.netcom.com>
Sender: alderson@netcom19.netcom.com
Reply-To: alderson@netcom.com
Organization: NETCOM On-line services
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu>
	<371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net>
	<Ci0U2.24$MF5.9024@srvr1.engin.umich.edu>
	<ALDERSON.99Apr23113531@netcom.netcom.com>
	<7fs7mv$hii$1@ligarius.ultra.net>
Date: Wed, 28 Apr 1999 00:01:36 GMT
Lines: 35
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!ix.netcom.com!netcom19!alderson

In article <7fs7mv$hii$1@ligarius.ultra.net> jmfbahciv@aol.com writes:

>In article <ALDERSON.99Apr23113531@netcom.netcom.com>,
>   alderson@netcom.netcom.com (Richard M. Alderson III) wrote:

>>In later versions of Tops-10, a user could interrupt a program, issue a
>>monitor command to keep it, run another program, and then resume the first
>>(up to four forks); before that change, a user could only run one program at
>>a time.

>Did you really mean TOPS-10 here, Rich?  Typos, when talking about monitors
>can get terribly misleading :-).

OK, second follow-up to same question:  Yes, I did mean Tops-10, at least as of
version 7.04 (and I think, though I'm certainly not gonna stake my life on it,
any 7.xx monitor).

What I have in mind is the PUSH, POP, and CONTEXT commands, rather than the
CSTART/CCONTINUE commands:  PUSH will allow you to exit from a running program,
get a new context, and issue any commands without thereby destroying the state
of the interrupted program; POP will take you back to the previous context; and
CONTEXT will tell you what you have running or interrupted.

CSTART and CCONTINUE would of course let you start (or continue) a program and
remain at the monitor prompt, but you could only issue commands that did not
cause a program to be run if you did not want to lose the running image.

Tops-20 was more forgiving and easier to use:  A program could be run in the
background, or put into the background after interruption, and other programs
run without affecting the process context of the first at all, without having
to issue a PUSH command (which would run another copy of the EXEC).
-- 
Rich Alderson			Last LOTS Tops-20 Systems Programmer, 1984-1991
				Current maintainer, MIT TECO EMACS (v. 170)
last name @ XKL dot COM		Chief systems administrator, XKL LLC, 1998-now

######

From: Robert Billing <unclebob@tnglwood.demon.co.uk>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 28 Apr 1999 07:44:17 +0000
Organization: Tanglewood
Message-ID: <3726BC51.E80F0C3D@tnglwood.demon.co.uk>
References: <371be3e4.0@newsfeed.one.net> <7g1b2h$ifj$1@shell3.ba.best.com> <7g1sdl$4vs@weyl.math.psu.edu> <7g47sv$rg5$2@antiochus.ultra.net> <7g4fkm$j34@bonkers.taronga.com>
NNTP-Posting-Host: localhost.demon.co.uk
X-NNTP-Posting-Host: tnglwood.demon.co.uk:158.152.132.30
X-Trace: news.demon.co.uk 925286117 nnrp-09:18579 NO-IDENT tnglwood.demon.co.uk:158.152.132.30
X-Complaints-To: abuse@demon.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.0.31 i586)
Lines: 17
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.nacamar.de!dispose.news.demon.net!demon!news.demon.co.uk!demon!tnglwood.demon.co.uk!not-for-mail

Peter da Silva wrote:

> drivers are a common elaboration, now, though of course you need to at least
> have all the drivers for boot compiled in.

 This is not strictly true of Linux, which loads, as part of the boot
process, an "initial ram disk", which is mounted as the root file
system, and then used to load the modular drivers for the boot devices.
I use this mechanism to boot this machine, which runs with a modular
driver for the SCSI disk, which is the boot device, permanently
installed.

-- 
I am Robert Billing, Christian, inventor, traveller, cook and animal
lover, I live near 0:46W 51:22N.  http://www.tnglwood.demon.co.uk/
"Bother," said Pooh, "Eeyore, ready two photon torpedoes and lock
phasers on the Heffalump, Piglet, meet me in transporter room three"

######

From: Kin Hoong CHUNG <khchung@maths.unsw.edu.au>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 28 Apr 1999 09:28:26 GMT
Organization: University of New South Wales
Lines: 16
Message-ID: <7g6kbq$d5c$3@mirv.unsw.edu.au>
References: <wk7lr7z16r.fsf@erols.com> <7fju8g$9ba$3@roch.zetnet.co.uk> <3724CC25.464B2C68@ebox.tninet.se>
NNTP-Posting-Host: alpha.maths.unsw.edu.au
User-Agent: tin/pre-1.4-980618 (UNIX) (OSF1/V4.0 (alpha))
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!howland.erols.net!news.maxwell.syr.edu!sunqbc.risq.qc.ca!news.uow.edu.au!news.usyd.edu.au!unsw.edu.au!not-for-mail

Stefan Skoglund <stetson@ebox.tninet.se> wrote:

: Hmm, i know of a sure way of detecting that.
: Install linux and  then have an egcs install fest ie
: get source, unpack, use the distribs gcc for a compile
: and then compile egcs with the just compiled version.

: Enough data so that a bad RAM should cause intermittent core-dumps.

: Emacs is also sensitive to this.

Why does this last note about emacs not suprise me :-p?

Cheers,

Kin Hoong

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 28 Apr 99 09:41:33 GMT
Organization: UltraNet Communications, Inc.
Lines: 52
Message-ID: <7g6r9i$9dc$1@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu>
	<371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net>
	<Ci0U2.24$MF5.9024@srvr1.engin.umich.edu>
	<ALDERSON.99Apr23113531@netcom.netcom.com>
	<7fs7mv$hii$1@ligarius.ultra.net> <ALDERSON.99Apr27170136@netcom19.netcom.com>
NNTP-Posting-Host: d14.dial-12.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 28 Apr 1999 11:26:42 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.ecrc.net!newsswitch.lcs.mit.edu!news.ultranet.com!d14

In article <ALDERSON.99Apr27170136@netcom19.netcom.com>,
   alderson@netcom19.netcom.com (Richard M. Alderson III) wrote:
>In article <7fs7mv$hii$1@ligarius.ultra.net> jmfbahciv@aol.com writes:
>
>>In article <ALDERSON.99Apr23113531@netcom.netcom.com>,
>>   alderson@netcom.netcom.com (Richard M. Alderson III) wrote:
>
>>>In later versions of Tops-10, a user could interrupt a program, issue a
>>>monitor command to keep it, run another program, and then 
>>>resume the first (up to four forks); before that change, 
>>>a user could only run one program at
>>>a time.
>
>>Did you really mean TOPS-10 here, Rich?  Typos, when 
>>talking about monitors can get terribly misleading :-).
>
>OK, second follow-up to same question:  Yes, I did mean 
>Tops-10, at least as of version 7.04 (and I think, though 
>I'm certainly not gonna stake my life on it, any 7.xx monitor).

Good thing, you'll live ;-).  I think those command got 
implemented with 7.03 (maybe, 7.02).
>
>What I have in mind is the PUSH, POP, and CONTEXT commands, 
>rather than the CSTART/CCONTINUE commands:  PUSH will allow 
>you to exit from a running program, get a new context, and 
>issue any commands without thereby destroying the state
>of the interrupted program; POP will take you back to the 
>previous context; and
>CONTEXT will tell you what you have running or interrupted.
>
>CSTART and CCONTINUE would of course let you start (or 
>continue) a program and remain at the monitor prompt, but 
>you could only issue commands that did not
>cause a program to be run if you did not want to lose the running image.
>
>Tops-20 was more forgiving and easier to use:  A program 
>could be run in the background, or put into the background 
>after interruption, and other programs run without affecting 
>the process context of the first at all, without having
>to issue a PUSH command (which would run another copy of the EXEC).


Oooohhhhh, _those_ commands.  I'd forgotten about them [slightly
abashed emoticon here groveling in the dust with abject apologies].
I guess I was old-fashioned enough that I never used (or trusted)
them.  Sorry, Rich.  Do you wish me lashed with noodles or 
something?

/BAH

Subtract a hundred and four for e-mail.

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 28 Apr 1999 09:44:42 -0500
Organization: none
Lines: 24
Message-ID: <7g76sq$f21@bonkers.taronga.com>
References: <371be3e4.0@newsfeed.one.net> <7g0con$68m@fohnix.metronet.com> <7g2vfl$akj$1@agate.berkeley.edu> <3728e2b4.8169296@news.ford.com>
NNTP-Posting-Host: bonkers.in.taronga.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.ecrc.net!newsfeed.berkeley.edu!newsfeed.stanford.edu!news.ems.psu.edu!bonkers!not-for-mail

In article <3728e2b4.8169296@news.ford.com>,
Clinton Pierce <cpierce1@ford.com> wrote:
>A process read-blocking on a hard-mounted NFS filesystem that's "gone
>away" will remain forever.  You can't get rid of this process.  You can
>kill it, but it won't go away completely.

This is one of many reasons that NFS should not be considered a UNIX file
system. The semantics that NFS provides to programs are too different from
local file semantics in too many ways.

I've used other network file systems that did a much better job (on OpenNET,
for example, this would eventually time out and return an EIO to the calling
process).

>A process that was waiting for a slow device (like a tape drive) and the
>device gets wedged--that process never goes away.

This is the only case I've seen... a bug in a device driver.

-- 
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
"Twenty-two points, plus triple-word-score, plus fifty points for using all
 my letters.  Game's over.  I'm outta here." -- Melissa ... by Kwyjibo.

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 28 Apr 1999 09:47:14 -0500
Organization: none
Lines: 19
Message-ID: <7g771i$f83@bonkers.taronga.com>
References: <371be3e4.0@newsfeed.one.net> <7g47sv$rg5$2@antiochus.ultra.net> <7g4fkm$j34@bonkers.taronga.com> <3726BC51.E80F0C3D@tnglwood.demon.co.uk>
NNTP-Posting-Host: bonkers.in.taronga.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!news-feed.inet.tele.dk!bofh.vszbr.cz!howland.erols.net!math.ohio-state.edu!news.cis.ohio-state.edu!news.ems.psu.edu!bonkers!not-for-mail

In article <3726BC51.E80F0C3D@tnglwood.demon.co.uk>,
Robert Billing  <unclebob@tnglwood.demon.co.uk> wrote:
>Peter da Silva wrote:
>> drivers are a common elaboration, now, though of course you need to at least
>> have all the drivers for boot compiled in.

> This is not strictly true of Linux, which loads, as part of the boot
>process, an "initial ram disk", which is mounted as the root file
>system, and then used to load the modular drivers for the boot devices.

So it basically boots off the RAM disk, and only needs that driver to boot.

How does this actually work, does LILO load this using the BIOS?

-- 
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
"Twenty-two points, plus triple-word-score, plus fifty points for using all
 my letters.  Game's over.  I'm outta here." -- Melissa ... by Kwyjibo.

######

From: Kin Hoong CHUNG <khchung@maths.unsw.edu.au>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 28 Apr 1999 09:55:27 GMT
Organization: University of New South Wales
Lines: 31
Message-ID: <7g6luf$d5c$4@mirv.unsw.edu.au>
References: <3723719a$0$496@news.zetnet.co.uk> <7g41g4$qnq$5@antiochus.ultra.net> <372620B7.A36411BB@snafu.de>
NNTP-Posting-Host: alpha.maths.unsw.edu.au
User-Agent: tin/pre-1.4-980618 (UNIX) (OSF1/V4.0 (alpha))
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!intgwpad.nntp.telstra.net!news1.optus.net.au!optus!news.usyd.edu.au!unsw.edu.au!not-for-mail

Walter Scherer <ws@snafu.de> wrote:
: Moin!

: jmfbahciv@aol.com wrote:
:>
:> I told you.  Let me parse the phrase for you...small [computer thinking].
:> The following are examples of small computer thinking and not to be
:> construed as the opinions of any individual, group, etc.
:> Now, to talk about this particular individual....

: Seems you are still afraid of *Him* ?

He is still alive, and therefore has access to lawyers... a problem in
most parts of the world.  This problem is worse in the USA, I believe.
Either that, or there is a _really_ large amount of ill-feeling...

: Are you talking about Dave Cutler?

At this point, I have a short story:  The only person I know who said that
he actually liked VMS told me something along the lines of "yes, well, you
do have to compile all your favourite commands like ls, cd etc before VMS
becomes usable."  As for me, I like the security features and robustness
of VMS, but never did figure out how to get it to do what I wanted, or where
anything I might want was, so at the first excuse, I went back to UNIX,
which I could, at least, get work done on.  Noting comments previously
posted by BAH, I would suspect that the answer is obvious.  Of course,
I could be wrong.

Cheers,

Kin Hoong

######

From: Tim Shoppa <shoppa@trailing-edge.com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 28 Apr 1999 10:49:56 -0400
Organization: Trailing Edge Technology
Lines: 21
Message-ID: <3726E7D4.34026666@trailing-edge.com>
References: <371be3e4.0@newsfeed.one.net> <7g47sv$rg5$2@antiochus.ultra.net> <7g4fkm$j34@bonkers.taronga.com> <7g748e$no4$2@ligarius.ultra.net> <FAwL3E.26G@midway.uchicago.edu>
NNTP-Posting-Host: trailing-edge.wdn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.0     DEC 3000 Model 300L)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newspeer1.nac.net!netnews.com!feeder.qis.net!yellow.newsread.com!news-toy.newsread.com!netaxs.com!newsread.com!netaxs.newsread.com!usenet

Eric Fischer wrote:
> 
> <jmfbahciv@aol.com> wrote:
> 
> > Device drivers in TOPS10 was the code that did the actual I/O to the
> > hardware.  Devices with file systems were just a tad more
> > complicated.  On TOPS10, there were actually three (I think I'll call
> > them) layers to maintain an I/O interface between the hardware and
> > the user.  John can describe this better than I.  Actually, I think
> > the only devices that had real file systems were disks.
> 
> Didn't DECtapes have file systems?  Or were those not "real" ones?

From the operating system's perspective, DECTapes are disks -
they're block-addressable devices.  Generally anywhere someone 
says "disks" they also include DECtapes.

The fact that DECtapes have access times that are potentially
ten seconds or more does have some impact on the system design.

Tim.

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 28 Apr 1999 10:53:33 -0700
Organization: Chez Inwap
Message-ID: <7g7hut$nnt$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7frcok$3ck@weyl.math.psu.edu> <7g1b2h$ifj$1@shell3.ba.best.com> <7g1sdl$4vs@weyl.math.psu.edu>
Lines: 72
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925322020 231 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!uunet!nyc.uu.net!newsfeed.mathworks.com!newshub.northeast.verio.net!netnews.com!feed1.news.rcn.net!rcn!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7g1sdl$4vs@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>	<chuckle> Could the equivalent be done by allocating 12 PTYs, putting
>TTY into the raw mode, reading from it, watching for switch characters and
>feeding the input to appropriate PTY? (with corresponding stuff for output)

The original implemenatation of PTYs had no line discipline.  It did
absolutely nothing special with control characters such as ^C, ^Q, ^S, ^U
or RUBOUT.  This was because the most common use of PTYs were BATCON
(which read lines from a file) and OPSER (in which case control characters
on the master console were acted upon before the OPSER program saw them).

Later on, there were two additions that made transparent mode possible.
1) PIM (Packed Image Mode) - put the terminal in "raw" mode where all
   8-bit characters were passed to the program verbatim.  The default was
   to pass each character to the program one at a time, but up to four
   end-on-line characters could be defined for line-buffered input.
2) Full SCNSER PTY - control characters and such were acted upon just
   like a real terminal.

>BTW, here's a serious design question - AFAICS large chunk of the
>functionality of shell was put into the terminal driver, right?

First off, there was no shell.  SCNSER was the device driver for TTYs,
PTYSER was the device driver for PTYs, and COMCON was the command controller.
It had the hardcoded list of commands that were valid in this Monitor.

Network terminals and PTYs hooked into the device-independent part of SCNSER.
SCNSER did all the line discipline, including translating control characters
(such as Control-C and Control-T) into pseudo commands that COMCON would act
on when it got a chance.  SCNSER was in the "bottom half" of the kernel; it
ran at device-driver interrupt level.  COMCON ran at an intermediate level;
commands that could be executed quickly were done during the 60 Hz clock-level
interrupt (when no other interrupts were in progress) and command that took a
long time (such as reading a core image from a disk file) were run at UUO
level.  UUO level was in the context of a job, meaning that it had a
address space, could be delayed if needed, and the Monitor could execute
UUOs on behalf of the user process.

>IOW, was there a separation between line discipline and terminal driver?

Logically, yes.  Physically, no - the two sets of routines were in the
same source file.

>How monolitic the monitor was?

The Monitor was monolithic: the entire thing was in a single file, which
meant that the Monitor could be loaded directly into physical memory from
magnetic tape.  (On a fresh install, the disks would have been low-level
formatted by the disk diagnostic program.  The Monitor would be read in
from tape, it would build a new file system, and run with or without
having SYSTEM.EXE anywhere on disk.)  [Side comment, TOPS10 did not have
disk partitions.  But it did support having logical disk structures composed
of multiple disk units, slightly similar to RAID-0.)

What you're really asking is "how formalized were the intefaces between
the various modules in the Monitor?".  "Could a new device-driver module
be added to the Monitor with a minimum of hassle?"

Device drivers had dispatch tables.  A device was selected by name using
the OPEN UUO.  When a job went to perform an operation (such as IN, OUT,
LOOKUP, ENTER, etc) on an open device, the Monitor would jump to the routine
pointed to by the appropriate entry in the dispatch table.  Operations
that were not applicable (such as file name selection on non-disk devices)
were dispatched to a dummy subroutine.

	-Joe

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 28 Apr 99 11:30:21 GMT
Organization: UltraNet Communications, Inc.
Lines: 40
Message-ID: <7g71lg$1d7$2@antiochus.ultra.net>
References: <3723719a$0$496@news.zetnet.co.uk> <7g41g4$qnq$5@antiochus.ultra.net> <372620B7.A36411BB@snafu.de>
NNTP-Posting-Host: d17.dial-16.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 28 Apr 1999 13:15:28 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.ecrc.net!newshub.northeast.verio.net!newspeer.monmouth.com!news.ultranet.com!d17

In article <372620B7.A36411BB@snafu.de>,
   Walter Scherer <ws@snafu.de> wrote:
>Moin!
>
>jmfbahciv@aol.com wrote:
>> >Tell us more. This is exactly the kind of perspective we could use for
>> >working out at least some of the reasons for NT being a steaming turd 
of
>> >an operating system...
>>
>> I told you.  Let me parse the phrase for you...small [computer 
thinking].
>> The following are examples of small computer thinking and not to be
>> construed as the opinions of any individual, group, etc.
>> Now, to talk about this particular individual....
>
>Seems you are still afraid of *Him* ?

Nope, I wasn't.  That was what my management was worried about.
>
>Are you talking about Dave Cutler?
>
>I have heard a story about such a guy at DEC. He was the chief
>designer of VMS. Someone enhanced the shell to do filename completion
>and after it was working and ready to ship they showed *Him*. Bad
>move! He turned it down and it never made it into the next release.
>His reasoning was that doing this would cost too much performance.
>Of course it would slow down the machine but improve productivity
>of the humans.
>
>Maybe I got it all wrong and it wasn't Dave Cutler or worse this 
>nonsense never happened?
>
I have no idea whether it happened or not.  My guess is that,
if it did happen, it was some other guy.

/BAH


Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 28 Apr 99 12:14:34 GMT
Organization: UltraNet Communications, Inc.
Lines: 66
Message-ID: <7g748e$no4$2@ligarius.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7g1b2h$ifj$1@shell3.ba.best.com> <7g1sdl$4vs@weyl.math.psu.edu> <7g47sv$rg5$2@antiochus.ultra.net> <7g4fkm$j34@bonkers.taronga.com>
NNTP-Posting-Host: d17.dial-16.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 28 Apr 1999 13:59:42 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!uunet!nyc.uu.net!lax.uu.net!ffx.uu.net!korova.insync.net!newsfeed.enteract.com!newsswitch.lcs.mit.edu!news.ultranet.com!d17

In article <7g4fkm$j34@bonkers.taronga.com>,
   peter@taronga.com (Peter da Silva) wrote:
>In article <7g47sv$rg5$2@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
>>On the other hand, the advantage of having the device drivers in
>>the monitor was that the code for that device driver did not
>>have to be included in the user's address space.  I have
>>no idea how UNIX deals with this.
>
>UNIX device drivers are typically relatively lightweight: 
>they provide just
>enough functionality to expose a uniform interface to the layer above (the
>filesystem layer, for disks, or the stream I/O layer for 
>devices like serial
>ports or tapes, or ptys). With System V a uniform "terminal input editing"
>layer was created so that ptys and serial ports could just expose a stream
>to that. You could also slide network stacks (PPP, for example) 
>or specialized
>I/O drivers (for real-time protocols, MIDI, and the like) in 
>at the streams layer.

Device drivers in TOPS10 was the code that did the actual 
I/O to the hardware.  Devices with file systems were just
a tad more complicated.  On TOPS10, there were actually
three (I think I'll call them) layers to maintain an I/O
interface between the hardware and the user.  John can
describe this better than I.  Actually, I think the only devices
that had real file systems were disks.  
>
>The number of these layers has grown as different kinds of devices become
>available (network interfaces or bitmal displays), and shrunk as common
>features allowed multiple kinds of devices to use a common interface 
(ptys,
>for example, or the System V streams model).
>
>Some things have never been moved below the kernel layer. 
>Cursor addressing, for example.

It shouldn't.

> Others have been pulled out and moved into user space (the
>Version 6 and version 7 printer drivers knew a lot about 
>line printers, but
>these days all parallel ports look the same and filters are 
>used to deal with eccentric hardware).

True.  But the line printer code hasn't been moved to every
user's address space.  Just the job that has the printer 
assigned.  Or is this a difference in UNIX?
>
>But in general the traditional UNIX model has had the 
>device drivers compiled
>into the kernel, and you needed a system build to 
>change them. Loadable device
>drivers are a common elaboration, now, though of course 
>you need to at least
>have all the drivers for boot compiled in.
>
Well, that was one of the next things that I wanted JMF and TW
to think about and tackle after SMP.  It would have been nice
to install a new device without having to reboot the monitor.
I was very ambitious in my younger years :-).  I wanted to
approach the networks in the same way.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 28 Apr 99 12:26:00 GMT
Organization: UltraNet Communications, Inc.
Lines: 67
Message-ID: <7g74tr$no4$3@ligarius.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7fv0l4$8v5$2@antiochus.ultra.net> <7fvhpn$241@bonkers.taronga.com> <7g0con$68m@fohnix.metronet.com> <7g2vfl$akj$1@agate.berkeley.edu> <3728e2b4.8169296@news.ford.com> <qhu2u1d8sq.fsf@ruckus.brouhaha.com>
NNTP-Posting-Host: d17.dial-16.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 28 Apr 1999 14:11:07 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!newsswitch.lcs.mit.edu!news.ultranet.com!d17

In article <qhu2u1d8sq.fsf@ruckus.brouhaha.com>,
   Eric Smith <eric-no-spam-for-me@brouhaha.com> wrote:
>cpierce1@ford.com (Clinton Pierce) writes:
>> It's quite possible in UNIX to create a process that 
>>will never completely
>> go away.
>
>As someone else pointed out, that is due to buggy device drivers, not due
>to "UNIX".
>
>> A process read-blocking on a hard-mounted NFS filesystem that's "gone
>> away" will remain forever.  You can't get rid of this process.  You can
>> kill it, but it won't go away completely.
>
>That's because NFS is seriously brain-damaged in both design and
>implementation.  Nothing about UNIX mandated that NFS be designed or
>implemented in such a way as to cause this problem.
>
>In fact, this "problem" is nothing more than the deliberately designed
>semantics of NFS hard mounts.  If this causes a problem 
>for you, you should
>ask yourself *why* you are using hard mounts.  And also, 
>perhaps, why you are
>using NFS.
>
>> A process that was waiting for a slow device (like a tape drive) and the
>> device gets wedged--that process never goes away.
>
>Only if the device driver is badly written.  On my systems, I routinely ^C
>out of lengthy I/O operations, *especially* if I know that 
>external conditions
>are such that the operation will never complete.
>
>Again, this is not a fundamental problem with UNIX.  Any monolithic
>operating system is unable to protect against broken device drivers.  A
>broken driver in TOPS-10 can just as easily cause problems.

Disasters would be a better description :-).
>
>AFAIK, though, the TOPS-10 monitor and device drivers were 
>usually pretty well
>debugged and tested by the time they shipped to customers.  
>This was a much different situation from UNIX, in that 
>TOPS-10 was only supplied by a single vendor, and only 
>had to support that vendor's hardware.  When you say "UNIX",
>you aren't referring to a specific operating system, much less 
>a specific vendor.

Yup.  That was a plus of that kind of control.  In the TOPS10
case, however, by shipping the sources allowed customers to
do their own developement but didn't cause us to be legally
responsible for their messes.  We did have to deal with those
kinds of messes but we didn't have to deal with everybody's 
mess.

The negative side to that kind of distribution is that only
DEC can use, develop, or sell the code.  It's not really
public domain.  So the art and feel of that soft/hardware
architecture is not available to just anybody who wants it.
I'm not sure we could have handled an operating system
business that allows just anybody's hardware to get plugged
in and used.  Just dealing with the SPRs and hot-line problems
is mind boggling to me.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 28 Apr 99 12:31:38 GMT
Organization: UltraNet Communications, Inc.
Lines: 38
Message-ID: <7g758d$no4$4@ligarius.ultra.net>
References: <3723719a$0$496@news.zetnet.co.uk> <7g41g4$qnq$5@antiochus.ultra.net> <372620B7.A36411BB@snafu.de> <7g6luf$d5c$4@mirv.unsw.edu.au> <37270A9D.6C847890@tnglwood.demon.co.uk>
NNTP-Posting-Host: d17.dial-16.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 28 Apr 1999 14:16:45 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!newsswitch.lcs.mit.edu!news.ultranet.com!d17

In article <37270A9D.6C847890@tnglwood.demon.co.uk>,
   Robert Billing <unclebob@tnglwood.demon.co.uk> wrote:
>Kin Hoong CHUNG wrote:
>
>> becomes usable."  As for me, I like the security features and robustness
>> of VMS, but never did figure out how to get it to do what I wanted, or 
>>where
>> anything I might want was, so at the first excuse, I went back to UNIX,
>
> Curiously, there was a time when I used VMS and Linux at the same time,
>and found most ideas freely interchangable between them. Perhaps this is
>because I grew up with DEC kit, and when I came to learn Linux (having
>briefly used solaris on the way), I simply had to look for the
>corresponding features.
>
> One thind that I still still forget some times is that "o" protection
>is "owner" on one and "other" on the other, or as the old song goes...
>
> The camel has but one hump
> The dromedary two
> Or perhaps it is the other way[1]
> I'm never sure, are you?
>
>[1] It is the other way IIRC.

That's awful!!  I'm talking about the o definition, not the
song. :-)  That's almost as bad as my "forgetting" which way
the copy goes on this machine.  Every once in a while, my
fingers remember TOPS10ese without checking in with my
forebrain for verification.

I know somebody who really, really likes VMS.  But, then,
it was his first timesharing operating system after spending
his formative years standing in IBM lines.  :-)

/BAH

Subtract a hundred and four for e-mail.

######

From: Robert Billing <unclebob@tnglwood.demon.co.uk>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 28 Apr 1999 13:18:21 +0000
Organization: Tanglewood
Message-ID: <37270A9D.6C847890@tnglwood.demon.co.uk>
References: <3723719a$0$496@news.zetnet.co.uk> <7g41g4$qnq$5@antiochus.ultra.net> <372620B7.A36411BB@snafu.de> <7g6luf$d5c$4@mirv.unsw.edu.au>
NNTP-Posting-Host: localhost.demon.co.uk
X-NNTP-Posting-Host: tnglwood.demon.co.uk:158.152.132.30
X-Trace: news.demon.co.uk 925306042 nnrp-02:25875 NO-IDENT tnglwood.demon.co.uk:158.152.132.30
X-Complaints-To: abuse@demon.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.0.31 i586)
Lines: 26
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!tnglwood.demon.co.uk!not-for-mail

Kin Hoong CHUNG wrote:

> becomes usable."  As for me, I like the security features and robustness
> of VMS, but never did figure out how to get it to do what I wanted, or where
> anything I might want was, so at the first excuse, I went back to UNIX,

 Curiously, there was a time when I used VMS and Linux at the same time,
and found most ideas freely interchangable between them. Perhaps this is
because I grew up with DEC kit, and when I came to learn Linux (having
briefly used solaris on the way), I simply had to look for the
corresponding features.

 One thind that I still still forget some times is that "o" protection
is "owner" on one and "other" on the other, or as the old song goes...

 The camel has but one hump
 The dromedary two
 Or perhaps it is the other way[1]
 I'm never sure, are you?

[1] It is the other way IIRC.
-- 
I am Robert Billing, Christian, inventor, traveller, cook and animal
lover, I live near 0:46W 51:22N.  http://www.tnglwood.demon.co.uk/
"Bother," said Pooh, "Eeyore, ready two photon torpedoes and lock
phasers on the Heffalump, Piglet, meet me in transporter room three"

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
From: eric@fudge.uchicago.edu (Eric Fischer)
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
X-Nntp-Posting-Host: fudge.uchicago.edu
Message-ID: <FAwL3E.26G@midway.uchicago.edu>
Sender: news@midway.uchicago.edu (News Administrator)
X-Newsposter: Pnews 4.0-test51 (15 Jan 97)
Organization: The University of Chicago
References: <371be3e4.0@newsfeed.one.net> <7g47sv$rg5$2@antiochus.ultra.net> <7g4fkm$j34@bonkers.taronga.com> <7g748e$no4$2@ligarius.ultra.net>
Date: Wed, 28 Apr 1999 14:18:49 GMT
Lines: 12
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!isdnet!howland.erols.net!vixen.cso.uiuc.edu!uchinews!not-for-mail

<jmfbahciv@aol.com> wrote:

> Device drivers in TOPS10 was the code that did the actual I/O to the
> hardware.  Devices with file systems were just a tad more
> complicated.  On TOPS10, there were actually three (I think I'll call
> them) layers to maintain an I/O interface between the hardware and
> the user.  John can describe this better than I.  Actually, I think
> the only devices that had real file systems were disks.  

Didn't DECtapes have file systems?  Or were those not "real" ones?

eric

######

From: alex*@*rockvax.rockefeller.edu (Alexandre Pechtchanski)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Organization: Rockefeller University Hospital (GCRC), New York
Message-ID: <37272743.607627324@Rockyd>
References: <3723719a$0$496@news.zetnet.co.uk> <7g41g4$qnq$5@antiochus.ultra.net> <372620B7.A36411BB@snafu.de> <7g6luf$d5c$4@mirv.unsw.edu.au>
X-Newsreader: Forte Agent 1.5/32.451
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Date: Wed, 28 Apr 1999 15:23:42 GMT
NNTP-Posting-Host: 129.85.24.56
X-Trace: rockyd.rockefeller.edu 925313042 129.85.24.56 (Wed, 28 Apr 1999 11:24:02 EDT)
NNTP-Posting-Date: Wed, 28 Apr 1999 11:24:02 EDT
Lines: 18
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!news.maxwell.syr.edu!newsfeed.nyu.edu!rockyd.rockefeller.edu!not-for-mail

On 28 Apr 1999 09:55:27 GMT, Kin Hoong CHUNG <khchung@maths.unsw.edu.au> wrote:

[ snip ]

>  As for me, I like the security features and robustness
>of VMS, but never did figure out how to get it to do what I wanted, or where
>anything I might want was, so at the first excuse, I went back to UNIX,
>which I could, at least, get work done on.  Noting comments previously
>posted by BAH, I would suspect that the answer is obvious.  Of course,
>I could be wrong.

Curiously, my experience was exactly opposite - probably because I grew up on
DEC systems (OS/8 -> RSX/11 -> VMS).  
I _do_ manage to have some work done on Unix (Ultrix, to be exact), but I always
have to look what the darn thing I want is called and where it is hidden ;-)

[ When replying, remove *'s from address ]
Alexandre Pechtchanski, Systems Manager, RUH, NY

######

From: Robert Billing <unclebob@tnglwood.demon.co.uk>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 28 Apr 1999 16:35:00 +0000
Organization: Tanglewood
Message-ID: <372738B4.A4E2E450@tnglwood.demon.co.uk>
References: <3723719a$0$496@news.zetnet.co.uk> <7g41g4$qnq$5@antiochus.ultra.net> <372620B7.A36411BB@snafu.de> <7g6luf$d5c$4@mirv.unsw.edu.au> <37270A9D.6C847890@tnglwood.demon.co.uk> <7g758d$no4$4@ligarius.ultra.net>
NNTP-Posting-Host: localhost.demon.co.uk
X-NNTP-Posting-Host: tnglwood.demon.co.uk:158.152.132.30
X-Trace: news.demon.co.uk 925320637 nnrp-80:2116 NO-IDENT tnglwood.demon.co.uk:158.152.132.30
X-Complaints-To: abuse@demon.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.0.31 i586)
Lines: 15
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!tnglwood.demon.co.uk!not-for-mail

jmfbahciv@aol.com wrote:

> I know somebody who really, really likes VMS.  But, then,
> it was his first timesharing operating system after spending

 Well, I quite like VMS. I suppose it was a case of losing the 16bit
restrictions implicit in RSX. Ever tried to get 4 frames of uncompressed
video in memeory at once on a machine with 4M, and then try to picture
manipulation in software...

-- 
I am Robert Billing, Christian, inventor, traveller, cook and animal
lover, I live near 0:46W 51:22N.  http://www.tnglwood.demon.co.uk/
"Bother," said Pooh, "Eeyore, ready two photon torpedoes and lock
phasers on the Heffalump, Piglet, meet me in transporter room three"

######

From: p98mccabe@alltel.net (Micheal H. McCabe)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 28 Apr 1999 17:00:01 -0400
Organization: Micheal McCabe Data Systems
Lines: 42
Message-ID: <p98mccabe-2804991700020001@r-103.54.alltel.net>
References: <371be3e4.0@newsfeed.one.net> <7g0usf$4lg@weyl.math.psu.edu> <H20V2.1086$WA4.216881@ptah.visi.com> <p98mccabe-2604991226580001@mepa2pp48.alltel.net> <7g4eu6$is7@bonkers.taronga.com>
NNTP-Posting-Host: r-103.54.alltel.net
X-Trace: news1.alltel.net 925333027 19707 166.102.103.54 (28 Apr 1999 20:57:07 GMT)
X-Complaints-To: abuse@client.alltel.net
NNTP-Posting-Date: 28 Apr 1999 20:57:07 GMT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!netnews.com!newspeer1.nac.net!newsin.agis.net!agis!news.central.agis.net!agis!news1.alltel.net!r-103.54.alltel.net!user

In article <7g4eu6$is7@bonkers.taronga.com>, peter@taronga.com (Peter da
Silva) wrote:

<SNIPPED>
> That's correct. UNIX processes are relatively lightweight, and in fact most
> traditional UNIX programs are intended to run as a part of a pipeline of
> cooperating processes, each of which does a part of the job: formatting
> a file or generating input, snipping out portions of that stream, sorting
> or reformatting it for display or further processing.
...
> Large interactive programs that serve a lot of different purposes were
> generally ports from other operating systems (emacs, for example) or written
> to emulate programs used in other environments (I believe that PINE falls
> into this category). As these have become more popular this non-UNIXy mode
> of use has spawned a lot of problems: buggy system calls (the looping
> process problem) and security problems (caused usually by having a large
> complex program that also requires special privileges to run).
...
> Hell, there's even things about Windows I like. But I better stop here before
> I'm lynched by BOTH sides.


The strange thing is, I find myself constantly writing short BASIC (for
lack of a better CHEAP programming language) that tend to operate this
way.... 

I have a file in format X, the BASIC program reformats it into format Y,
another program sorts it, finally another short program generates a
report.  User I/O tends to be conversational and directed at the glass
teletype.

Generally, I'm doing this under MS-DOS/Windows or within the Mac OS (Thank
god for Chipmunk BASIC!), where STDIN and STDOUT have largely been
forgotten.  The Mac in particular has few provisions for these "batch"
operations.

I guess Windows (3.x/95/NT/98) has its strong points (cheap, available,
familiar), but the instability of the system gets pretty annoying.

As always, thanks for the thoughtful reply.

p98mccabe@alltel.net

######

From: michael.wojcik@merant.com (Michael Wojcik)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 28 Apr 1999 19:07:37 GMT
Organization: MERANT Inc.
Lines: 46
Message-ID: <7g7m9p$222h@news2.newsguy.com>
References: <3723719a$0$496@news.zetnet.co.uk> <7g41g4$qnq$5@antiochus.ultra.net> <372620B7.A36411BB@snafu.de> <7g6luf$d5c$4@mirv.unsw.edu.au> <37272743.607627324@Rockyd>
NNTP-Posting-Host: p-019.newsdawg.com
X-Newsreader: xrn 9.00
Originator: mww@raederle.microfocus.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.enteract.com!netnews.com!newspeer1.nac.net!newsfeed.concentric.net!pln-w!spln!extra.newsguy.com!newsp.newsguy.com!michael.wojcik


In article <37272743.607627324@Rockyd>, alex*@*rockvax.rockefeller.edu (Alexandre Pechtchanski) writes:
> On 28 Apr 1999 09:55:27 GMT, Kin Hoong CHUNG <khchung@maths.unsw.edu.au> wrote:
> >  As for me, I like the security features and robustness
> >of VMS, but never did figure out how to get it to do what I wanted, or where
> >anything I might want was, so at the first excuse, I went back to UNIX,
> >which I could, at least, get work done on.
> 
> Curiously, my experience was exactly opposite - probably because I grew up on
> DEC systems (OS/8 -> RSX/11 -> VMS).  

I had to use VMS in school, for assembly and Fortran programming
labs, and I liked many aspects of it.  VAX VMS is a great platform
for CISC assembly-language programming.

> I _do_ manage to have some work done on Unix (Ultrix, to be exact),
> but I always have to look what the darn thing I want is called and
> where it is hidden ;-)

Indeed.  Unix (in assorted flavors) is my personal OS of choice, but
that's because I use it every day and have large portions memorized.
I certainly wouldn't recommend it for everyone, or for every application.
(For many business functions, OS/400 - the police state of operating
systems - is much more suitable, for example.)

ObSomewhatOnTopic:  NT, on the other hand, doesn't become much easier
to use even after spending a lot of time with it.  It's too GUI - you
plateau early on.  Oh, there are still "keyboard shortcuts" and macro
recorders and the like, but they're an afterthought and rarely done
well.  I was recently forced to upgrade from the crummy Exchange
email client to the heinous Outlook 98 one, and had to relearn most
of the shortcuts, as the Microsoft nitwits had changed most of them.

Windows 2.0 and Windows/286 were awkward but usable without a mouse
(at least as much so as they were with one).  Windows 3 was only
marginally workable without a mouse.  With Win32, forget it.  You
can expect to waste most of your time pushing that stupid cursor
around the screen.


Michael Wojcik          	        michael.wojcik@merant.com
AAI Development, MERANT                 (block capitals are a company mandate)
Department of English, Miami University

Please enjoy the stereo action fully that will surprise you.
-- Pizzicato Five

######

From: Ariel Scolnicov <ariels@compugen.co.il>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 28 Apr 1999 20:35:10 +0300
Organization: NetVision Israel	
Lines: 52
Message-ID: <yzqzp3shcmp.fsf@selena.compugen.co.il>
References: <371be3e4.0@newsfeed.one.net> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu> <7g1b2h$ifj$1@shell3.ba.best.com> <7g1sdl$4vs@weyl.math.psu.edu> <7g47sv$rg5$2@antiochus.ultra.net>
NNTP-Posting-Host: selena.compugen.co.il
X-Trace: news.netvision.net.il 925320828 14236 194.90.227.168 (28 Apr 1999 17:33:48 GMT)
X-Complaints-To: abuse@netvision.net.il
NNTP-Posting-Date: 28 Apr 1999 17:33:48 GMT
X-Newsreader: Gnus v5.5/XEmacs 20.4 - "Emerald"
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.tli.de!newsfeed.amsterdam.nl.net!sun4nl!uunet!ams.uu.net!ffx.uu.net!in3.uu.net!news-feed.netvision.net.il!194.90.1.15.MISMATCH!news!not-for-mail

jmfbahciv@aol.com writes:

> In article <7g1sdl$4vs@weyl.math.psu.edu>,
>    viro@weyl.math.psu.edu (Alexander Viro) wrote:
> >[...]
> >	<chuckle> Could the equivalent be done by allocating 
> >12 PTYs, 
> 
> Yes.
> 
> >putting TTY into the raw mode, 
> 
> I don't know what you mean by raw mode.

"Raw mode" is when the terminal driver does the absolute minimum -- no 
CR/LF processing, no backspace/DEL processing, interrupts, or
anything.  Also no buffering, so byte-at-a-time I/O to the terminal
will work.

There's also "cbreak", which is similar, but does process the
interrupt, quit, suspend, and flow control characters (list stolen
from curses documentation).

Commonly used for programs that want keystroke control (like vi, pine, 
etc.).

> [...]
> 
> >BTW, here's a serious design question - AFAICS large chunk of the
> >functionality of shell was put into the terminal driver, right? 
> 
> If I understand your question, the answer is no.  Most of the
> "shell" part was in the stand alone program we called OPSER.
> The address space of the user didn't mix with the address space
> of the monitor.
> 
> >I.e. together with the equivalent of cooked mode. 
> 
> I don't know what "cooked mode" means.

"Cooked" is, well, the opposite of "raw".  It's what you normally use
in programs -- read full lines of text, let the terminal driver handle 
what little "editing" it does (delete last character/word/entire
line), etc.

> [...]

-- 
Ariel Scolnicov        |"GCAAGAATTGAACTGTAG"            |ariels@compugen.co.il
Compugen Ltd.          |Tel: +972-2-6795059 (Jerusalem)	\  THIS SPACE TO LET
72 Pinhas Rosen St.    |Tel: +972-3-7658520 (Main office)`--------------------
Tel-Aviv 69512, ISRAEL |Fax: +972-3-7658555  http://www.compugen.co.il/~ariels

######

From: Ariel Scolnicov <ariels@compugen.co.il>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Followup-To: alt.folklore.computers
Date: 28 Apr 1999 20:37:28 +0300
Organization: NetVision Israel	
Lines: 27
Message-ID: <yzqyajchciu.fsf@selena.compugen.co.il>
References: <371be3e4.0@newsfeed.one.net> <H20V2.1086$WA4.216881@ptah.visi.com> <7g23p5$559@weyl.math.psu.edu> <7g485m$rg5$3@antiochus.ultra.net> <7g5cc4$6lf@weyl.math.psu.edu>
NNTP-Posting-Host: selena.compugen.co.il
X-Trace: news.netvision.net.il 925320973 14236 194.90.227.168 (28 Apr 1999 17:36:13 GMT)
X-Complaints-To: abuse@netvision.net.il
NNTP-Posting-Date: 28 Apr 1999 17:36:13 GMT
X-Newsreader: Gnus v5.5/XEmacs 20.4 - "Emerald"
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.tli.de!newsfeed.amsterdam.nl.net!sun4nl!uunet!ams.uu.net!ffx.uu.net!in3.uu.net!news-feed.netvision.net.il!194.90.1.15.MISMATCH!news!not-for-mail

viro@weyl.math.psu.edu (Alexander Viro) writes:

> In article <7g485m$rg5$3@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
> >In article <7g23p5$559@weyl.math.psu.edu>,
> >   viro@weyl.math.psu.edu (Alexander Viro) wrote:
> >>	Been there, seen that. Program that ignores the return value of
> >>function it called *is* badly written. AFAIK no OS is able to stop program
> >>from loop: goto loop;. Fresh example: nethack. Function reads a character,
> >>checks if it belongs to <set> and either returns the index 
> >>in set or rereads.  Ignores EOF.
> >
> >Why should an OS stop a user mode program from looping?  If that's
> 	Exactly.
> >what the user wants, the OS should give it him.  It's just another
> >job, albeit probably doing useless work.  It is the job of the
> 	See above - 99:1 that it's a bug in the program and OS has no business
> stopping it.
> >OS to ensure that a looping job doesn't hog the system precluding
> >any other job from running.  If runnable jobs don't get a slice
> >of execution time, then it's a very bad bug.
> 	It gets a slice, indeed.

Of course, an infinite loop is only until the next reboot, which can
come surprisingly quickly under *SOME* operating systems...

-- 
Ariel Scolnicov

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 28 Apr 1999 20:39:20 -0400
Organization: -ENOENT
Lines: 13
Message-ID: <7g89no$dv5@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7g4fkm$j34@bonkers.taronga.com> <3726BC51.E80F0C3D@tnglwood.demon.co.uk> <7g771i$f83@bonkers.taronga.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news.cis.ohio-state.edu!news.ems.psu.edu!news3.cac.psu.edu!not-for-mail

In article <7g771i$f83@bonkers.taronga.com>,
Peter da Silva <peter@taronga.com> wrote:
>In article <3726BC51.E80F0C3D@tnglwood.demon.co.uk>,
>So it basically boots off the RAM disk, and only needs that driver to boot.
>
>How does this actually work, does LILO load this using the BIOS?

Yup. It can be booted in a normal way, indeed, but for precanned kernel in
the distribution it's easier to go that way.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: norris@mech.eng.usyd.edu.au
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 29 Apr 1999 00:51:49 GMT
Organization: The University of Sydney, Australia
Lines: 13
Message-ID: <7g8af5$a6d$2@metro.ucc.usyd.edu.au>
References: <371be3e4.0@newsfeed.one.net> <7fv0l4$8v5$2@antiochus.ultra.net> <7fvhpn$241@bonkers.taronga.com> <7g0con$68m@fohnix.metronet.com> <7g2vfl$akj$1@agate.berkeley.edu> <3728e2b4.8169296@news.ford.com>
NNTP-Posting-Host: ebb.mech.eng.usyd.edu.au
X-Trace: metro.ucc.usyd.edu.au 925347109 10445 129.78.14.203 (29 Apr 1999 00:51:49 GMT)
X-Complaints-To: usenet@news.usyd.edu.au
NNTP-Posting-Date: 29 Apr 1999 00:51:49 GMT
User-Agent: tin/pre-1.4-981114 ("The Watchman") (UNIX) (Linux/2.0.30 (i586))
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!sunqbc.risq.qc.ca!news.uow.edu.au!news.usyd.edu.au!not-for-mail

In alt.folklore.computers Clinton Pierce <cpierce1@ford.com> wrote:

: A process read-blocking on a hard-mounted NFS filesystem that's "gone
: away" will remain forever.  You can't get rid of this process.  You can
: kill it, but it won't go away completely.

Yeah. Thats why you use "hard intr" as options when mounting nfs
partitions.

-- 
Stuart Norris                                   norris@mech.eng.usyd.edu.au
Mechanical Engineering,University of Sydney,NSW 2006   wk:+(61 2) 9351-2272
http://www.maths.unsw.edu.au/~norris                   hm:+(61 2) 9326-5276

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7fv0l4$8v5$2@antiochus.ultra.net>
     <7fvhpn$241@bonkers.taronga.com> <7g0con$68m@fohnix.metronet.com>
     <7g2vfl$akj$1@agate.berkeley.edu> <3728e2b4.8169296@news.ford.com>
     <qhu2u1d8sq.fsf@ruckus.brouhaha.com> <7g74tr$no4$3@ligarius.ultra.net>
X-Newsreader: NN version 6.5.0 CURRENT #119
From: werme@werme.ne.mediaone.net (Ric Werme)
Lines: 49
Message-ID: <GmOV2.24090$tY1.15521@wbnws01.ne.mediaone.net>
Date: Thu, 29 Apr 1999 01:13:10 GMT
NNTP-Posting-Host: 24.128.109.10
X-Trace: wbnws01.ne.mediaone.net 925348390 24.128.109.10 (Wed, 28 Apr 1999 21:13:10 EDT)
NNTP-Posting-Date: Wed, 28 Apr 1999 21:13:10 EDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newspump.monmouth.com!newspeer.monmouth.com!netnews.com!chnws02.mediaone.net!24.128.1.101!chnws05.ne.mediaone.net!24.128.44.7!wbnws01.ne.mediaone.net.POSTED!not-for-mail

jmfbahciv@aol.com writes:

>In article <qhu2u1d8sq.fsf@ruckus.brouhaha.com>,

>>In fact, this "problem" is nothing more than the deliberately designed
>>semantics of NFS hard mounts.  If this causes a problem 
>>for you, you should
>>ask yourself *why* you are using hard mounts.  And also, 
>>perhaps, why you are
>>using NFS.

I'm _the_ NFS person at Compaq, but I'm doing a pretty good job biting
my tounge on all this.  After all, this is not comp.protocols.nfs and
I'm not fond of USENET subject drift, unless it pertains to PDP-10s.
Yes, NFS has warts, but so does SMB, AFS, and DFS.  Still, more than 50% of
the network traffic at work is NFS.

One of the reasons hard mounts are so hard goes back to diskless Sun days
when servers actually crashed once in a while.  Timing out page outs or ins
tended to have kill the client too.  Still, we could do better, but most
of my work is on the server side and I others spend more time on the client.

The major alternative to hard mounts is soft mounts.  My predecessor,
Chet Juszczak, refused to talk to _anyone_ with NFS complaints who
used soft mounts.  Our flaky network then pretty much guaranteed
timeouts.  Chet was about as irascible (sp?) as Tony Wachs.

The compromise is hard, interruptible mounts, but there are couple
warts there, our data cache is really serious about getting pages out,
and that needs to be tamed a bit.  I'm not sure if other vendors have
similar problems.  Still, its a good middle ground.

Note - anyone squawking about NFS will probably get direct replies
instead of followups.  NFS _folklore_ I might post to.

>Yup.  That was a plus of that kind of control.  In the TOPS10
>case, however, by shipping the sources allowed customers to
>do their own development but didn't cause us to be legally
>responsible for their messes.

Another big win thanks to shipping sources is that DEC hired a lot
of OS engineers who didn't need training - they came well skilled
in TOPS-10 design.  I was about the only monitor person from C-MU,
though we had several C-MU alums elsewhere, Worcester Poly, Univ of
Arizona, etc. were good sources.
--
Ric Werme                  | http://people.ne.mediaone.net/werme
werme@nospam.mediaone.net  | http://www.cyberportal.net/werme
      ^^^^^^^ delete

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 29 Apr 1999 01:57:05 -0400
Organization: -ENOENT
Lines: 96
Message-ID: <7g8sbh$e88@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7g1b2h$ifj$1@shell3.ba.best.com> <7g1sdl$4vs@weyl.math.psu.edu> <7g7hut$nnt$1@shell3.ba.best.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!nyd.news.ans.net!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7g7hut$nnt$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>In article <7g1sdl$4vs@weyl.math.psu.edu>,
>Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>	<chuckle> Could the equivalent be done by allocating 12 PTYs, putting
>>TTY into the raw mode, reading from it, watching for switch characters and
>>feeding the input to appropriate PTY? (with corresponding stuff for output)
>
>The original implemenatation of PTYs had no line discipline.  It did
>absolutely nothing special with control characters such as ^C, ^Q, ^S, ^U
>or RUBOUT.  This was because the most common use of PTYs were BATCON
>(which read lines from a file) and OPSER (in which case control characters
>on the master console were acted upon before the OPSER program saw them).

	Umm... That is, in original implementation PTY == something similar
to bidirectional pipe (socketpair in AF_UNIX, actually), right? AFAICS
model implied one terminal - one process relation (BTW, were there unattached
processes?). Interesting...
	I can see where it could come from, but it should have a
nasty trade-off - UNIX model (and as far as I heard TOPS-20 one) allows
signal delivery to group. I.e. if I run a pipeline on a tty3 all processes
in this pipeline belong to a group controlling tty3 and pressing ^Z will
cause tty3 to deliver SIGSTOP to all of them. This usage of terminal ID
is impossible with terminal ID == process ID model. Was there an equivalent
mechanism? (No, I'm aware that pipelines are UNIXism and all such. Having a
group of cooperating processes that can be stopped/resumed/killed together
isn't).

>Later on, there were two additions that made transparent mode possible.
>1) PIM (Packed Image Mode) - put the terminal in "raw" mode where all
>   8-bit characters were passed to the program verbatim.  The default was
>   to pass each character to the program one at a time, but up to four
>   end-on-line characters could be defined for line-buffered input.
>2) Full SCNSER PTY - control characters and such were acted upon just
>   like a real terminal.
>
>>BTW, here's a serious design question - AFAICS large chunk of the
>>functionality of shell was put into the terminal driver, right?
>
>First off, there was no shell.  SCNSER was the device driver for TTYs,
>PTYSER was the device driver for PTYs, and COMCON was the command controller.
>It had the hardcoded list of commands that were valid in this Monitor.
>
>Network terminals and PTYs hooked into the device-independent part of SCNSER.
>SCNSER did all the line discipline, including translating control characters
>(such as Control-C and Control-T) into pseudo commands that COMCON would act
>on when it got a chance.  SCNSER was in the "bottom half" of the kernel; it
>ran at device-driver interrupt level. COMCON ran at an intermediate level;
>commands that could be executed quickly were done during the 60 Hz clock-level
>interrupt (when no other interrupts were in progress) and command that took a
>long time (such as reading a core image from a disk file) were run at UUO
>level. 
	Aha... I'm not completely sure how it translates to UNIX terms. Let's
see... There are hardware interrupt handlers (normally most urgent stuff +
planning a software interrupt for less time-critical parts) and there are
software interrupts (normally planned by something else (including prior
invocations of themselves) and run with hardware interrupts enabled at
reasonable frequency). Both types are in bottom half (async to process, have
no context, can preempt upper half, can't call scheduler). Do you mean that
SCNSER was in the equivalent of hardware interrupt? Or just that it sat in
the software interrupt and didn't use continuation mechanisms?
	
>UUO level was in the context of a job, meaning that it had a
>address space, could be delayed if needed, and the Monitor could execute
>UUOs on behalf of the user process. 
	Aha, so UUO level == normal upper halh. Wait... Do you mean that
in some cases COMCON ran in the context of some job? Could you elaborate?

>>IOW, was there a separation between line discipline and terminal driver?
>
>Logically, yes.  Physically, no - the two sets of routines were in the
>same source file.
[snip]

>What you're really asking is "how formalized were the intefaces between
>the various modules in the Monitor?".  "Could a new device-driver module
>be added to the Monitor with a minimum of hassle?"

>Device drivers had dispatch tables.  A device was selected by name using
>the OPEN UUO.  When a job went to perform an operation (such as IN, OUT,
>LOOKUP, ENTER, etc) on an open device, the Monitor would jump to the routine
>pointed to by the appropriate entry in the dispatch table.  Operations
	I.e. the same setup as in UNIX, modulo different set of methods.
<chuckle> Always fine to watch C++ fanatics claiming that OOP is impossible
in C. Wonder how do they react on OOP in assembler...
>that were not applicable (such as file name selection on non-disk devices)
				   ^^^^^^^^^^^^^^^^^^^
					What???
>were dispatched to a dummy subroutine.
	No questions wrt to N/A methods, but... do you imply that filesystem
was folded into the disk driver? <shudder><shudder><cold sweat> My deepest
condolence to those who had to maintain that. Seriously. BTW, what kind of fs
did you have and how did it evolve?

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 29 Apr 1999 08:05:32 -0500
Organization: none
Lines: 19
Message-ID: <7g9les$7j5@bonkers.taronga.com>
References: <371be3e4.0@newsfeed.one.net> <qhu2u1d8sq.fsf@ruckus.brouhaha.com> <7g74tr$no4$3@ligarius.ultra.net> <GmOV2.24090$tY1.15521@wbnws01.ne.mediaone.net>
NNTP-Posting-Host: bonkers.in.taronga.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!newsfeed.stanford.edu!news.ems.psu.edu!bonkers!not-for-mail

In article <GmOV2.24090$tY1.15521@wbnws01.ne.mediaone.net>,
Ric Werme <werme@werme.ne.mediaone.net> wrote:
>Yes, NFS has warts, but so does SMB, AFS, and DFS.  Still, more than 50% of
>the network traffic at work is NFS.

Intel was shipping a SMB-based remote file system, not exactly compatible
with Lan Manager and running over ISO CONS/TP4, that provided pretty damn
close to full UNIX semantics over a network connection.

The Sprite network file system also had some nice features.

Unfortunately, you can't compete against NFS or SMB in the open market, because
they're ubiquitous and you have to remain compatible with one or the other.

-- 
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
"Twenty-two points, plus triple-word-score, plus fifty points for using all
 my letters.  Game's over.  I'm outta here." -- Melissa ... by Kwyjibo.

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 29 Apr 1999 08:08:54 -0500
Organization: none
Lines: 28
Message-ID: <7g9ll6$7k8@bonkers.taronga.com>
References: <371be3e4.0@newsfeed.one.net> <7g47sv$rg5$2@antiochus.ultra.net> <7g4fkm$j34@bonkers.taronga.com> <7g748e$no4$2@ligarius.ultra.net>
NNTP-Posting-Host: bonkers.in.taronga.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!news-feed.inet.tele.dk!bofh.vszbr.cz!newsfeed.stanford.edu!news.ems.psu.edu!bonkers!not-for-mail

In article <7g748e$no4$2@ligarius.ultra.net>,  <jmfbahciv@aol.com> wrote:
>>Some things have never been moved below the kernel layer. 
>>Cursor addressing, for example.

>It shouldn't.

I was under the impression that the device drivers in TOPS-20 knew a lot
about the terminal, so (for example) kill-line actually moved the cursor
to the end of the prompt and then issued an erase-line code.

(that really impressed me, for some reason)

>>Version 6 and version 7 printer drivers knew a lot about 
>>line printers, but
>>these days all parallel ports look the same and filters are 
>>used to deal with eccentric hardware).

>True.  But the line printer code hasn't been moved to every
>user's address space.  Just the job that has the printer 
>assigned.  Or is this a difference in UNIX?

No.

-- 
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
"Twenty-two points, plus triple-word-score, plus fifty points for using all
 my letters.  Game's over.  I'm outta here." -- Melissa ... by Kwyjibo.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Thu, 29 Apr 99 10:04:52 GMT
Organization: UltraNet Communications, Inc.
Lines: 22
Message-ID: <7g9h1i$2jc$1@antiochus.ultra.net>
References: <3723719a$0$496@news.zetnet.co.uk> <7g41g4$qnq$5@antiochus.ultra.net> <372620B7.A36411BB@snafu.de> <7g6luf$d5c$4@mirv.unsw.edu.au> <37270A9D.6C847890@tnglwood.demon.co.uk> <7g758d$no4$4@ligarius.ultra.net> <372738B4.A4E2E450@tnglwood.demon.co.uk>
NNTP-Posting-Host: d1.dial-12.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 29 Apr 1999 11:50:10 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d1

In article <372738B4.A4E2E450@tnglwood.demon.co.uk>,
   Robert Billing <unclebob@tnglwood.demon.co.uk> wrote:
>jmfbahciv@aol.com wrote:
>
>> I know somebody who really, really likes VMS.  But, then,
>> it was his first timesharing operating system after spending
>
> Well, I quite like VMS. I suppose it was a case of losing the 16bit
>restrictions implicit in RSX. Ever tried to get 4 frames of uncompressed
>video in memeory at once on a machine with 4M, and then try to picture
>manipulation in software...
>
Nope.  That's not where my talents lay.  From the sounds of it,
you did.  

I've been told that VMS is good because a user doesn't have
to be an expert user in order to get his/er work done.  One
just has to remember to type ? [or whatever it is on VMS].

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Thu, 29 Apr 99 10:39:22 GMT
Organization: UltraNet Communications, Inc.
Lines: 31
Message-ID: <7g9j27$88h$1@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7g47sv$rg5$2@antiochus.ultra.net> <7g4fkm$j34@bonkers.taronga.com> <7g748e$no4$2@ligarius.ultra.net> <FAwL3E.26G@midway.uchicago.edu>
NNTP-Posting-Host: d2.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 29 Apr 1999 12:24:39 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!howland.erols.net!newsswitch.lcs.mit.edu!news.ultranet.com!d2

In article <FAwL3E.26G@midway.uchicago.edu>,
   eric@fudge.uchicago.edu (Eric Fischer) wrote:
><jmfbahciv@aol.com> wrote:
>
>> Device drivers in TOPS10 was the code that did the actual I/O to the
>> hardware.  Devices with file systems were just a tad more
>> complicated.  On TOPS10, there were actually three (I think I'll call
>> them) layers to maintain an I/O interface between the hardware and
>> the user.  John can describe this better than I.  Actually, I think
>> the only devices that had real file systems were disks.  
>
>Didn't DECtapes have file systems?  Or were those not "real" ones?

Now I want to mount a DECtape and take a look at the directory.
IIRC, there weren't really PPNs on a DECtape, although one
could tell whether it was an -11 DECtape or a -10 DECtape
by looking at the PPN displayed by DIRECT.  One could have 
a limited number of files (with an overall limit of blocks).
The way the -11s handled placing data on DECtape was different
from the way TOPS10 handled placing data on DECtape.  I remember
that read an -11 DECtape on an -11 was ssssoooooo sssslllloooowwwww
because the tape had to be read all the way to the EOT and back
again.  Again, I don't know the implementation details.  Maybe
somebody else can talk about that.  One could random access
a DECtape on the -10 but couldn't on a magtape.  I remember
even swapping on a DECtape but with the advent of installing
Level D software that stopped being a necessity.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Thu, 29 Apr 99 10:48:21 GMT
Organization: UltraNet Communications, Inc.
Lines: 92
Message-ID: <7g9jj5$88h$2@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7g0usf$4lg@weyl.math.psu.edu> <H20V2.1086$WA4.216881@ptah.visi.com> <p98mccabe-2604991226580001@mepa2pp48.alltel.net> <7g26qv$55r@weyl.math.psu.edu>
NNTP-Posting-Host: d2.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 29 Apr 1999 12:33:41 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.ecrc.net!newsswitch.lcs.mit.edu!news.ultranet.com!d2

In article <7g26qv$55r@weyl.math.psu.edu>,
   viro@weyl.math.psu.edu (Alexander Viro) wrote:
>In article <p98mccabe-2604991226580001@mepa2pp48.alltel.net>,
>Micheal H. McCabe <p98mccabe@alltel.net> wrote:
>>
>>OK folks, here's where my ignorance really starts to show...
>>
>>To begin with, I know nothing about UNIX internals.  My experiences with
>>time-sharing are limited to TS/OS on the Univac 90/60 (An IBM 360 clone),
>>VAX/VMS (Somewhere around version 2.? - Later upgraded to 4.?), and RT-11
>>on the pdp-11.
>>
>>ISTR, when you submitted a DO file under VMS or RT-11 (hazy recall of an
>>@filename command), it appeared as if commands were entered from your
>>terminal and executed directly.  Output was displayed (or printed) on 
your
>>terminal unless redirected to a file.  During execution of this command
>>file, the only actions possible from the terminal keyboard were CTRL-S to
>>stop scrolling, CTRL-Q to resume, CTRL-C to interrupt execution of the
>>current command, or CTRL-Z to abort the whole thing.
>>
>>Likewise, when you used SUBMIT to enter a batch job, an independant
>>process was created that would execute until it completed its task, ran
>>out of CPU time, or was killed by the user (or system operator.)  Once 
the
>>job was submitted, your terminal was free to do additional work.
>>
>>Did I miss some fine distinction here?
>>
>>How did you do this sort of thing under TOPS-10 or TOPS-20?  Also, my
>	<AOL></AOL>
>>understanding of UNIX was that each invocation of a command generally
>>created an independant process.  Is this an incorrect assumption?
>	In case of UNIX you have this functionality divided between several
>layers.
>	First of all, creation of new process is *fast*. There are two
>primitives - one of them (fork()) creates a copy of the virtual machine 
(==
>process) and lets it run in parallel with everything else. It's lazy, i.e.
>pages are marked as copy-on-write rather than being copied. If a process
>tries to modify such page it gets a copy (transparently) and c-o-w bit on
>that copy is reset. fork() doesn't load a new program - it's done by
>exec() (as far as I could understand TOPS-10 chain execution is done via
>something pretty similar). exec() drops all memory of the virtual machine
>that had called it and maps it on the executable you've asked for.
>	Now, since the creation of processes is cheap it's offloaded from
>the kernel. Any program can fork a copy, exec whatever you want in said
>copy and either wait for the child's completion or simply go on with its
>business. It doesn't have to bother with dumping its state, loading the
>child instead of itself and restoring the state later. I.e. invocation
>of child process is easy. Shell is a regular program that reads from input
>(either terminal or file with a script) and invokes appropriate commands.
>How much will be implemented by internal functions and how much will be
>in utilities depends on the authors' tastes. Many other programs do the
>same - e.g. mailer can invoke one of the standard editors instead of
>implementing an internal one.
>	Processing of ^U, ^C and friends is a different story - you can
>either put the terminal into raw mode and make all processing yourself or
>you can use a line discipline to process this stuff in the kernel. In the
>latter case it's transparent for you. You can make a LDISC driver to 
interpret
>a non-trivial language if you want (and are willing to write it). One can
>implement a shell-alike as a LDISC and get a situation similar to (as far
>as I could understand) one in TOPS-10. Generally nobody does it (e.g. 
since
>debugging the user program is much safer and easier than debugging the
>kernel code ;-/), but it is doable - I did something similar when I hacked
>on EXT2 and needed a twisted equivalent of kdb. It didn't launch new
>processes, but upon Esc-!-A it gave a prompt and allowed to watch/modify
>VFS data, set/reset semaphores and call inode methods. Ugly, but it saved 
my
>ass quite a few times. Attempt to do it from userspace was, to put it
>mildly, unwise - half of operations would be impossible and it would have
>very good chances to cause additional harm. So it can be done, but unless 
you
>are doing pretty hairy kernel hacking it's simply not needed.
>

When one wanted to debug the TOPS10 monitor, one took the whole
system stand-alone.  The monitor could not be debugged as a user
mode program.  A very special DDT, called EDDT, was loaded with
the monitor for symbolic debugging purposes.  On our systems,
one could go to the system console, type <ctrl-D>, and stop
the system by hitting that special break point.  Since our
shops were run based on something we called "gentleman's 
timesharing" this was not done unless there existed a great
need.  There a few times that the command was typed by
mistake but an <altmode>P would continue system operation.

/BAH

Subtract a hundred and four for e-mail.

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 29 Apr 1999 19:32:06 -0700
Organization: Chez Inwap
Message-ID: <7gb4n6$795$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7g1sdl$4vs@weyl.math.psu.edu> <7g7hut$nnt$1@shell3.ba.best.com> <7g8sbh$e88@weyl.math.psu.edu>
Lines: 257
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925439530 220 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.ecrc.net!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7g8sbh$e88@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>	Umm... That is, in original implementation PTY == something similar
>to bidirectional pipe (socketpair in AF_UNIX, actually), right?

Not really.  The PTY setup is extremely asymmetric.  On a Monitor configured
for 64 terminals and 32 PTYs, the names TTY0 through TTY77 corresponded to
real terminals, and TTY100 through TTY137 corresponded to the slave half of
PTY0 through PTY37.

After a program did an OPEN on a PTY, writing to the PTY via an OUT UUO
caused characters to appear in the input buffer on the slave terminal.
Doing an IN UUO performed a non-blocking read of the slave terminal's output
buffer.  There was a special UUO to read the condition of the PTY.
Conditions included "output available for reading", "slave job is waiting for
input (user mode)" and "slave job is waiting for input (Monitor mode)".

>signal delivery to group. Was there an equivalent mechanism?

There are no groups.  TOPS-10 did not have pipelines.  There was no concept of
STDIN or STDOUT.  Programs had two ways of doing I/O to the terminal: it could
do an OPEN on device 'TTY' and use normal IN, INPUT, OUT, or OUTPUT UUOs, or
it do TTCALLs.  The most common TTCALLs were: output one 7-bit character,
output a string of characters (first character had to start on a word
boundary, the string was terminated by a null character), input one character
(cbreak), and input one character after waiting for a full line (to allow the
user a chance to use backspace, delete, Control-U, etc).  If you wanted to
turn off echo, or to be able to read all possible characters (including
Control-C and/or parity), you had to use the OPEN call.  But that used up a
software I/O channel, and there were only 16 of those available.

>implied one terminal - one process relation (BTW, were there unattached
>processes?)

Each device on the system had a DDB (Device Data Block).  In the DDB,
there was a 9-bit number indicating which job owned the device.  There
was a separate bit for TTY DDBs to indicate that the job was owning the device
was using it as the controlling terminal.  Each job had an entry in a table
pointing to its TTY.

The implications were this:
  *) A job could have control over more than one TTY.  For an example,
  a job running on a VT52 that emulated a lineprinter by sending characters
  to an LA36 that no-one was logged in on at the time.

  *) If Control-C was seen arriving on a TTY, and that was controlling
  a job, then said job would get an interrupt condition.  (Control-C
  on TOPS-10 was like Control-Z on Unix; the stopped process could be
  continued by entering the CONTINUE or CCONTINUE commands.)

  *) The DETACH command and DETACH system call would disassociate a job
  from its terminal.  A properly written program could continue running
  even when the job no longer had a controlling terminal.

>	Aha... I'm not completely sure how it translates to UNIX terms. Let's
>see... There are hardware interrupt handlers (normally most urgent stuff +
>planning a software interrupt for less time-critical parts) and there are
>software interrupts (normally planned by something else (including prior
>invocations of themselves) and run with hardware interrupts enabled at
>reasonable frequency). Both types are in bottom half (async to process, have
>no context, can preempt upper half, can't call scheduler). Do you mean that
>SCNSER was in the equivalent of hardware interrupt? Or just that it sat in
>the software interrupt and didn't use continuation mechanisms?

SCNSER was the name of the module (assembly language source file) that had the
TTY interrupt routines and the TTY UUO routines.  COMMON was the name of
the module that set up the interrupt vectors, determined which physical
device caused the interrupt, then passed control to the device-specific
interrupt handler via the device's dispatch table.

To output "Hello world!", the job would execute a UUO which did a context
switch into the UUOCON module.  UUOCON would locate the terminal device's
DDB, then use the DDB's dispatch table to jump to the routine that handled
the OUT UUO.  This routine lived in SCNSER's "top half".  It would disable
TTY interrupts, copy the characters from user address space to kernel
address space, then update the count of characters in the device's output
buffer.  Then it would jump to a routine in SCNSER's "bottom half" to
start the device driver (if it was not already running) and re-enable
TTY interrupts.

Also involved were LDBs (Line Data Blocks).  They were adjuncts to the
TTY DDBs and held the TTY input and output buffer chunks.




>>UUO level was in the context of a job, meaning that it had a
>>address space, could be delayed if needed, and the Monitor could execute
>>UUOs on behalf of the user process. 
>	Aha, so UUO level == normal upper half. Wait... Do you mean that
>in some cases COMCON ran in the context of some job? Could you elaborate?

COMCON had to deal with many types of commands.
  *) Quick, no output (such as SET TERMINAL WIDTH 80).

  *) Quick, with output from the Monitor.  Such as PJOB (print job number),
     DAYTIME (print time of day), Control-T (print name of program currently
     in core, whether it is running or blocked on I/O).  If the terminal's
     output buffer was already full, the characters from the Monitor were
     thrown into the bit bucket, because COMCON was not in a situation
     where it could afford to wait for the output buffer to empty.

  *) Slow - anything a file into the user's address space or writing
     a core image to a file.  (Note: the terms "user address space" and
     "core image" are synonymous in this context.)

Take, for example, the case where someone types SYSTAT and hits the RETURN
key.

  *) First character arrives from device driver.  It is stored in the
     input buffer for that terminal line.  The input buffer count is
     incremented and the LDB is marked as having at least one character.
     Unless SET TERMINAL NO ECHO is in effect, the character is copied to
     the output buffer and echoed back to the user.

  *) Additional characters arrive.  The user can edit the input by using
     RUBOUT (DEL), Control-H (backspace) or Control-U.

  *) The user hits the RETURN key.  The device driver pretends that the
     user has entered both carriage-return and line-feed.  Both characters
     go into the input buffer, and both get echoed to the output buffer.

  *) Line-feed is one of several characters that TOPS-10 takes as an
     end-of-line condition.  The LDB is marked as having a line of input
     available.  Since the terminal is in Monitor mode (instead of user
     mode), SCNSER marks a bit saying that this line needs attention from
     COMCON.

  Everything described above was done at Priority Interrupt level 3, the
  one usually assigned to SCNSER and other device drivers.

  *) An interrupt comes in a PI level 1 from the line-frequency clock.
     This happens 60 times per second in North America, 50 Hz elsewhere.
     It increments the time-of-day counters, and causes a software
     initiated interrupt at PI level 7.

  *) If no other interrupts are in progress, the interrupt handler for PI
     level 7 runs.  This is clock-level, where scheduling and CPU usage
     accounting is done.  COMCON runs at clock-level, and is expected to
     complete all its tasks before the next clock tick that will arrive
     in 16.67 or 20 milliseconds.

  *) COMCON locates the terminal line that has a line of input waiting.
     It looks for a command, skipping leading blanks and stopping at
     a semicolon, exclaimation mark, or line-feed.  The first word of
     the line is converted to uppercase, truncated to 6 characters max,
     then searched for in the table of legal Monitor commands.

  *) The command dispatch table entry for SYSTAT says that this command
     destroys the current core image (the current job's physical and
     virtual memory resources have to be released), is legal even when
     no one is logged in on that terminal, requires a job number (the next
     available job number is assigned if no one is logged in on that
     terminal), will put the terminal in user mode, requires the running
     of a program, and the name of the program to run is SYS:SYSTAT.EXE.

  *) The process of reading SYS:SYSTAT.EXE from disk may take around one
     hundred milliseconds.  (Back in the days when TOPS-10 ran from
     DECtapes, reading a file from tape could take 2 to 50 seconds).

COMCON can't afford to wait around while that happens; it has other lines
that need to be monitored for commands.  Besides, reading a program into
user's address space requires that physical and virtual core be assigned
and a page map created.  The job is marked as needing attention, and COMCON
continues with its clock-level tasks.

  *) The scheduler also runs at clock-level.  It notices that there is
     a job that needs to run.  The job that was running before clock-level
     interrupt came in is examined.  If it had used up its time slice,
     the scheduler finishes the context switch by storing the job's
     registers and PC in the job's process table.  The scheduler changes
     the user page map for the new job, loads its registers and PC.
     The flag bits in the PC indicate that the job will be run in kernel
     mode and the PC points to the routine in COMCON that will resume
     loading SYS:SYSTAT.EXE into memory.

  *) The last instruction in the clock-level routine is one that restores
     the jobs PC and PC flags, and dismisses the current (level 7) interrupt.

Now COMCON is running on the behalf of a particular job, at UUO level.
It pretends that the job had executed a RUN UUO to get here.  TOPS-10
allows a UUO handling routine to execute another UUO.  The RUN UUO
does an OPEN on logical device SYS (which typically translated to
DSKB:[1,4]), a LOOKUP on SYSTAT.EXE, and then as many IN UUOs as needed
to read the program into core.  (Later versions of TOPS-10 allowed on-demand
paging, where pages of the EXE file would be read in as needed by the page
fault handler, but that's a different topic.)

>>Device drivers had dispatch tables.  A device was selected by name using
>>the OPEN UUO.  When a job went to perform an operation (such as IN, OUT,
>>LOOKUP, ENTER, etc) on an open device, the Monitor would jump to the routine
>>pointed to by the appropriate entry in the dispatch table.  Operations
>	I.e. the same setup as in UNIX, modulo different set of methods.
><chuckle> Always fine to watch C++ fanatics claiming that OOP is impossible
>in C. Wonder how do they react on OOP in assembler...

Say "amen!" brother.  New grad students are re-inventing stuff 30 years old.

>>that were not applicable (such as file name selection on non-disk devices)
>				   ^^^^^^^^^^^^^^^^^^^
>					What???
>>were dispatched to a dummy subroutine.

To create a file on disk, the program would OPEN device DSK, do an ENTER UUO
to specify a 6-character file name and a 3-character extension.  The fourth
word of the ENTER UUO would be zero to specify the current directory,
nonzero in both halves to specify a particular user's top-level directory,
or zero in the left half and nonzero in the right to point to a path block
(directory and up to five levels of subdirectory).  Then the file could
be written to with OUT UUOs.

To create a file on DECtape, OPEN and ENTER were mandatory before OUT.

To send output to a lineprinter, paper tape punch, card punch, plotter,
magnetic tape, or terminal, only the OPEN was required.  The ENTER UUO was a
no-op.  (Although with printer spooling, the name from the ENTER would be used
in the print queue.  And an ENTER was significant on ANSI-labled tapes.)

Well written programs would do an ENTER even if they did an OPEN to device
TTY.  This allowed for
	.ASSIGN DSK TTY
	.RUN PROGRA.EXE
	.DEASSIGN TTY
to redirect the output of the program to a file on the disk.	

>	No questions wrt to N/A methods, but... do you imply that filesystem
>was folded into the disk driver?

Not at all.  The entry points to the filesystem were stored in a "struct"
that was accessable to the file system, the disk driver, and the UUO
handling routines.
all devices that were file oriented.  The filesystem itself was stored

When an ENTER UUO was done, UUOCON would get the arguments to the ENTER and
copy them to the device DDB.  Then it would use the device's dispatch table to
locate the device-specific routine to handle an ENTER and call it.

For devices that were not file-oriented, the routine was nothing but a simple
return-from-subroutine.  For devices that were read-only, the routine would
cause the UUO to return with an "invalid operation" error code.

For DECtapes, the routine for ENTER was in DTASER, the same module that
had the DECtape device driver.  (The file system semantics for DECtapes
was similar to MSDOS version 1; a limited number of directory entries,
6.3 naming convention, no subdirectories.)

For disk devices, the routine for ENTER was in FILUUO.  That was the
device-independent module for doing LOOKUP, ENTER, RENAME and other file
related operations.  That module implemented the disk directory structure,
mandatory file locks and the ability to read the old contents of a file that
was being superseded.
	-Joe

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

Newsgroups: alt.folklore.computers
From: nickerson@pundit.ds.boeing.com ()
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
X-Nntp-Posting-Host: pundit.ds.boeing.com
Message-ID: <FAyuBw.MBI@news.boeing.com>
Lines: 31
Sender: nickerson@mirage.boeing.com ()
Reply-To: nickerson@pundit.ds.boeing.com ()
Organization: Boeing Defense & Space Group / Software Systems
X-Newsreader: mxrn 6.18-32
References: <3723719a$0$496@news.zetnet.co.uk> <7g41g4$qnq$5@antiochus.ultra.net> <372620B7.A36411BB@snafu.de> <7g6luf$d5c$4@mirv.unsw.edu.au> <37270A9D.6C847890@tnglwood.demon.co.uk> <7g758d$no4$4@ligarius.ultra.net> <372738B4.A4E2E450@tnglwood.demon.co.uk> <7g9h1i$2jc$1@antiochus.ultra.net>
Date: Thu, 29 Apr 1999 19:33:32 GMT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!howland.erols.net!newsfeed.cwix.com!192.220.250.21!netnews1.nw.verio.net!netnews.nwnet.net!news.verio.net!xyzzy!mirage.boeing.com!nickerson


In article <7g9h1i$2jc$1@antiochus.ultra.net>, jmfbahciv@aol.com writes:

|>I've been told that VMS is good because a user doesn't have
|>to be an expert user in order to get his/er work done.  One
|>just has to remember to type ? [or whatever it is on VMS].
|>
|>/BAH


$ help			! parenthetically (man man) and thinking 1960s 


I believe your observation on expertise is correct; when MacOS or Windows
are mentioned the converse is hard to explain to people; ie) those are
user friendly but obtuse and hard to manage; this discussion has again 
gotten me thinking about why I like VMS and dislike UNIX; 

maybe one item would be UNIX shell/termcap and VMS DCL/drivers for the 
VT100 et. al.; another post mentioned that when running a command file 
($ @command.com) use of certain control keys is possible; actually the
use of control keys is a bit richer than mentioned and has evolved over
time; control keys are used for DCL commands, command files, cursor 
position or state, command line recall, command line editing, and screen 
display;  the VMS users manual has serveral pages of explanation; thus
there is a large post that could expound on this topic, but suffice it 
to say, "gotta love that VMS";

--bn (Bart Nickerson)
nickerson@pundit.ds.boeing.com
(206) 662-0183

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 29 Apr 1999 19:39:08 -0700
Organization: Chez Inwap
Message-ID: <7gb54c$9i3$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7g4fkm$j34@bonkers.taronga.com> <7g748e$no4$2@ligarius.ultra.net> <7g9ll6$7k8@bonkers.taronga.com>
Lines: 22
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925439954 210 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!newsfeed.berkeley.edu!news2.best.com!news1.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7g9ll6$7k8@bonkers.taronga.com>,
Peter da Silva <peter@taronga.com> wrote:
>I was under the impression that the device drivers in TOPS-20 knew a lot
>about the terminal, so (for example) kill-line actually moved the cursor
>to the end of the prompt and then issued an erase-line code.
>
>(that really impressed me, for some reason)

That was in the TTY driver, not the device driver.

Device drivers include the console terminal, KLINIK, DZ11, DL11, ANF10,
DC68, network virtual terminals and other things that ran at device
interrupt level.

The TTY driver was a software abstraction on top of the device drivers.  Both
TOPS-10 and TOPS-20 had the VT100 ESCape-leftsquarebracket-K hardcoded into
the system.
	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 30 Apr 1999 06:19:07 -0400
Organization: -ENOENT
Lines: 60
Message-ID: <7gc02r$go9@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7g7hut$nnt$1@shell3.ba.best.com> <7g8sbh$e88@weyl.math.psu.edu> <7gb4n6$795$1@shell3.ba.best.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!news-feed.inet.tele.dk!bofh.vszbr.cz!howland.erols.net!newsxfer.itd.umich.edu.MISMATCH!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7gb4n6$795$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>Take, for example, the case where someone types SYSTAT and hits the RETURN
>key.
[reading the line]
>     available.  Since the terminal is in Monitor mode (instead of user
>     mode), SCNSER marks a bit saying that this line needs attention from
>     COMCON.

>  Everything described above was done at Priority Interrupt level 3, the
>  one usually assigned to SCNSER and other device drivers.

[upon the timer interrupt timer bh is scheduled and COMCON is run]
>  *) COMCON locates the terminal line that has a line of input waiting.
[parses]

Umhm... And since device can't be used by several processes COMCON can't
be implemented as a process and has to sit in kernel.

>  *) The command dispatch table entry for SYSTAT says that this command
>     destroys the current core image (the current job's physical and
>     virtual memory resources have to be released), is legal even when
>     no one is logged in on that terminal, requires a job number (the next
>     available job number is assigned if no one is logged in on that
>     terminal), will put the terminal in user mode, requires the running
>     of a program, and the name of the program to run is SYS:SYSTAT.EXE.
[snip]
>The job is marked as needing attention, and COMCON
>continues with its clock-level tasks.

>  *) The scheduler also runs at clock-level.
[Joe Average schedule()]
[new process was set up sitting in the kernel mode and going to do]
[equivalent of exec() on SYSTAT.EXE. The rest is obvious.]

Aha. It looks like a hybrid between getty and (minimal) shell, put into the
kernel...

>To create a file on disk, the program would OPEN device DSK, do an ENTER UUO
>to specify a 6-character file name and a 3-character extension.  The fourth
>word of the ENTER UUO would be zero to specify the current directory,
>nonzero in both halves to specify a particular user's top-level directory,
>or zero in the left half and nonzero in the right to point to a path block
>(directory and up to five levels of subdirectory).  Then the file could
>be written to with OUT UUOs.
[snip]

>For disk devices, the routine for ENTER was in FILUUO.  That was the
>device-independent module for doing LOOKUP, ENTER, RENAME and other file
>related operations.  That module implemented the disk directory structure,
>mandatory file locks and the ability to read the old contents of a file that
>was being superseded.

So you did sequence of lookups, followed by the operation. All from the
userspace. Either you had damn good namespace locking available to processes
or there should be a helluva lot of races... BTW, what happened if one
process tried to remove or rename a file opened by another one?

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 30 Apr 1999 09:09:24 -0500
Organization: none
Lines: 15
Message-ID: <7gcdik$38c@bonkers.taronga.com>
References: <371be3e4.0@newsfeed.one.net> <7g0con$68m@fohnix.metronet.com> <37241351.F947285A@tnglwood.demon.co.uk> <Pine.NXT.4.20.9904261200000.15191-100000@Tomobiki-Cho.CAC.Washington.EDU>
NNTP-Posting-Host: bonkers.in.taronga.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!math.ohio-state.edu!news.cis.ohio-state.edu!news.ems.psu.edu!bonkers!not-for-mail

In article <Pine.NXT.4.20.9904261200000.15191-100000@Tomobiki-Cho.CAC.Washington.EDU>,
Mark Crispin  <mrc@CAC.Washington.EDU> wrote:
>Be careful.  There are those of us who know quite a bit about UNIX and
>Christianity, and can say with some authority that the main thing wrong
>with criticism of either is that it is understated.
>
>This is not the forum to continue this discussion.

May I suggest alt.religion.computers ?

-- 
This is The Reverend Peter da Silva's Boring Sig File - there are no references
to Wolves, Kibo, Discordianism, or The Church of the Subgenius in this document
"Twenty-two points, plus triple-word-score, plus fifty points for using all
 my letters.  Game's over.  I'm outta here." -- Melissa ... by Kwyjibo.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 30 Apr 99 09:30:49 GMT
Organization: UltraNet Communications, Inc.
Lines: 47
Message-ID: <7gc3e0$ek8$1@antiochus.ultra.net>
References: <3723719a$0$496@news.zetnet.co.uk> <7g41g4$qnq$5@antiochus.ultra.net> <372620B7.A36411BB@snafu.de> <7g6luf$d5c$4@mirv.unsw.edu.au> <37270A9D.6C847890@tnglwood.demon.co.uk> <7g758d$no4$4@ligarius.ultra.net> <372738B4.A4E2E450@tnglwood.demon.co.uk> <7g9h1i$2jc$1@antiochus.ultra.net> <FAyuBw.MBI@news.boeing.com>
NNTP-Posting-Host: d6.dial-12.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 30 Apr 1999 11:16:16 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!sunqbc.risq.qc.ca!newsswitch.lcs.mit.edu!news.ultranet.com!d6

In article <FAyuBw.MBI@news.boeing.com>,
   nickerson@pundit.ds.boeing.com () wrote:
>
>In article <7g9h1i$2jc$1@antiochus.ultra.net>, jmfbahciv@aol.com writes:
>
>|>I've been told that VMS is good because a user doesn't have
>|>to be an expert user in order to get his/er work done.  One
>|>just has to remember to type ? [or whatever it is on VMS].
>|>
>|>/BAH
>
>
>$ help			! parenthetically (man man) and thinking 1960s

I think it's the in-line help character prompt that gets used the
most instead of the command.  But I've never "watched" the VMS
usage so it's pure speculation on my part.  One can learn an
awful lot just by watching a user work on a system...especially
where improvements are required.

 
>
>
>I believe your observation on expertise is correct; when MacOS or Windows
>are mentioned the converse is hard to explain to people; ie) those are
>user friendly but obtuse and hard to manage; this discussion has again 
>gotten me thinking about why I like VMS and dislike UNIX; 
>
>maybe one item would be UNIX shell/termcap and VMS DCL/drivers for the 
>VT100 et. al.; another post mentioned that when running a command file 
>($ @command.com) use of certain control keys is possible; actually the
>use of control keys is a bit richer than mentioned and has evolved over
>time; control keys are used for DCL commands, command files, cursor 
>position or state, command line recall, command line editing, and screen 
>display;  the VMS users manual has serveral pages of explanation; thus
>there is a large post that could expound on this topic, but suffice it 
>to say, "gotta love that VMS";

Well, one of the reasons I preferred TOPS10 to TOPS20 is that the -10
didn't require verbosity.  VMS does require it and UNIX doesn't.
(I'm talking about the garden variety without preferential 
enhancements here.  I know that user interfaces can be developed
to match the tastes of the user.)

/BAH

Subtract a hundred and four for e-mail.

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 30 Apr 1999 10:17:04 -0700
Organization: Chez Inwap
Message-ID: <7gcoig$63f$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7g8sbh$e88@weyl.math.psu.edu> <7gb4n6$795$1@shell3.ba.best.com> <7gc02r$go9@weyl.math.psu.edu>
Lines: 70
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925492631 223 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!isdnet!logbridge.uoregon.edu!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gc02r$go9@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>In article <7gb4n6$795$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>>For disk devices, the routine for ENTER was in FILUUO.  That was the
>>device-independent module for doing LOOKUP, ENTER, RENAME and other file
>>related operations.  That module implemented the disk directory structure,
>>mandatory file locks and the ability to read the old contents of a file that
>>was being superseded.
>
>So you did sequence of lookups, followed by the operation. All from the
>userspace. Either you had damn good namespace locking available to processes
>or there should be a helluva lot of races... BTW, what happened if one
>process tried to remove or rename a file opened by another one?

The process of loading a program into memory was in userspace.  File
manipulation was done by the kernel.  No problems, no races.

Here is one command on TOPS-10 that cannot done the same way on Unix:
	COPY RESUME.TXT=HEADER.TXT,RESUME.TXT,FOOTER.TXT

*) Open channel 1 to disk.

*) Do an ENTER on channel 1 to RESUME.TXT.  If the file is already open for
   output by another job, return error code 3 (File Being Modified). Since the
   file name already exists, this will be a superceding enter.  Copy the file
   permissions for the existing file into the DDB.  (The old file remains on
   disk, and can still be read.)

*) Open channel 2 to disk.

*) LOOKUP HEADER.TXT on channel 2. Do a series of IN UUOs on channel 2 and OUT
   UUOs on channel 1.  CLOSE open file on channel 2 at EOF.

*) LOOKUP RESUME.TXT on channel 2.  This locates the original file; the new
   version being written channel 1 is known only to the Monitor and is not
   pointed to by any directory.  Do a series of IN UUOs on channel 2 and OUT
   UUOs on channel 1.  CLOSE open file on channel 2 at EOF.

*) LOOKUP FOOTER.TXT on channel 2. Do a series of IN UUOs on channel 2 and OUT
   UUOs on channel 1.  CLOSE open file on channel 2 at EOF.

*) RELEASE channel 2.  Free up the DDB.

*) CLOSE channel 1.  Update the RIB (Retrieval Information Block, much
   like an inode).  Update the directory entry to point to the new RIB.

*) If no job had the old RESUME.TXT open, free up the disk blocks belonging
   to the old file, including the old RIB.

*) If RESUME.TXT is open for reading on another channel or by another job,
   the old RIB is still accessable via the open DDBs.  There can be any
   number of superseded versions of the file still open for reading.

*) RELEASE channel 1 and terminate the COPY program (SYS:PIP.EXE).

TOPS-20 has explicit version numbers.  It would create RESUME.TXT.2 while
reading RESUME.TXT.1; both file names would exist in the directory when
the output file was closed.

This is in contrast to Unix and other operating systems which truncate
the file to zero bytes when it is openned for output.

The down side was that, in general, it was not possible to watch the progress
of a slowly growing file.  Programs that wanted to make their log files
visible in real time would need to open+append+close for each block written.
	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7g47sv$rg5$2@antiochus.ultra.net> <7g4fkm$j34@bonkers.taronga.com> <3726BC51.E80F0C3D@tnglwood.demon.co.uk>
Organization: D Bit, Troy, NY
From: wilson@dbit.com (John Wilson)
NNTP-Posting-Host: dbit.dbit.com
Message-ID: <3729c2b2.0@news.wizvax.net>
Date: 30 Apr 1999 10:48:18 -0500
X-Trace: 30 Apr 1999 10:48:18 -0500, dbit.dbit.com
Lines: 14
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!howland.erols.net!netnews.com!newspeer.monmouth.com!newsfeed.wizvax.net!news.wizvax.net!dbit.com!wilson

In article <3726BC51.E80F0C3D@tnglwood.demon.co.uk>,
Robert Billing  <unclebob@tnglwood.demon.co.uk> wrote:
>> drivers are a common elaboration, now, though of course you need to at least
>> have all the drivers for boot compiled in.
>
> This is not strictly true of Linux, which loads, as part of the boot
>process, an "initial ram disk", which is mounted as the root file
>system, and then used to load the modular drivers for the boot devices.

Of course, in this case the kernel has to have the RAM disk driver and
its file system compiled in.  Both can be pretty primitive, but still...

John Wilson
D Bit

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7g748e$no4$2@ligarius.ultra.net> <7g9ll6$7k8@bonkers.taronga.com> <7gb54c$9i3$1@shell3.ba.best.com>
Organization: D Bit, Troy, NY
From: wilson@dbit.com (John Wilson)
NNTP-Posting-Host: dbit.dbit.com
Message-ID: <3729c731.0@news.wizvax.net>
Date: 30 Apr 1999 11:07:29 -0500
X-Trace: 30 Apr 1999 11:07:29 -0500, dbit.dbit.com
Lines: 10
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.wirehub.nl!newspeer1.nac.net!newspeer.monmouth.com!newsfeed.wizvax.net!news.wizvax.net!dbit.com!wilson

In article <7gb54c$9i3$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>The TTY driver was a software abstraction on top of the device drivers.  Both
>TOPS-10 and TOPS-20 had the VT100 ESCape-leftsquarebracket-K hardcoded into
>the system.

So was VTSOP% just an OZism, or was it a regular part of TOPS-20?
It looked a bit like it wanted to be the ITS TTY driver.

John Wilson
D Bit

######

From: Mark Crispin <mrc@CAC.Washington.EDU>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 30 Apr 1999 12:50:33 -0700
Organization: Networks & Distributed Computing
Lines: 74
Message-ID: <Pine.NXT.4.20.9904301203070.19802-100000@Tomobiki-Cho.CAC.Washington.EDU>
References: <371be3e4.0@newsfeed.one.net> <7g0con$68m@fohnix.metronet.com> <37241351.F947285A@tnglwood.demon.co.uk> <Pine.NXT.4.20.9904261200000.15191-100000@Tomobiki-Cho.CAC.Washington.EDU> <m7lW2.2057$WA4.398156@ptah.visi.com>
NNTP-Posting-Host: tomobiki-cho.cac.washington.edu
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Trace: nntp1.u.washington.edu 925501837 23750 (None) 140.142.17.35
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: fwa
To: Peter Seebach <seebs@plethora.net>
In-Reply-To: <m7lW2.2057$WA4.398156@ptah.visi.com>
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.eurocyber.net!newsfeed.mathworks.com!nntp-out.monmouth.com!newspeer.monmouth.com!netnews1.nw.verio.net!netnews.nwnet.net!news.verio.net!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc

On Fri, 30 Apr 1999, Peter Seebach wrote:
> So far, I've mostly found that "problems with Unix" are really not problems
> with Unix - not always, but very often.

UNIX file locking is broken, when it works at all:
 1) doesn't work with NFS on some systems; on others, it sort of works
    until lockd/statd wedge and you have to reboot the entire cluster
    after zapping all the locks.
 2) no upgrade from shared to exclusive, or downgrade from exclusive to
    shared.  What?  You say it has it?  Read the man page carefully.  Very
    carefully.
 3) only one lock resource per file.  You can have multiple section locks,
    but only one resource.

UNIX file semantics are untrustworthy:
 1) stat() and file data can get out of synchronization with each other in
    the buffer cache, necessitating you to cat through a few megabytes to
    clear the buffer cache and get current data.
 2) O_EXCL is not guaranteed atomic.
 3) inconsistant implementations of disk quota vis a vis write().  Some
    systems will fail the write(); others will return partial success.
    Some systems won't let you overwrite existing file data if you're
    over quota (which screws over trying to fix it).
 4) Don't get me started about the difference between write()+fsync()
    and write()+write()+fsync().  Or about the timing race that plops
    NULs on top of perfectly good file data because you had the temercity
    to append to the file.
 5) Don't get me started about the bugs in O_RDWR mode either.

UNIX is much less device independent than any PDP-10 operating system was.
Not only does it have a huge pile of device-dependent operations, but
they're split into zillions of odd system calls.  The very worst of the
worst is that abomination called sockets.

The list goes on and on.  Time and time again, the answer is "It's a
well-known bug/misfeature in UNIX.  Program around it."

> I recall PINE used to have quite a bit of trouble porting - because someone
> somewhere got the idea that, depsite several standards, you should *never*
> include <unistd.h> if you could get away with it, so several ports ended up
> not working - because they needed (as is correct) <unistd.h>.

Please get your facts straight.

Pine was written on a UNIX system that did not have <unistd.h>.  The very
oldest c-client code was written on a Mac.  The other UNIX ports that we
did at the time were to UNIX systems that did not have <unistd.h>.

Almost all of the other ports were contributed code from people who had
those other systems.  We did not, for example, do the original SVR4 port;
nor did we even touch an SVR4 system for several years.  I won't name
names, but the people who did the original SVR4 work simply made the code
compile on SVR4, without checking for SVR4 vs. BSD semantic differences.

Things were made worse by the fact that as UNIX vendors got religion, some
of them added <unistd.h> or some other new "standard" and promptly made
everything dependent upon it. The problem with this is that lots of people
still had the old release. "Ve ist ze UNIX geeks, ve zu not need ze
upwards compatibility!"  Often, this was a system type that we didn't
have.  So, we get sent a patch to fix this, then a patch to fix the patch,
then...  SGI and Data General were the worst offenders, but Linux has done
its part.

When get find out (more likely figure out) what's going on, we clean it up
and split up ports as necessary.  It takes time to do that when the
information is second or third hand.

-- Mark --

* RCW 19.190 notice: This email address is located in Washington State.	*
* Unsolicited commercial email may be billed $500 per message.		*
Science does not emerge from voting, party politics, or public debate.

######

From: Mark Crispin <mrc@CAC.Washington.EDU>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 30 Apr 1999 12:50:33 -0700
Organization: Networks & Distributed Computing
Lines: 74
Message-ID: <Pine.NXT.4.20.9904301203070.19802-100000@Tomobiki-Cho.CAC.Washington.EDU>
References: <371be3e4.0@newsfeed.one.net> <7g0con$68m@fohnix.metronet.com> <37241351.F947285A@tnglwood.demon.co.uk> <Pine.NXT.4.20.9904261200000.15191-100000@Tomobiki-Cho.CAC.Washington.EDU> <m7lW2.2057$WA4.398156@ptah.visi.com>
NNTP-Posting-Host: tomobiki-cho.cac.washington.edu
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Trace: nntp1.u.washington.edu 925501837 23750 (None) 140.142.17.35
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: fwa
To: Peter Seebach <seebs@plethora.net>
In-Reply-To: <m7lW2.2057$WA4.398156@ptah.visi.com>
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.eurocyber.net!newsfeed.mathworks.com!nntp-out.monmouth.com!newspeer.monmouth.com!netnews1.nw.verio.net!netnews.nwnet.net!news.verio.net!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc

On Fri, 30 Apr 1999, Peter Seebach wrote:
> So far, I've mostly found that "problems with Unix" are really not problems
> with Unix - not always, but very often.

UNIX file locking is broken, when it works at all:
 1) doesn't work with NFS on some systems; on others, it sort of works
    until lockd/statd wedge and you have to reboot the entire cluster
    after zapping all the locks.
 2) no upgrade from shared to exclusive, or downgrade from exclusive to
    shared.  What?  You say it has it?  Read the man page carefully.  Very
    carefully.
 3) only one lock resource per file.  You can have multiple section locks,
    but only one resource.

UNIX file semantics are untrustworthy:
 1) stat() and file data can get out of synchronization with each other in
    the buffer cache, necessitating you to cat through a few megabytes to
    clear the buffer cache and get current data.
 2) O_EXCL is not guaranteed atomic.
 3) inconsistant implementations of disk quota vis a vis write().  Some
    systems will fail the write(); others will return partial success.
    Some systems won't let you overwrite existing file data if you're
    over quota (which screws over trying to fix it).
 4) Don't get me started about the difference between write()+fsync()
    and write()+write()+fsync().  Or about the timing race that plops
    NULs on top of perfectly good file data because you had the temercity
    to append to the file.
 5) Don't get me started about the bugs in O_RDWR mode either.

UNIX is much less device independent than any PDP-10 operating system was.
Not only does it have a huge pile of device-dependent operations, but
they're split into zillions of odd system calls.  The very worst of the
worst is that abomination called sockets.

The list goes on and on.  Time and time again, the answer is "It's a
well-known bug/misfeature in UNIX.  Program around it."

> I recall PINE used to have quite a bit of trouble porting - because someone
> somewhere got the idea that, depsite several standards, you should *never*
> include <unistd.h> if you could get away with it, so several ports ended up
> not working - because they needed (as is correct) <unistd.h>.

Please get your facts straight.

Pine was written on a UNIX system that did not have <unistd.h>.  The very
oldest c-client code was written on a Mac.  The other UNIX ports that we
did at the time were to UNIX systems that did not have <unistd.h>.

Almost all of the other ports were contributed code from people who had
those other systems.  We did not, for example, do the original SVR4 port;
nor did we even touch an SVR4 system for several years.  I won't name
names, but the people who did the original SVR4 work simply made the code
compile on SVR4, without checking for SVR4 vs. BSD semantic differences.

Things were made worse by the fact that as UNIX vendors got religion, some
of them added <unistd.h> or some other new "standard" and promptly made
everything dependent upon it. The problem with this is that lots of people
still had the old release. "Ve ist ze UNIX geeks, ve zu not need ze
upwards compatibility!"  Often, this was a system type that we didn't
have.  So, we get sent a patch to fix this, then a patch to fix the patch,
then...  SGI and Data General were the worst offenders, but Linux has done
its part.

When get find out (more likely figure out) what's going on, we clean it up
and split up ports as necessary.  It takes time to do that when the
information is second or third hand.

-- Mark --

* RCW 19.190 notice: This email address is located in Washington State.	*
* Unsolicited commercial email may be billed $500 per message.		*
Science does not emerge from voting, party politics, or public debate.

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7gc02r$go9@weyl.math.psu.edu> <7gcoig$63f$1@shell3.ba.best.com> <7gcr9l$hn5@weyl.math.psu.edu>
Organization: D Bit, Troy, NY
From: wilson@dbit.com (John Wilson)
NNTP-Posting-Host: dbit.dbit.com
Message-ID: <3729fa95.0@news.wizvax.net>
Date: 30 Apr 1999 14:46:45 -0500
X-Trace: 30 Apr 1999 14:46:45 -0500, dbit.dbit.com
Lines: 41
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feed2.nntp.acc.ca!feed.nntp.acc.ca!newspump.monmouth.com!newspeer.monmouth.com!newsfeed.wizvax.net!news.wizvax.net!dbit.com!wilson

In article <7gcr9l$hn5@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>Here is one command on TOPS-10 that cannot done the same way on Unix:
>>	COPY RESUME.TXT=HEADER.TXT,RESUME.TXT,FOOTER.TXT
>
>[snip]
>
>	Can be done almost the same way. rename() over the open file
>unlinks it, but leaves the thing intact. And it's atomic (if you
>don't care for atomicity - simpler yet, open();unlink();open(); will
>do the thing).

Maybe you can do flips and twists to get the same effect on UNIX if you
*know* that the input and output files have the same name, but the point is,
on TOPS-10 (and many other DEC OSes) it will Just Work, even with simple-minded
programs that plow ahead and open the files the most obvious way.  The
concept of "file marked for deletion" etc. is definitely handled properly
by the monitor, so it doesn't matter if the directory appears to be changing
before a program's eyes, the file processor still knows what is what.

I feel sure TOPS-10 had this feature too:  it's really nice to be able to
purge an I/O channel, rather than officially closing it, and have all trace
of the output file (including the implicit deletion of any other file of
the same name) disappear.  So if things go seriously wrong (my favorite
mistake is accidentally interchanging the names of your source and object
files), the program can punt and leave things as they used to be.

>Actually I was asking about a bit different thing - races between
>(in UNIX terms) rename("a","/b/b/b/b"); and rename("b","/a/a/a/a");

Although UNIX likes to believe otherwise, renaming a file (i.e. changing
the contents of its dir entry) and moving it to another directory (i.e.
deleting its dir entry and creating a new one elsewhere) are two completely
different things.  I don't know if TOPS-10 even supported the latter, and in
any case I strongly doubt it was possible to change the PPN of a UFD in-place
(except if you patch the disk by hand).  Constantly fiddling with the shape
of the directory structure on a whim is definitely a UNIX kind of thing,
not so common on many other systems.

John Wilson
D Bit

######

From: Tim Shoppa <shoppa@trailing-edge.com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 30 Apr 1999 15:26:00 -0400
Organization: Trailing Edge Technology
Lines: 46
Message-ID: <3729CB88.37554CD3@trailing-edge.com>
References: <371be3e4.0@newsfeed.one.net> <7g47sv$rg5$2@antiochus.ultra.net> <7g4fkm$j34@bonkers.taronga.com> <7g748e$no4$2@ligarius.ultra.net> <FAwL3E.26G@midway.uchicago.edu> <7g9j27$88h$1@antiochus.ultra.net> <925499428.5752.0.nnrp-07.9e98250c@news.demon.co.uk>
NNTP-Posting-Host: trailing-edge.wdn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.0     DEC 3000 Model 300L)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news-fra.pop.de!newsfeed.nacamar.de!feeder.qis.net!yellow.newsread.com!netaxs.com!newsread.com!netaxs.newsread.com!usenet

Gareth Alun Evans wrote:
> 
> jmfbahciv@aol.com wrote in message <7g9j27$88h$1@antiochus.ultra.net>...
> > I remember
> >that read an -11 DECtape on an -11 was ssssoooooo sssslllloooowwwww
> >because the tape had to be read all the way to the EOT and back
> >again.  Again, I don't know the implementation details.  Maybe
> >somebody else can talk about that.
> 
> ISTR writing a device driver for DECTape on a PDP-11, back
> in  73/74. There were 576 blocks, and the block number was
> permanently encoded on the tape. Certainly there was some
> jiggery-pokery necessary to get to the correct block on the
> tape. Once you found it, you needed to back up 3 blocks
> and then come past again at speed.

Most of the distributed DECTape drivers got around this by
searching for the (block you want minus three or four) and then
getting to the right speed.  For example, in the RT-11 5.7
DECTape driver:

RETRY:  TSTB    @#TCST          ;TAPE UP TO SPEED?
        BPL     10$             ;NOPE...
        JMP     FORW1           ;YES-AVOID STOPPING TAPE

10$:    ADD     #4,BWANT        ;NO-IT TAKES 4 BLOCKS TO START AND STOP
        CMP     (PC)+,(SP)      ;MAKE AN ATTEMPT TO START IN THE
BWANT:   .WORD  0               ;RIGHT DIRECTION BASED ON LAST BLOCK
        JMP     DIRECT          ;DESIRED

> DECCassettes were another matter. You could only write
> (append) to the end of the tape. There were block gaps
> and file gaps for navigating. ISTR that you had to find
> the EOT mark, back up a couple of blocks, come
> past again at speed, reading those blocks and
> then switch to writing at the last minute.

Yes, DECCassettes require all sorts of special
code in the utilities to handle their unique nature.  DECTapes,
though, are for most purposes just block-addressible devices.

-- 
 Tim Shoppa                        Email: shoppa@trailing-edge.com
 Trailing Edge Technology          WWW:   http://www.trailing-edge.com/
 7328 Bradley Blvd                 Voice: 301-767-5917
 Bethesda, MD, USA 20817           Fax:   301-767-5927

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7g0con$68m@fohnix.metronet.com> <37241351.F947285A@tnglwood.demon.co.uk> <Pine.NXT.4.20.9904261200000.15191-100000@Tomobiki-Cho.CAC.Washington.EDU>
Organization: Plethora . Net - More net, less spam!
X-Newsreader: trn 4.0-test70 (17 January 1999)
From: seebs@plethora.net (Peter Seebach)
Lines: 28
Message-ID: <m7lW2.2057$WA4.398156@ptah.visi.com>
Date: Fri, 30 Apr 1999 16:46:10 GMT
NNTP-Posting-Host: 205.166.146.8
X-Complaints-To: abuse@visi.com
X-Trace: ptah.visi.com 925490770 205.166.146.8 (Fri, 30 Apr 1999 11:46:10 CDT)
NNTP-Posting-Date: Fri, 30 Apr 1999 11:46:10 CDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newspeer.clara.net!news.clara.net!newsfeed.nacamar.de!europa.netcrusader.net!209.98.98.32!chippy.visi.com!news-out.visi.com!ptah.visi.com!not-for-mail

In article <Pine.NXT.4.20.9904261200000.15191-100000@Tomobiki-Cho.CAC.Washington.EDU>,
Mark Crispin  <mrc@CAC.Washington.EDU> wrote:
>Be careful.  There are those of us who know quite a bit about UNIX and
>Christianity, and can say with some authority that the main thing wrong
>with criticism of either is that it is understated.

So far, I've mostly found that "problems with Unix" are really not problems
with Unix - not always, but very often.

I recall PINE used to have quite a bit of trouble porting - because someone
somewhere got the idea that, depsite several standards, you should *never*
include <unistd.h> if you could get away with it, so several ports ended up
not working - because they needed (as is correct) <unistd.h>.

Generally, people who have elaborate "porting" mechanisms between Unixes are
just creating problems; most of the time, it turns out that the existing
portable mechanism that would work on all but a few boxes was already there,
and, well, if you try to reinvent the wheel, you'll lose.

(And if you want to argue the other one, I'd be happy to do it, but this is
not the place.)

-s
-- 
Copyright 1999, All rights reserved.  Peter Seebach / seebs@plethora.net
C/Unix wizard, Pro-commerce radical, Spam fighter.  Boycott Spamazon!
Will work for interesting hardware.  http://www.plethora.net/~seebs/
Visit my new ISP <URL:http://www.plethora.net/> --- More Net, Less Spam!

######

From: "Gareth Alun Evans" <gareth@cemetery.demon.co.uk>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 30 Apr 1999 19:47:46 +0100
Message-ID: <925499428.5752.0.nnrp-07.9e98250c@news.demon.co.uk>
References: <371be3e4.0@newsfeed.one.net> <7g47sv$rg5$2@antiochus.ultra.net> <7g4fkm$j34@bonkers.taronga.com> <7g748e$no4$2@ligarius.ultra.net> <FAwL3E.26G@midway.uchicago.edu> <7g9j27$88h$1@antiochus.ultra.net>
NNTP-Posting-Host: cemetery.demon.co.uk
X-NNTP-Posting-Host: cemetery.demon.co.uk:158.152.37.12
X-Trace: news.demon.co.uk 925499428 nnrp-07:5752 NO-IDENT cemetery.demon.co.uk:158.152.37.12
X-Complaints-To: abuse@demon.net
X-Newsreader: Microsoft Outlook Express 4.72.2106.4
X-MimeOLE: Produced By Microsoft MimeOLE V4.72.2106.4
Lines: 26
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.icl.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!cemetery.demon.co.uk!not-for-mail

jmfbahciv@aol.com wrote in message <7g9j27$88h$1@antiochus.ultra.net>...
> I remember
>that read an -11 DECtape on an -11 was ssssoooooo sssslllloooowwwww
>because the tape had to be read all the way to the EOT and back
>again.  Again, I don't know the implementation details.  Maybe
>somebody else can talk about that.

ISTR writing a device driver for DECTape on a PDP-11, back
in  73/74. There were 576 blocks, and the block number was
permanently encoded on the tape. Certainly there was some
jiggery-pokery necessary to get to the correct block on the
tape. Once you found it, you needed to back up 3 blocks
and then come past again at speed. ISTR also that you
could read and write in both directions.

DECCassettes were another matter. You could only write
(append) to the end of the tape. There were block gaps
and file gaps for navigating. ISTR that you had to find
the EOT mark, back up a couple of blocks, come
past again at speed, reading those blocks and
then switch to writing at the last minute.

9 track tapes never programmed them, only used
them as backup media.

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 30 Apr 1999 20:46:55 -0400
Organization: -ENOENT
Lines: 24
Message-ID: <7gditv$il6@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7gb4n6$795$1@shell3.ba.best.com> <7gc02r$go9@weyl.math.psu.edu> <7gcoig$63f$1@shell3.ba.best.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!newspeer.monmouth.com!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7gcoig$63f$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>
>*) Open channel 1 to disk.
>
>*) Do an ENTER on channel 1 to RESUME.TXT.  If the file is already open for
>   output by another job, return error code 3 (File Being Modified). Since the
>   file name already exists, this will be a superceding enter.  Copy the file
>   permissions for the existing file into the DDB.  (The old file remains on
>   disk, and can still be read.)

>*) LOOKUP RESUME.TXT on channel 2.  This locates the original file; the new
>   version being written channel 1 is known only to the Monitor and is not
>   pointed to by any directory.  Do a series of IN UUOs on channel 2 and OUT
>   UUOs on channel 1.  CLOSE open file on channel 2 at EOF.

D'oh. So new link is created only upon close() and two processes can't write
into the same file, right? What happens upon the crash?

Another question: how did you work with subdirectories? BTW, is there such
thing as list of UUOs with descriptions? In some accessible place, that is.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 30 Apr 1999 21:05:55 -0400
Organization: -ENOENT
Lines: 30
Message-ID: <7gdk1j$ilu@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7gcoig$63f$1@shell3.ba.best.com> <7gcr9l$hn5@weyl.math.psu.edu> <3729fa95.0@news.wizvax.net>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!sunqbc.risq.qc.ca!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <3729fa95.0@news.wizvax.net>, John Wilson <wilson@dbit.com> wrote:
>>Actually I was asking about a bit different thing - races between
>>(in UNIX terms) rename("a","/b/b/b/b"); and rename("b","/a/a/a/a");
>
>Although UNIX likes to believe otherwise, renaming a file (i.e. changing
>the contents of its dir entry) and moving it to another directory (i.e.
>deleting its dir entry and creating a new one elsewhere) are two completely
>different things.  I don't know if TOPS-10 even supported the latter, and in
	Depends on which semantics you prefer to use.

>any case I strongly doubt it was possible to change the PPN of a UFD in-place
							^^^^^
	Excuse me? Do you imply that user could have only one directory?
AFAICS PPN == UID+GID and it looks like per-user thing, not per-directory one.

>(except if you patch the disk by hand).  Constantly fiddling with the shape
>of the directory structure on a whim is definitely a UNIX kind of thing,
>not so common on many other systems.
	Religious issues aside (files as separate objects vs. files as
something hanging from the directory entry, etc.), how did the namespace
look like on TOPS-10? If you want a holy war - let's take it to email.
I suspect that our tastes are mutually incompatible in that area and all
of those flamewars were beaten to death *way* too many times (case sensibility,
unified tree, files as fundamentally nameless objects, versioning lossage,
yodda, yodda). Actual information about TOPS-10 is somewhat more interesting
than reshuffling this stuff again.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: Robert Billing <unclebob@tnglwood.demon.co.uk>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sat, 01 May 1999 07:45:17 +0000
Organization: Tanglewood
Message-ID: <372AB10D.3F5A3E51@tnglwood.demon.co.uk>
References: <371be3e4.0@newsfeed.one.net> <7g47sv$rg5$2@antiochus.ultra.net> <7g4fkm$j34@bonkers.taronga.com> <3726BC51.E80F0C3D@tnglwood.demon.co.uk> <3729c2b2.0@news.wizvax.net>
NNTP-Posting-Host: localhost.demon.co.uk
X-NNTP-Posting-Host: tnglwood.demon.co.uk:158.152.132.30
X-Trace: news.demon.co.uk 925545250 nnrp-10:22424 NO-IDENT tnglwood.demon.co.uk:158.152.132.30
X-Complaints-To: abuse@demon.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.0.31 i586)
Lines: 13
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!tnglwood.demon.co.uk!not-for-mail

John Wilson wrote:

> Of course, in this case the kernel has to have the RAM disk driver and
> its file system compiled in.  Both can be pretty primitive, but still...

 Yes. The ramdisk driver is of course minimal. The file system is the
ext2 file system that it will need anyway when it is up and running.

-- 
I am Robert Billing, Christian, inventor, traveller, cook and animal
lover, I live near 0:46W 51:22N.  http://www.tnglwood.demon.co.uk/
"Bother," said Pooh, "Eeyore, ready two photon torpedoes and lock
phasers on the Heffalump, Piglet, meet me in transporter room three"

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sat, 01 May 99 08:45:32 GMT
Organization: UltraNet Communications, Inc.
Lines: 20
Message-ID: <7gel5d$j3t$1@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7g0con$68m@fohnix.metronet.com> <37241351.F947285A@tnglwood.demon.co.uk> <Pine.NXT.4.20.9904261200000.15191-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gcdik$38c@bonkers.taronga.com>
NNTP-Posting-Host: d1.dial-16.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 1 May 1999 10:31:09 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d1

In article <7gcdik$38c@bonkers.taronga.com>,
   peter@taronga.com (Peter da Silva) wrote:
>In article 
<Pine.NXT.4.20.9904261200000.15191-100000@Tomobiki-Cho.CAC.Washington.EDU>,
>Mark Crispin  <mrc@CAC.Washington.EDU> wrote:
>>Be careful.  There are those of us who know quite a bit about UNIX and
>>Christianity, and can say with some authority that the main thing wrong
>>with criticism of either is that it is understated.
>>
>>This is not the forum to continue this discussion.
>
>May I suggest alt.religion.computers ?
>
I would rather that Mark talk about how TOPS20 approached 
these problems.  He knows all this stuff and is usually
willing to talk about it.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sat, 01 May 99 08:49:54 GMT
Organization: UltraNet Communications, Inc.
Lines: 33
Message-ID: <7geldj$j3t$2@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7gcoig$63f$1@shell3.ba.best.com> <7gcr9l$hn5@weyl.math.psu.edu> <3729fa95.0@news.wizvax.net> <7gdk1j$ilu@weyl.math.psu.edu>
NNTP-Posting-Host: d1.dial-16.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 1 May 1999 10:35:31 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d1

In article <7gdk1j$ilu@weyl.math.psu.edu>,
   viro@weyl.math.psu.edu (Alexander Viro) wrote:
>In article <3729fa95.0@news.wizvax.net>, John Wilson <wilson@dbit.com> 
wrote:
<snip>

>>(except if you patch the disk by hand).  Constantly fiddling 
>>with the shape
>>of the directory structure on a whim is definitely a UNIX kind of thing,
>>not so common on many other systems.

>	Religious issues aside (files as separate objects vs. files as
>something hanging from the directory entry, etc.), how did the namespace
>look like on TOPS-10? If you want a holy war - let's take it to email.
>I suspect that our tastes are mutually incompatible in that area and all
>of those flamewars were beaten to death *way* too many times 
>(case sensibility, unified tree, files as fundamentally nameless 
>objects, versioning lossage,
>yodda, yodda). Actual information about TOPS-10 is somewhat more 
>interesting than reshuffling this stuff again.
>
I don't think John was trying to start a flame war.  One of the
reasons I didn't care for UNIX was its approach to files and 
data integrity.  There was definitely a difference in philosophy
between the three operating systems.

Please don't take this discussion off line.  I probably should
also say that some of our discussion style in the TOPS10 monitor
group included yelling.  That's not flaming.

/BAH

Subtract a hundred and four for e-mail.

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 1 May 1999 10:07:23 -0400
Organization: -ENOENT
Lines: 26
Message-ID: <7gf1qr$j8q@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <3729fa95.0@news.wizvax.net> <7gdk1j$ilu@weyl.math.psu.edu> <7geldj$j3t$2@antiochus.ultra.net>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!news.idt.net!nyd.news.ans.net!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7geldj$j3t$2@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
>I don't think John was trying to start a flame war.  One of the
>reasons I didn't care for UNIX was its approach to files and 
>data integrity.  There was definitely a difference in philosophy
>between the three operating systems.
>
>Please don't take this discussion off line.  I probably should
>also say that some of our discussion style in the TOPS10 monitor
>group included yelling.  That's not flaming.
	Yup. Probably I should formulate it more clear - judging by
the long history both here and in other froups *any* discussion of
UNIX-style vs. DEC-style filesystem namespace handling degenerates
into common flame in 3-4 hours and goes for months. I *like* UNIX-style
variant and it's more natural for me. YMMV. Certainly I'm biased - I'm
UNIX hacker and most of my kernel work is on Linux VFS and filesystems.
As for the files tied to directories - bletch! Sorry for ranting,
but I've spent last three (fsck, four) months rewriting the drivers of
FAT-derived filesystems. Maybe in July I'll be able to discuss the idea
of folding bits of file state (size, location, etc.) into the directories
without references to anatomy and physiology. At least I can hope ;-/
Until then... I'ld rather sit on my hands and abstain from describing
what I think of that.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: "Alan H. Martin" <AMartin@MA.UltraNet.Com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sat, 01 May 1999 11:26:02 -0400
Organization: UltraNet Communications , an RCN Company http://www.ultranet.com/
Lines: 55
Message-ID: <372B1D0A.2AB40504@MA.UltraNet.Com>
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu><371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net><Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7fs7mv$hii$1@ligarius.ultra.net> <7fsjbg$91o$1@zingo.tninet.se> <3723DBA5.E5B2FF0@MA.UltraNet.Com> <7g1gvb$bid$1@ligarius.ultra.net>
NNTP-Posting-Host: 207-172-216-103.s103.tnt1.sbo.ma.dialup.rcn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 1 May 1999 15:26:37 GMT
X-Mailer: Mozilla 4.5 [en] (Win95; U)
X-Accept-Language: en,en-US,en-GB,es
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!not-for-mail

jmfbahciv@aol.com wrote:
> 
> In article <3723DBA5.E5B2FF0@MA.UltraNet.Com>,
>    "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
> >"Peter Hendén" wrote:
> >> jmfbahciv@aol.com wrote:
> >....
> >> >The reason I didn't like to use the -20 is because ...; it was
> >> >too slow for my taste ...
> >>
> >> Wasn't this dependent on whether your programs were
> >> TOPS-10 (used PA1050) or TOPS-20 native. As I recall
> >> (1977 - 80) we had a lot of users on our machine running
> >> Interlisp-20 and a bunch of other native programs, and they
> >> didn't get sluggish until long after the TOPS-10 progs did.
> >
> >Could be.  However, here's something to consider about -10 vs. -20 I/O
> >performance:
> >
> >On the one hand, I don't know how fast PA1050-simulated buffered I/O went
> >compared to PMAP.  However, I definitely recall hearing that the DUMPI and
> >DUMPO JSY' that were intended for simulation of unbuffered I/O were *the*
> >fastest form of I/O on Tops-20.
> 
> You completely missed my point.  ...

Peter asserted that under a heavy system load, native programs in the mix had
better responsiveness than programs which used PA1050.

You asserted that a system running Tops-20 could handle fewer users than a
system running Tops-10.


>... Just because one user used
> something that did UUOs, should NOT have affected another user's
> performance who is using JSYSes.  ...

What should we conclude?

1.  Peter's Tops-20 system didn't actually behave the way he recalls.
	-or-
2.  Tops-20's performance didn't actually degrade the way you would like.


I asserted that programs that used the native Tops-20 calls intended for use
by PA1050 itself ran faster than programs that used native Tops-20 calls
intended for use by native applications (implicitly, PMAP).  (This was
regardless of whether programs that ran on top of PA1050 were slower or faster
than programs that ran right on top of Tops-20).


Gee, would you have liked PMAP to have better performance than DUMPI/DUMPO?
				/AHM
-- 
Alan Howard Martin				AMartin@MA.UltraNet.Com

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <3729fa95.0@news.wizvax.net> <7gdk1j$ilu@weyl.math.psu.edu> <7geldj$j3t$2@antiochus.ultra.net>
Organization: D Bit, Troy, NY
From: wilson@dbit.com (John Wilson)
NNTP-Posting-Host: dbit.dbit.com
Message-ID: <372b2c32.0@news.wizvax.net>
Date: 1 May 1999 12:30:42 -0500
X-Trace: 1 May 1999 12:30:42 -0500, dbit.dbit.com
Lines: 38
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newspeer.monmouth.com!newsfeed.wizvax.net!news.wizvax.net!dbit.com!wilson

In article <7geldj$j3t$2@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
>I don't think John was trying to start a flame war.  One of the
>reasons I didn't care for UNIX was its approach to files and 
>data integrity.  There was definitely a difference in philosophy
>between the three operating systems.

I certainly wasn't trying to flame, I was seriously trying to make a point --
I used flat-directory systems for years before the first time I saw a
tree-structured system and never felt like something was missing.  Actually,
16 years later I *still* haven't gotten the hang of controlling clutter on
tree-structured dirs, and I notice that every tree-structured OS comes with
some kind of utility to help you find the files you've lost somewhere in
the tree, so I don't think I'm the only one.

My point is just that when you're used to the flat system, you *don't*
sit around cursing its inability to move wads of files from one subtree to
another etc., the way many UNIX users do when exposed to a non-tree system.
So the fact that you may not be able to "rename" a file to another directory
(i.e. delete its dir entry and retrieval list, and create new ones in another
directory, which is hardly a "rename" operation unless you think that files
exist independently of their dir entries, which again is a taste issue and
while it's central to the UNIX file systems, it's the opposite of many other
systems, whose users probably don't see their way as a deficiency) doesn't
chafe on your nerves so much, especially since the issue may be clouded by
disk structures where files from multiple physical disks are intermingled in
the same directory structure, so two files that UNIX users might think differ
only in their "names" actually have their data (and dir entries/retrieval)
on entirely different packs.  Copying a file to another drive and calling
that a "rename" is just too weird!

Similarly, creating a directory may be a relatively big deal on non-UNIXy
systems, that's the other thing I meant about tossing files around on a whim.
When creating a new directory is something you have to beg a system operator
to do for you (and even then you need a password to get into it), it doesn't
happen so much.

John Wilson
D Bit

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 1 May 1999 12:57:33 -0700
Organization: Chez Inwap
Message-ID: <7gfmbd$ndv$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gc02r$go9@weyl.math.psu.edu> <7gcoig$63f$1@shell3.ba.best.com> <7gcr9l$hn5@weyl.math.psu.edu>
Lines: 25
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925588658 208 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gcr9l$hn5@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>In article <7gcoig$63f$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>>Here is one command on TOPS-10 that cannot done the same way on Unix:
>>	COPY RESUME.TXT=HEADER.TXT,RESUME.TXT,FOOTER.TXT
>
>	Can be done almost the same way. rename() over the open file
>unlinks it, but leaves the thing intact.

Won't work.  Renaming the old file means that it is no longer accessable
via its old name, therefore the open-for-input (which comes later) will
fail with "file not found".

>Actually I was asking about a bit different thing - races between
>(in UNIX terms) rename("a","/b/b/b/b"); and rename("b","/a/a/a/a");

The UFD or SFD involved would be marked as being busy by the first
RENAME.  The second one would be blocked until the first one had gone
to completion.

	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 1 May 1999 13:22:04 -0700
Organization: Chez Inwap
Message-ID: <7gfnpc$3in$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gcr9l$hn5@weyl.math.psu.edu> <3729fa95.0@news.wizvax.net> <7gdk1j$ilu@weyl.math.psu.edu>
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925590128 226 inwap@206.184.139.134
Lines: 43
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news-fra.pop.de!newsfeed.nacamar.de!dispose.news.demon.net!demon!newsfeed.berkeley.edu!news2.best.com!news1.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gdk1j$ilu@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>any case I strongly doubt it was possible to change the PPN of a UFD in-place
>							^^^^^
>	Excuse me? Do you imply that user could have only one directory?
>AFAICS PPN == UID+GID and it looks like per-user thing, not per-directory one.

Each user could have only one top-level directory per disk.

If LOGIN determined that my PPN was 13,17 and that I was allowed to use
DSKA and DSKB, then the logical name "DSK:" refered to both
DSKA:[13,17] and DSKB:[13,17].  When I typed the DIR command, SYS:DIRECT.EXE
would read DSKA:[1,1]#000013000017.UFD and DSKB:[1,1]#000013000017.UFD.

(The poundsign notation says to take the 12 octal digits which follow and
convert them to a 36-bit quantity.  In the example above, the first 6
digits are the project number, and the next 6 digits are the programmer
number; combined, they are the PPN.)

The ersatz device name "MFD" was an alias for "DSK:[1,1]".  The only entries
in the Master File Directory were the UFDs (User File Directories).
UFDs were the login directorys, and could contain files or SFDs (Sub File
Directories).  The limit was 5 levels if SFDs.

If I said "DIR [,,TECO,SRC]", it would read DSKA:[13,17,TECO]SRC.SFD and/or
DSKB:[13,17,TECO]SRC.SFD.

Like Unix, TOPS-10 had file permissions specified as three octal digits,
one each for owner, project(group) and other.  But it had a different
method of determining the owner.  If you created a directory that was
writeable by world, then any new files created there belonged to you
(the owner of the directory) and not to the PPN of the creator.

You were a member of the group owning the file if the project part of the
file's directory matched the left half of your PPN.  You were the owner of a
file if the programmer part of the file's directory matched the right half
PPN.

	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 1 May 1999 14:12:06 -0400
Organization: -ENOENT
Lines: 72
Message-ID: <7gfg5m$jen@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7gdk1j$ilu@weyl.math.psu.edu> <7geldj$j3t$2@antiochus.ultra.net> <372b2c32.0@news.wizvax.net>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newspump.monmouth.com!newspeer.monmouth.com!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

>I certainly wasn't trying to flame, I was seriously trying to make a point --
>I used flat-directory systems for years before the first time I saw a
	Ditto.
>tree-structured system and never felt like something was missing.  Actually,
>16 years later I *still* haven't gotten the hang of controlling clutter on
>tree-structured dirs, and I notice that every tree-structured OS comes with
>some kind of utility to help you find the files you've lost somewhere in
>the tree, so I don't think I'm the only one.
	Dunno. Unless you have a habit of leaving your files in others'
homedirs it's a matter of not creating a chaos on your own account ;-)
Keep everything in a single directory if you like it better, but IMO creating
a subdirectories for (large enough) projects you are doing is more convenient.

>My point is just that when you're used to the flat system, you *don't*
>sit around cursing its inability to move wads of files from one subtree to
>another etc., the way many UNIX users do when exposed to a non-tree system.
>So the fact that you may not be able to "rename" a file to another directory
>(i.e. delete its dir entry and retrieval list, and create new ones in another
>directory, which is hardly a "rename" operation unless you think that files
>exist independently of their dir entries, which again is a taste issue and

	Yes, I do. Please, let's not go there. Besides, if I understood
the situation with TOPS-10 right, it also had the same opinion. I.e. file
remains alive even if you remove directory entry; better yet, opening file
for write (ENTER) postpones directory entry creation until you close the file.

>while it's central to the UNIX file systems, it's the opposite of many other
>systems, whose users probably don't see their way as a deficiency) doesn't
>chafe on your nerves so much, especially since the issue may be clouded by
>disk structures where files from multiple physical disks are intermingled in
>the same directory structure, so two files that UNIX users might think differ
>only in their "names" actually have their data (and dir entries/retrieval)
>on entirely different packs.  Copying a file to another drive and calling
>that a "rename" is just too weird!

	Hold on. If you are talking about syscall (rename()) it will return
an error in case you are crossing disk boundary (-EXDEV). If you are talking
about standard utility - it's called mv. 'move' seems to be perfectly
reasonable verb to describe an action (different from syscall - it tries to
use rename() and if it says that target is on the other device mv copies
the file and unlinks the original).

>Similarly, creating a directory may be a relatively big deal on non-UNIXy
>systems, that's the other thing I meant about tossing files around on a whim.
>When creating a new directory is something you have to beg a system operator
>to do for you (and even then you need a password to get into it), it doesn't
>happen so much.

	Oh, yes. Now, whether it's win or lossage is a different question ;-)
I worked with flat system, actually I've started there (RSX). I was not happy
when I tried to recall which of three projects FOO.FOR belonged to (each of
them was beyond 20 files, IIRC). Getting a separate account for each project
meant the hassles you've described above. I had Bourne's book and I could only
grit my teeth - I realized that it would be just a matter of creating 3
subdirectories and moving the stuff there... and switching to a different OS.

	But you really left me puzzled - earlier posting mentioned passing
a subdirectory as an argument to ENTER. So, was it really flat namespace?
What about TOPS-20 and ITS?
	Another question: exact semantics of ENTER and LOOKUP. How did it
actually work? As far as I understood second ENTER was impossible until the
file was closed. Was it done via creation of temporary entry for a new
version and "upgrading' it to the normal upon close? If it was kept
in core and written on disk only upon close - what happened in case of crash?
Another question being: suppose FOO.BAR is a large file and I want to change
10 words in the middle. How it would be done? (words as in 'machine word',
not 'sequence of letters, surrounded by non-letters'). Yet another one:
suppose you want to add something to the end of file. What would it take?

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
From: lisard@zetnet.co.uk
X-Everything: Net-Tamer V 1.08X
Date: 01 May 1999 14:22:24 GMT
Message-ID: <372b0e20$0$494@news.zetnet.co.uk>
NNTP-Posting-Host: man-022.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 925568544 494 194.247.41.27
Lines: 14
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!dispose.news.demon.net!demon!newspeer.clara.net!news.clara.net!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail



On 1999-04-29 viro@weyl.math.psu.edu(AlexanderViro) said:
   :<chuckle> Always fine to watch C++
   :fanatics claiming that OOP is impossible in C. Wonder how do they
   :>react on OOP in assembler...

Well, OS6 did OOP in BCPL (as did Amiga Exec, many years later).
Dispatch tables have always been a part of the competent programmer's
vocabulary. Apparently, Spacewar used them (I think I read that here),
and that's waaaaay back. OOP just formalised the concept, gave it a
name, and <grrr> bastardised the whole thing for ever more </grrr>.
--
Communa (lisard@zetnet.co.uk) -- you know soft spoken changes nothing

######

From: rob.nicholson@zetnet.co.uk (Rob Nicholson)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sat, 01 May 1999 17:12:36 GMT
Organization: Rainow Systems Ltd.
Reply-To: rob.nicholson@zetnet.co.uk
Message-ID: <372b323c.1198971@news.zetnet.co.uk>
References: <371be3e4.0@newsfeed.one.net> <3729fa95.0@news.wizvax.net> <7gdk1j$ilu@weyl.math.psu.edu> <7geldj$j3t$2@antiochus.ultra.net> <372b2c32.0@news.wizvax.net>
X-Newsreader: Forte Agent 1.5/32.452
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
NNTP-Posting-Host: manr-050.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 925578763 495 194.247.43.180
Lines: 11
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!unlisys!news.snafu.de!newsfeed.nacamar.de!dispose.news.demon.net!demon!newspeer.clara.net!news.clara.net!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail

>tree-structured system and never felt like something was missing.  Actually,
>16 years later I *still* haven't gotten the hang of controlling clutter on
>tree-structured dirs, and I notice that every tree-structured OS comes with
>some kind of utility to help you find the files you've lost somewhere in
>the tree, so I don't think I'm the only one.

True but try managing a site with hundreds of users all sharing files together.
With a flat structure you end up hundreds of folders each containing many files.
It's an administration nightmare.

Rob.

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 1 May 1999 17:35:27 -0400
Organization: -ENOENT
Lines: 36
Message-ID: <7gfs2v$jlq@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7gcoig$63f$1@shell3.ba.best.com> <7gcr9l$hn5@weyl.math.psu.edu> <7gfmbd$ndv$1@shell3.ba.best.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!sunqbc.risq.qc.ca!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7gfmbd$ndv$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>In article <7gcr9l$hn5@weyl.math.psu.edu>,
>Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>In article <7gcoig$63f$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>>>Here is one command on TOPS-10 that cannot done the same way on Unix:
>>>	COPY RESUME.TXT=HEADER.TXT,RESUME.TXT,FOOTER.TXT
>>
>>	Can be done almost the same way. rename() over the open file
>>unlinks it, but leaves the thing intact.
>
>Won't work.  Renaming the old file means that it is no longer accessable
>via its old name, therefore the open-for-input (which comes later) will
>fail with "file not found".

Open sources first, then create a temp. file and rename it over
the destination. Oh, darn... 16 channels only...

>>Actually I was asking about a bit different thing - races between
>>(in UNIX terms) rename("a","/b/b/b/b"); and rename("b","/a/a/a/a");
>
>The UFD or SFD involved would be marked as being busy by the first
>RENAME.  The second one would be blocked until the first one had gone
>to completion.

<chuckle> That's what 4.4BSD does. And has the race in question. Let's modify
it a bit:
process A - PWD = /a/a/a/a, rename("/b/b","a");
process B - PWD = /b/b/b/b, rename("/a/a","b");
They don't come anywhere near each other, even on lookup phase.

And if you have lookup and rename phases separated you are in worse trouble -
all lookups may suceed before any of renames will happen.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 1 May 1999 17:50:17 -0400
Organization: -ENOENT
Lines: 22
Message-ID: <7gfsup$jmk@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <3729fa95.0@news.wizvax.net> <7gdk1j$ilu@weyl.math.psu.edu> <7gfnpc$3in$1@shell3.ba.best.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newspump.monmouth.com!newspeer.monmouth.com!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7gfnpc$3in$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>In article <7gdk1j$ilu@weyl.math.psu.edu>,
>Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>>any case I strongly doubt it was possible to change the PPN of a UFD in-place
>>							^^^^^
>>	Excuse me? Do you imply that user could have only one directory?
>>AFAICS PPN == UID+GID and it looks like per-user thing, not per-directory one.
>
[namespace wasn't flat, homedirs were all in root]
[disks were kinda-sorta unionfs'ed]
[2^18 possible UIDs and 2^18 possible GIDs]
[maximal depth was 5 (or 6?)]
[for UID it had the same policy as BSD has for GID in SGID directories]
(BTW, what about GID (== project) part?)

D'oh. It's much nicer than the picture implied by John. Was there a notion
of current working directory? And how did one open [42,6,FOO,BAR]BAZ (if
I didn't mess the notation) for write?

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7gdk1j$ilu@weyl.math.psu.edu> <7geldj$j3t$2@antiochus.ultra.net> <372b2c32.0@news.wizvax.net>
Organization: Plethora . Net - More net, less spam!
X-Newsreader: trn 4.0-test70 (17 January 1999)
From: seebs@plethora.net (Peter Seebach)
Lines: 36
Message-ID: <xRIW2.2253$WA4.441694@ptah.visi.com>
Date: Sat, 01 May 1999 19:45:33 GMT
NNTP-Posting-Host: 205.166.146.8
X-Complaints-To: abuse@visi.com
X-Trace: ptah.visi.com 925587933 205.166.146.8 (Sat, 01 May 1999 14:45:33 CDT)
NNTP-Posting-Date: Sat, 01 May 1999 14:45:33 CDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!chippy.visi.com!news-out.visi.com!ptah.visi.com!not-for-mail

In article <372b2c32.0@news.wizvax.net>, John Wilson <wilson@dbit.com> wrote:
>I certainly wasn't trying to flame, I was seriously trying to make a point --
>I used flat-directory systems for years before the first time I saw a
>tree-structured system and never felt like something was missing.  Actually,
>16 years later I *still* haven't gotten the hang of controlling clutter on
>tree-structured dirs, and I notice that every tree-structured OS comes with
>some kind of utility to help you find the files you've lost somewhere in
>the tree, so I don't think I'm the only one.

I don't think that's a fair criticism.

You're implying that it's a fault of the tree-structured environment.

I think it's a limitation of my memory.  I have a fair number of files; a
quick scan of my personal home directory shows 107,196 files.  That's a lot of
things to remember.  Mostly, I don't have to remember them all - partially
because the tree structure allows me to remember broader rules, and partially
because I have utilities that allow me to find things.

Without the tree, I'd be unable to keep track of a tenth as many files.

>Similarly, creating a directory may be a relatively big deal on non-UNIXy
>systems, that's the other thing I meant about tossing files around on a whim.
>When creating a new directory is something you have to beg a system operator
>to do for you (and even then you need a password to get into it), it doesn't
>happen so much.

Which would suck; the ability to create some temporary namespace to sort
things in is a gigantic win.

-s
-- 
Copyright 1999, All rights reserved.  Peter Seebach / seebs@plethora.net
C/Unix wizard, Pro-commerce radical, Spam fighter.  Boycott Spamazon!
Will work for interesting hardware.  http://www.plethora.net/~seebs/
Visit my new ISP <URL:http://www.plethora.net/> --- More Net, Less Spam!

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 1 May 1999 23:17:25 -0700
Organization: Chez Inwap
Message-ID: <7ggqll$u0$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7geldj$j3t$2@antiochus.ultra.net> <372b2c32.0@news.wizvax.net> <7gfg5m$jen@weyl.math.psu.edu>
Lines: 81
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925625850 215 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!nntp.primenet.com!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gfg5m$jen@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>	But you really left me puzzled - earlier posting mentioned passing
>a subdirectory as an argument to ENTER. So, was it really flat namespace?
>What about TOPS-20 and ITS?

Format of the four-word LOOKUP/ENTER block:
  E+0: Filename; 6 SIXBIT characters for file names, PPN for UFDs.
  E+1: Extension; 3 SIXBIT characters in LH, access date in RH.
  E+2: File protection, modify date and time
  E+3: Directory, one of:
	P,,PN - file will be looked for or created in specified UFD.
	0,,E2 - E2 is address of PATH block, which contains PPN and
		up to 5 levels of SFD.
	0,,0  - current working directory.

The current working directory was changed via the PATH. UUO and/or the PATH
command.  The PATH. UUO also was used to enable something that predates Sun's
"translucent file system".  On LOOKUP, if the file is not found in the
current SFD, scan higher-level SFDs until a UFD is reached.  If still not
found, search the directory declared as the LIB: PPN.  If still not found,
search SYS: and/or NEW:.  These features can be enabled and disabled
separately.

This was quite handy for experimental sources.  If the sources for
TECO were in [,,SOURCE,TECO] and your current path was
[,,SOURCE,TECO,NEW]/SCAN, then only the modified sources need be copied
to [,,SOURCE,TECO,NEW] and the compiler would find all the unmodified
files in [,,SOURCE,TECO].

TOPS-20 uses different punctuation than TOPS-10 to indicate directories.

>	Another question: exact semantics of ENTER and LOOKUP. How did it
>actually work? As far as I understood second ENTER was impossible until the
>file was closed.

Yep.

>Was it done via creation of temporary entry for a new
>version and "upgrading' it to the normal upon close?

Yep.

>If it was kept
>in core and written on disk only upon close - what happened in case of crash?

It's gone.  That's why programs that want to be sure that their log file
survives a crash uses the open-append-close method instead of keeping the
file open.

>Another question being: suppose FOO.BAR is a large file and I want to change
>10 words in the middle. How it would be done?

That is a topic I was holding back until explictly asked.

If a program does an OPEN and LOOKUP, and then does an ENTER on the same
channel using the same four-word LOOKUP/ENTER/RENAME block, then the file
would be in update mode.  Use USETI to set the input pointer to the desired
block in the file.  Use the IN UUO to read the current block.

  1 block = 128 36-bit words = 256 halfwords = 512 9-bit bytes = 640 7-bit
  ASCII characters = 768 SIXBIT bytes = 1152 4-bit nybbles (576 8-bit bytes @
  2 nybbles per byte).

Update the copy of the block in memory.  Use USETO to set the output pointer
back to the specified block number.  Use OUTPUT to rewrite the block.
 
>suppose you want to add something to the end of file. What would it take?

Use -1 as the argument to USETI to select the end of the file.  The
next OUTPUT will append to the file.

One of the file protection codes was for append-only.  Set properly, other
people could append to your file (until your disk quota was used up) but could
not delete, supersede or update your file.

	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: dpeschel@u.washington.edu (Derek Peschel)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 1 May 1999 23:39:59 GMT
Organization: University of Washington, Seattle
Lines: 19
Message-ID: <7gg3cf$c8u$1@nntp1.u.washington.edu>
References: <372b0e20$0$494@news.zetnet.co.uk>
NNTP-Posting-Host: saul6.u.washington.edu
X-Trace: nntp1.u.washington.edu 925601999 12574 (None) 140.142.17.37
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: dpeschel
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!masternews.telia.net!news.algonet.se!algonet!newsfeed.berkeley.edu!news.u.washington.edu!dpeschel

In article <372b0e20$0$494@news.zetnet.co.uk>,  <lisard@zetnet.co.uk> wrote:

>Well, OS6 did OOP in BCPL (as did Amiga Exec, many years later).
>Dispatch tables have always been a part of the competent programmer's
>vocabulary. Apparently, Spacewar used them (I think I read that here),
>and that's waaaaay back. OOP just formalised the concept, gave it a
>name, and <grrr> bastardised the whole thing for ever more </grrr>.

I've always wanted to know more about BCPL.  Specifically, I've heard about
the "global vector".  Is this related to the dispatch table you talked
about?  Or was that only an OS6 creation?

I've also wondered if BCPL's run-time overhead (i.e., the global vector) is
too heavy.  Amiga owners have complained about the overhead.  I would think
C is better, but perhaps the C people have simply brainwashed me.  What do
you think?  Is it good to try to create a "no-overhead" language?

-- Derek

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 1 May 1999 23:42:51 -0700
Organization: Chez Inwap
Message-ID: <7ggs5b$7vc$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gdk1j$ilu@weyl.math.psu.edu> <7gfnpc$3in$1@shell3.ba.best.com> <7gfsup$jmk@weyl.math.psu.edu>
Lines: 41
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925627377 205 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!newsfeed.berkeley.edu!news2.best.com!news1.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gfsup$jmk@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>D'oh. It's much nicer than the picture implied by John. Was there a notion
>of current working directory? And how did one open [42,6,FOO,BAR]BAZ (if
>I didn't mess the notation) for write?

OPNBLK:	0			; Mode, 0=ASCII, 10-17=binary
	SIXBIT	/DSK/		; Device
	OBUF,,0			; Addr of output buffer header

OBUF:	BLOCK	3		; Will point to start of buffer ring

FILBLK:	SIXBIT	/BAZ/		; File name
	SIXBIT	//		; Extension
	0			; Use default file permissions
	0,,PTHBLK		; Pointer to PATH block

PTHBLK:	0			; For compatibility with PATH. UUO (device)
	0			; For compatibility with PATH. UUO (flags)
	42,,6			; Directory
	SIXBIT	/FOO/		; 1st SFD
	SIXBIT	/BAR/		; 2nd SFD
	EXP	0,0,0		; 3rd, 4th, 5th SFDs

START:	OPEN	1,OPNBLK	; Open channel 1 to disk
	  HALT			; Can never fail
	ENTER	1,FILBLK	; Create or supersede output file
	  HALT			; Should check error code in RH of FILBLK+1
	JFCL			; Insert code here to write to file
	CLOSE	1,		; Close file, update directory entry
	RELEAS	1,		; Done with channel
	EXIT			; Exit to monitor

	END	START		; This tells the assembler our start address


		-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

rom: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 1 May 1999 23:46:02 -0700
Organization: Chez Inwap
Message-ID: <7ggsba$8ta$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gc02r$go9@weyl.math.psu.edu> <7gcoig$63f$1@shell3.ba.best.com> <7gditv$il6@weyl.math.psu.edu>
Lines: 15
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925627567 209 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.wirehub.nl!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gditv$il6@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>Another question: how did you work with subdirectories? BTW, is there such
>thing as list of UUOs with descriptions? In some accessible place, that is.

Eventually they will be online on Eric Smith's site
	http://www.36bit.org/dec/manual/
but I do not know of any Monitor Calls Manual online at this time.

		-Joe

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: David Scheidt <dscheidt@enteract.com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 1 May 1999 23:50:20 GMT
Organization: EnterAct L.L.C. Turbo-Elite News Server
Lines: 13
Message-ID: <7gg3vs$lcc$1@eve.enteract.com>
References: <371be3e4.0@newsfeed.one.net> <3729fa95.0@news.wizvax.net> <7gdk1j$ilu@weyl.math.psu.edu> <7gfnpc$3in$1@shell3.ba.best.com> <7gfsup$jmk@weyl.math.psu.edu>
NNTP-Posting-Host: 207.229.143.40
User-Agent: tin/pre-1.4-19990216 ("Styrofoam") (UNIX) (FreeBSD/3.1-STABLE (i386))
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!netnews.com!newsfeed.enteract.com!news.enteract.com!not-for-mail

In alt.folklore.computers Alexander Viro <viro@weyl.math.psu.edu> wrote:
: [for UID it had the same policy as BSD has for GID in SGID directories]

BSD has this behavior by default, in all directories.  System V does this if 
the SGID bit is set.  There is some US government standard that requires this
behavior.

David Scheidt

-- 
dscheidt@enteract.com
Ketchup, therefore, shows both thixotropic and pseudoplastic rheological
properties.   -- John Schmitt, in AFU

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 1 May 1999 23:59:18 -0700
Organization: Chez Inwap
Message-ID: <7ggt46$cv3$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7fs7mv$hii$1@ligarius.ultra.net> <7fsjbg$91o$1@zingo.tninet.se> <3723DBA5.E5B2FF0@MA.UltraNet.Com>
Lines: 27
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925628362 227 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <3723DBA5.E5B2FF0@MA.UltraNet.Com>,
Alan H. Martin <AMartin@MA.UltraNet.Com> wrote:
>On the one hand, I don't know how fast PA1050-simulated buffered I/O went
>compared to PMAP.  However, I definitely recall hearing that the DUMPI and
>DUMPO JSY' that were intended for simulation of unbuffered I/O were *the*
>fastest form of I/O on Tops-20.

TOPS-20 allowed the use of SIN and SOUT and BIN and BOUT to do I/O as a string
of bytes or individual bytes.  Very convenient for lazy programmers.  TOPS-10
did I/O in blocks only.

One of the programmers in Colorado noticed that a particular file copy
program, highly optimized for TOPS-10 (doing unbuffered I/O in large chunks),
was quicker than the standard TOPS-20 programs for copying files on TOPS-20.

In spite of the overhead of the TOPS-10 program running under PA1050, it
was more efficient than the TOPS-20 native programs using SIN and SOUT.

Changing the TOPS-10 program to detect when it was running under TOPS-20
and use DUMPI/DUMPO made it faster.  And using PMAP to map file pages
into user address space was even faster.
	-Joe

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 2 May 1999 02:14:44 -0400
Organization: -ENOENT
Lines: 13
Message-ID: <7ggqgk$k3e@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7gfnpc$3in$1@shell3.ba.best.com> <7gfsup$jmk@weyl.math.psu.edu> <7gg3vs$lcc$1@eve.enteract.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newspeer.monmouth.com!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7gg3vs$lcc$1@eve.enteract.com>,
David Scheidt  <dscheidt@enteract.com> wrote:
>In alt.folklore.computers Alexander Viro <viro@weyl.math.psu.edu> wrote:
>: [for UID it had the same policy as BSD has for GID in SGID directories]

<looking at the idiocy above><boggle>
Arrgh. Mea culpa. Halfway-edited sentences make nice fsckups when posted.

<selfLART> proofread before sending. *whack* </selfLART>

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 2 May 1999 05:25:39 -0400
Organization: -ENOENT
Lines: 60
Message-ID: <7gh5mj$kb3@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <372b2c32.0@news.wizvax.net> <7gfg5m$jen@weyl.math.psu.edu> <7ggqll$u0$1@shell3.ba.best.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!newshub.northeast.verio.net!howland.erols.net!newsxfer.itd.umich.edu.MISMATCH!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7ggqll$u0$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>In article <7gfg5m$jen@weyl.math.psu.edu>,
>Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>	But you really left me puzzled - earlier posting mentioned passing
>>a subdirectory as an argument to ENTER. So, was it really flat namespace?
>>What about TOPS-20 and ITS?
>
>Format of the four-word LOOKUP/ENTER block:
>  E+0: Filename; 6 SIXBIT characters for file names, PPN for UFDs.
>  E+1: Extension; 3 SIXBIT characters in LH, access date in RH.
>  E+2: File protection, modify date and time
>  E+3: Directory, one of:
>	P,,PN - file will be looked for or created in specified UFD.
>	0,,E2 - E2 is address of PATH block, which contains PPN and
>		up to 5 levels of SFD.
>	0,,0  - current working directory.

Aha. What about relative names (a/b/c)? Not that they are critical
(especially if one has getpwd() equivalent), though...

>>Was it done via creation of temporary entry for a new
>>version and "upgrading' it to the normal upon close?
>
>Yep.

Was the temporary entry written on disk or was it in-core only beast?

>>Another question being: suppose FOO.BAR is a large file and I want to change
>>10 words in the middle. How it would be done?
>
>That is a topic I was holding back until explictly asked.
>
>If a program does an OPEN and LOOKUP, and then does an ENTER on the same
>channel using the same four-word LOOKUP/ENTER/RENAME block, then the file
>would be in update mode.  Use USETI to set the input pointer to the desired
>block in the file.  Use the IN UUO to read the current block.
[snip]
>Update the copy of the block in memory.  Use USETO to set the output pointer
>back to the specified block number.  Use OUTPUT to rewrite the block.

Grrrmmm... So you could upgrade read-only channel to rw one. What happened in
case of the following sequence: process A does LOOKUP on foo; process B does
LOOKUP on foo; process A does ENTER on the same channel;?

Another thing being: was there an equivalent of buffer cache? Darn useful
when you are doing a big compile... Since you had a big lock over write
access it shouldn't give any problems with cache consistency. IOW: how many
disk reads would happen if two processes had the same file opened for read
and both requested read on the same block? OTOH I got used to situation when
CPU is much faster than any IO, so relative costs might be completely
different...

>One of the file protection codes was for append-only.  Set properly, other
>people could append to your file (until your disk quota was used up) but could
>not delete, supersede or update your file.
	Erm... I hope it does include protection from rename() too.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 02 May 99 09:01:03 GMT
Organization: UltraNet Communications, Inc.
Lines: 22
Message-ID: <7ghaeq$hnv$1@ligarius.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7gc02r$go9@weyl.math.psu.edu> <7gcoig$63f$1@shell3.ba.best.com> <7gditv$il6@weyl.math.psu.edu> <7ggsba$8ta$1@shell3.ba.best.com>
NNTP-Posting-Host: d14.dial-17.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 2 May 1999 10:46:50 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.ecrc.net!newsswitch.lcs.mit.edu!news.ultranet.com!d14

In article <7ggsba$8ta$1@shell3.ba.best.com>,
   inwap@best.com (Joe Smith) wrote:
>In article <7gditv$il6@weyl.math.psu.edu>,
>Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>Another question: how did you work with subdirectories? BTW, is there 
such
>>thing as list of UUOs with descriptions? In some accessible place, that 
is.
>
>Eventually they will be online on Eric Smith's site
>	http://www.36bit.org/dec/manual/
>but I do not know of any Monitor Calls Manual online at this time.

Alas, that is one of the wars that I lost.  As part of my "wrap
it up and turn the lights out" plan for TOPS10, I wanted all 
the manuals that we had in bits put on a tape and shipped to
our customers as a last hurrah.  I do apologize although I
did fight long and hard and have a number of scars. :-)

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 02 May 99 09:07:18 GMT
Organization: UltraNet Communications, Inc.
Lines: 78
Message-ID: <7ghaqi$hnv$2@ligarius.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu><371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net><Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7fs7mv$hii$1@ligarius.ultra.net> <7fsjbg$91o$1@zingo.tninet.se> <3723DBA5.E5B2FF0@MA.UltraNet.Com> <7g1gvb$bid$1@ligarius.ultra.net> <372B1D0A.2AB40504@MA.UltraNet.Com>
NNTP-Posting-Host: d14.dial-17.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 2 May 1999 10:53:06 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!news-feed.inet.tele.dk!bofh.vszbr.cz!newspeer.monmouth.com!news.ultranet.com!d14

In article <372B1D0A.2AB40504@MA.UltraNet.Com>,
   "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
>jmfbahciv@aol.com wrote:
>> 
>> In article <3723DBA5.E5B2FF0@MA.UltraNet.Com>,
>>    "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
>> >"Peter Hendén" wrote:
>> >> jmfbahciv@aol.com wrote:
>> >....
>> >> >The reason I didn't like to use the -20 is because ...; it was
>> >> >too slow for my taste ...
>> >>
>> >> Wasn't this dependent on whether your programs were
>> >> TOPS-10 (used PA1050) or TOPS-20 native. As I recall
>> >> (1977 - 80) we had a lot of users on our machine running
>> >> Interlisp-20 and a bunch of other native programs, and they
>> >> didn't get sluggish until long after the TOPS-10 progs did.
>> >
>> >Could be.  However, here's something to consider about -10 vs. -20 I/O
>> >performance:
>> >
>> >On the one hand, I don't know how fast PA1050-simulated buffered I/O 
went
>> >compared to PMAP.  However, I definitely recall hearing that the DUMPI 
and
>> >DUMPO JSY' that were intended for simulation of unbuffered I/O were 
*the*
>> >fastest form of I/O on Tops-20.
>> 
>> You completely missed my point.  ...
>
>Peter asserted that under a heavy system load, native programs in the mix 
had
>better responsiveness than programs which used PA1050.
>
>You asserted that a system running Tops-20 could handle fewer users than a
>system running Tops-10.

That's right.  It will always be the case.  I'm talking about
a computer system delivering service to the users in such a
way that they don't have to take numerous coffee breaks just
to get some simple work done.

>
>
>>... Just because one user used
>> something that did UUOs, should NOT have affected another user's
>> performance who is using JSYSes.  ...
>
>What should we conclude?
>
>1.  Peter's Tops-20 system didn't actually behave the way he recalls.
>	-or-
>2.  Tops-20's performance didn't actually degrade the way you would like.
>
>
>I asserted that programs that used the native Tops-20 calls intended for 
use
>by PA1050 itself ran faster than programs that used native Tops-20 calls
>intended for use by native applications (implicitly, PMAP).  (This was
>regardless of whether programs that ran on top of PA1050 were slower or 
faster
>than programs that ran right on top of Tops-20).
>
>
>Gee, would you have liked PMAP to have better performance than 
DUMPI/DUMPO?

Reread what I wrote, Alan.  I would use the -20 in the early morning
until somebody else logged on (I didn't have to know about the
person..I could feel the system degrade).  Then, about that time,
TW would give the TOPS10 systems up for timesharing and I would
move to the -10.  If the -20 system was so great, how come 
builds were taken over to the -10 (or didn't you know about that?)

/BAH

Subtract a hundred and four for e-mail.

######

From: Bill Westfield <billw@flipper.cisco.com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 02 May 1999 11:59:56 -0700
Organization: Ye 'Ol Disorganized NNTPCache groupie
Lines: 22
Message-ID: <544slv1emr.fsf@flipper.cisco.com>
References: <371be3e4.0@newsfeed.one.net> <7fs7mv$hii$1@ligarius.ultra.net>
	<7fsjbg$91o$1@zingo.tninet.se> <3723DBA5.E5B2FF0@MA.UltraNet.Com>
	<7ggt46$cv3$1@shell3.ba.best.com>
NNTP-Posting-Host: usenet.cisco.com
X-Newsreader: Gnus v5.3/Emacs 19.34
Cache-Post-Path: newsproxy.cisco.com!unknown@flipper.cisco.com
X-Cache: nntpcache 2.3.3 (see http://www.nntpcache.org/)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!tank.news.pipex.net!pipex!uunet!ams.uu.net!ffx.uu.net!in2.uu.net!news-master.cisco.com!not-for-mail

    Changing the TOPS-10 program to detect when it was running under TOPS-20
    and use DUMPI/DUMPO made it faster.  And using PMAP to map file pages
    into user address space was even faster.

Using PMAP always bothered me a bit on some conceptual plane.  It seemed
that the monitor could/would do some quite sensible "read-ahead"
behavior if a file was opened for sequential IO, and of course you could
do this yourself with async IO under tops10 (IIRC.)  However, using
PMAP, it seemed that you lost the ability for such lookahead.  If you
didn't touch a page, it may or may not have been in core, and when you
did touch it you'd have to wait for the paging code to read it from disk
(probably.)  I didn't see any sort of "start getting me this page, and
I'll do the actual accesss in a bit, so don't block now." (You could set
up a whole file to be prefetched, I think, but the calls didn't seem
applicable to the scenario I describe here.)

Yes, no?

BillW

-- 
(remove spam food from return address)

######

From: "Alan H. Martin" <AMartin@MA.UltraNet.Com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 02 May 1999 13:23:58 -0400
Organization: UltraNet Communications , an RCN Company http://www.ultranet.com/
Lines: 18
Message-ID: <372C8A2E.F0CC6590@MA.UltraNet.Com>
References: <371be3e4.0@newsfeed.one.net> <7fs7mv$hii$1@ligarius.ultra.net> <7fsjbg$91o$1@zingo.tninet.se> <3723DBA5.E5B2FF0@MA.UltraNet.Com> <7ggt46$cv3$1@shell3.ba.best.com>
NNTP-Posting-Host: 207-172-216-241.s241.tnt1.sbo.ma.dialup.rcn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 2 May 1999 17:24:36 GMT
X-Mailer: Mozilla 4.5 [en] (Win95; U)
X-Accept-Language: en,en-US,en-GB,es
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!newsfeed.mathworks.com!newsswitch.lcs.mit.edu!news.ultranet.com!not-for-mail

Joe Smith wrote:

> One of the programmers in Colorado noticed that a particular file copy
> program, highly optimized for TOPS-10 (doing unbuffered I/O in large chunks),
> was quicker than the standard TOPS-20 programs for copying files on TOPS-20.
> 
> In spite of the overhead of the TOPS-10 program running under PA1050, it
> was more efficient than the TOPS-20 native programs using SIN and SOUT.
> 
> Changing the TOPS-10 program to detect when it was running under TOPS-20
> and use DUMPI/DUMPO made it faster.  And using PMAP to map file pages
> into user address space was even faster.

Interesting; that could have been the deal.  It makes sens that SIN/SOUT was
slower than other approaches, since it required an extra copy of the data.
				/AHM
-- 
Alan Howard Martin				AMartin@MA.UltraNet.Com

######

From: "Alan H. Martin" <AMartin@MA.UltraNet.Com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 02 May 1999 14:14:21 -0400
Organization: UltraNet Communications , an RCN Company http://www.ultranet.com/
Lines: 35
Message-ID: <372C95FD.8F899038@MA.UltraNet.Com>
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu><371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net><Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7fs7mv$hii$1@ligarius.ultra.net> <7fsjbg$91o$1@zingo.tninet.se> <3723DBA5.E5B2FF0@MA.UltraNet.Com> <7g1gvb$bid$1@ligarius.ultra.net> <372B1D0A.2AB40504@MA.UltraNet.Com> <7ghaqi$hnv$2@ligarius.ultra.net>
NNTP-Posting-Host: 207-172-216-241.s241.tnt1.sbo.ma.dialup.rcn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 2 May 1999 18:15:01 GMT
X-Mailer: Mozilla 4.5 [en] (Win95; U)
X-Accept-Language: en,en-US,en-GB,es
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!not-for-mail

jmfbahciv@aol.com wrote:
> 
> In article <372B1D0A.2AB40504@MA.UltraNet.Com>,
>    "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
> >jmfbahciv@aol.com wrote:
> >>
> >> In article <3723DBA5.E5B2FF0@MA.UltraNet.Com>,
> >>    "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
> >> >"Peter Hendén" wrote:
> >> >> jmfbahciv@aol.com wrote:
> >> >....
> >> >> >The reason I didn't like to use the -20 is because ...; it was
> >> >> >too slow for my taste ...
> >> >>
> >> >> Wasn't this dependent on whether your programs were
> >> >> TOPS-10 (used PA1050) or TOPS-20 native. As I recall
> >> >> (1977 - 80) we had a lot of users on our machine running
> >> >> Interlisp-20 and a bunch of other native programs, and they
> >> >> didn't get sluggish until long after the TOPS-10 progs did.
...
> >>... Just because one user used
> >> something that did UUOs, should NOT have affected another user's
> >> performance who is using JSYSes.  ...
...
> Reread what I wrote, Alan.  I would use the -20 in the early morning
> until somebody else logged on (I didn't have to know about the
> person..I could feel the system degrade).  Then, about that time,
> TW would give the TOPS10 systems up for timesharing and I would
> move to the -10.  If the -20 system was so great, how come
> builds were taken over to the -10 (or didn't you know about that?)

Could you sense whether the person was running native or compatibility apps?
				/AHM
-- 
Alan Howard Martin				AMartin@MA.UltraNet.Com

######

From: "Alan H. Martin" <AMartin@MA.UltraNet.Com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 02 May 1999 14:14:21 -0400
Organization: UltraNet Communications , an RCN Company http://www.ultranet.com/
Lines: 35
Message-ID: <372C95FD.8F899038@MA.UltraNet.Com>
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu><371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net><Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7fs7mv$hii$1@ligarius.ultra.net> <7fsjbg$91o$1@zingo.tninet.se> <3723DBA5.E5B2FF0@MA.UltraNet.Com> <7g1gvb$bid$1@ligarius.ultra.net> <372B1D0A.2AB40504@MA.UltraNet.Com> <7ghaqi$hnv$2@ligarius.ultra.net>
NNTP-Posting-Host: 207-172-216-241.s241.tnt1.sbo.ma.dialup.rcn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 2 May 1999 18:15:01 GMT
X-Mailer: Mozilla 4.5 [en] (Win95; U)
X-Accept-Language: en,en-US,en-GB,es
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!not-for-mail

jmfbahciv@aol.com wrote:
> 
> In article <372B1D0A.2AB40504@MA.UltraNet.Com>,
>    "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
> >jmfbahciv@aol.com wrote:
> >>
> >> In article <3723DBA5.E5B2FF0@MA.UltraNet.Com>,
> >>    "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
> >> >"Peter Hendén" wrote:
> >> >> jmfbahciv@aol.com wrote:
> >> >....
> >> >> >The reason I didn't like to use the -20 is because ...; it was
> >> >> >too slow for my taste ...
> >> >>
> >> >> Wasn't this dependent on whether your programs were
> >> >> TOPS-10 (used PA1050) or TOPS-20 native. As I recall
> >> >> (1977 - 80) we had a lot of users on our machine running
> >> >> Interlisp-20 and a bunch of other native programs, and they
> >> >> didn't get sluggish until long after the TOPS-10 progs did.
...
> >>... Just because one user used
> >> something that did UUOs, should NOT have affected another user's
> >> performance who is using JSYSes.  ...
...
> Reread what I wrote, Alan.  I would use the -20 in the early morning
> until somebody else logged on (I didn't have to know about the
> person..I could feel the system degrade).  Then, about that time,
> TW would give the TOPS10 systems up for timesharing and I would
> move to the -10.  If the -20 system was so great, how come
> builds were taken over to the -10 (or didn't you know about that?)

Could you sense whether the person was running native or compatibility apps?
				/AHM
-- 
Alan Howard Martin				AMartin@MA.UltraNet.Com

######

From: "Alan H. Martin" <AMartin@MA.UltraNet.Com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 02 May 1999 14:38:52 -0400
Organization: UltraNet Communications , an RCN Company http://www.ultranet.com/
Lines: 15
Message-ID: <372C9BBC.FDC22D8D@MA.UltraNet.Com>
References: <371be3e4.0@newsfeed.one.net> <7g8sbh$e88@weyl.math.psu.edu> <7gb4n6$795$1@shell3.ba.best.com> <7gc02r$go9@weyl.math.psu.edu> <7gcoig$63f$1@shell3.ba.best.com>
NNTP-Posting-Host: 207-172-216-241.s241.tnt1.sbo.ma.dialup.rcn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 2 May 1999 18:39:31 GMT
X-Mailer: Mozilla 4.5 [en] (Win95; U)
X-Accept-Language: en,en-US,en-GB,es
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.ecrc.net!newsswitch.lcs.mit.edu!news.ultranet.com!not-for-mail

Joe Smith wrote:

> The down side was that, in general, it was not possible to watch the progress
> of a slowly growing file.  Programs that wanted to make their log files
> visible in real time would need to open+append+close for each block written.

(Unless you watched the disk's free blocks decrease (and knew no one else was
writing).

I accidentally discovered in high school that the disk space consumed by the
prime and spare RIBs, and perhaps the first data block, were debited upon
ENTER, not OUT (let alone CLOSE).  It made sense to me in retrospect, though.
				/AHM
-- 
Alan Howard Martin				AMartin@MA.UltraNet.Com

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
From: lisard@zetnet.co.uk
X-Everything: Net-Tamer V 1.08X
Date: 02 May 1999 14:41:29 GMT
Lines: 54
Message-ID: <372c6419$0$496@news.zetnet.co.uk>
NNTP-Posting-Host: man-052.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 925656089 496 194.247.41.64
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newspeer.clara.net!news.clara.net!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail



On 1999-05-01 dpeschel@u.washington.edu(DerekPeschel) said:
   :I've always wanted to know more about BCPL.  Specifically, I've
   :heard about the "global vector".  Is this related to the dispatch
   :table you talked about?  Or was that only an OS6 creation?

No, the dispatch table was an OS6 thing. OS6 had a concept called
streams, which were basically a precursor of the Unix-like streams, and
every stream had a number of parameters with it. Included in those
parameters were pointers to a number of function calls, for reading,
writing, etc. etc. There's a paper on it in one of the journals,
somewhere, and if someone has a webbed version of it or can exhume one,
I for one would be a very happy pussycat.

The global vector was what BCPL had to avoid using a linker... but
before explaining that, I should probably go into BCPL's semantics
first. BCPL has one datatype, the word, and one namespace, which
contains a set of associations between names and variables (except for
constants, which were different). So if you declared a variable, you got
a word-sized thing whose address was associated with a name; but if you
declared a vector (always one-dimensional) you got a word-sized thing
containing the address of that vector, which is the number of words you
requested, plus one.

The same applies to functions. You can declare a function, and your
declaration names not a constant, but a variable in memory storing the
address of that function. One of the things this means is that you can
suddenly change the function a particular name calls. Another is that
you can have arrays of function calls.

And this is what the global vector is; it's a long vector in which each
element is associated with a name. You make the association yourself, in
a header file somewhere, and the onus is on you to avoid scribbling over
other functions. It's an interesting way of avoiding the requirement to
have a linker. From reading about Amiga Exec, it seems that the address
of the Amiga's global vector is held at absolute address 4.

   :I've also wondered if BCPL's run-time overhead (i.e., the global
   :vector) is too heavy.  Amiga owners have complained about the
   :overhead.  I would think C is better, but perhaps the C people have
   :simply brainwashed me.  What do you think?  Is it good to try to
   :create a "no-overhead" language?

The global vector doesn't in itself add overhead; the overhead comes
because *every* function call is indirected through a pointer, and also
because you have to scale pointers. BCPL's other quirk is that a
word-addressed machine is an inbuilt assumption. The earlier 68Ks can't
scale addresses in the way that the 386 (for example) can, which means
that every memory addressing operation needs a left shift by 2 as well.
Now that's overhead. :> However, I think a 386 would be a near-perfect
BCPL machine, if someone fancies porting AmigaDOS...?
--
Communa (lisard@zetnet.co.uk) -- this film's crap, let's slash the seats

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
From: lisard@zetnet.co.uk
X-Everything: Net-Tamer V 1.08X
Date: 02 May 1999 14:41:29 GMT
Lines: 54
Message-ID: <372c6419$0$496@news.zetnet.co.uk>
NNTP-Posting-Host: man-052.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 925656089 496 194.247.41.64
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newspeer.clara.net!news.clara.net!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail



On 1999-05-01 dpeschel@u.washington.edu(DerekPeschel) said:
   :I've always wanted to know more about BCPL.  Specifically, I've
   :heard about the "global vector".  Is this related to the dispatch
   :table you talked about?  Or was that only an OS6 creation?

No, the dispatch table was an OS6 thing. OS6 had a concept called
streams, which were basically a precursor of the Unix-like streams, and
every stream had a number of parameters with it. Included in those
parameters were pointers to a number of function calls, for reading,
writing, etc. etc. There's a paper on it in one of the journals,
somewhere, and if someone has a webbed version of it or can exhume one,
I for one would be a very happy pussycat.

The global vector was what BCPL had to avoid using a linker... but
before explaining that, I should probably go into BCPL's semantics
first. BCPL has one datatype, the word, and one namespace, which
contains a set of associations between names and variables (except for
constants, which were different). So if you declared a variable, you got
a word-sized thing whose address was associated with a name; but if you
declared a vector (always one-dimensional) you got a word-sized thing
containing the address of that vector, which is the number of words you
requested, plus one.

The same applies to functions. You can declare a function, and your
declaration names not a constant, but a variable in memory storing the
address of that function. One of the things this means is that you can
suddenly change the function a particular name calls. Another is that
you can have arrays of function calls.

And this is what the global vector is; it's a long vector in which each
element is associated with a name. You make the association yourself, in
a header file somewhere, and the onus is on you to avoid scribbling over
other functions. It's an interesting way of avoiding the requirement to
have a linker. From reading about Amiga Exec, it seems that the address
of the Amiga's global vector is held at absolute address 4.

   :I've also wondered if BCPL's run-time overhead (i.e., the global
   :vector) is too heavy.  Amiga owners have complained about the
   :overhead.  I would think C is better, but perhaps the C people have
   :simply brainwashed me.  What do you think?  Is it good to try to
   :create a "no-overhead" language?

The global vector doesn't in itself add overhead; the overhead comes
because *every* function call is indirected through a pointer, and also
because you have to scale pointers. BCPL's other quirk is that a
word-addressed machine is an inbuilt assumption. The earlier 68Ks can't
scale addresses in the way that the 386 (for example) can, which means
that every memory addressing operation needs a left shift by 2 as well.
Now that's overhead. :> However, I think a 386 would be a near-perfect
BCPL machine, if someone fancies porting AmigaDOS...?
--
Communa (lisard@zetnet.co.uk) -- this film's crap, let's slash the seats

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 2 May 1999 16:01:28 -0700
Organization: Chez Inwap
Message-ID: <7gilg8$pvp$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gfg5m$jen@weyl.math.psu.edu> <7ggqll$u0$1@shell3.ba.best.com> <7gh5mj$kb3@weyl.math.psu.edu>
Lines: 110
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925686093 201 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!uunet!ams.uu.net!nyc.uu.net!newsfeed.mathworks.com!newsfeed1.earthlink.net!newsfeed.berkeley.edu!news2.best.com!news1.best.com!news.idiom.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gh5mj$kb3@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>Aha. What about relative names (a/b/c)?

The concept did not exist, unfortunatly.

>Was the temporary entry written on disk or was it in-core only beast?

It was written to disk.  The bits in the SAT (Storate Allocation Table)
corresponding to the block containing the RIB and to the data blocks written
thus far were marked as being in use.  The Monitor was very paranoid at
making sure the disk structure was not damaged if a crash occurred at any
time.  To create a new file:

  *) Allocate blocks by find free bits in the SAT.  Mark 16 consecutive
     blocks in use for the RIB and the first 15 data blocks.

  *) Write the modified SAT to disk.  Wait for disk I/O to complete.

  *) Mark the RIB has having 16 blocks; itself and the contiguous
     data blocks.  Set the file's size to be 0 words.  Set the file's
     allocated size to 16 blocks.  Write the RIB to disk, wait for I/O.

  *) When the program does an OUT UUO, update the RIB and write the user's
     data to disk.  Do not return to the user program until I/O is done.

If the system crashed before the file was closed, there would not be any
pointer in any directory for the RIB.  The RIB and its data blocks would
become "lost blocks", and not reclaimed until CHKDSK was run with the
system in single-user mode.

>Grrrmmm... So you could upgrade read-only channel to rw one. What happened in
>case of the following sequence: process A does LOOKUP on foo; process B does
>LOOKUP on foo; process A does ENTER on the same channel;?

Process A has update/write access.  Process B still has read access.
If process A modifies blocks that process B has not yet read, then B will
see the modified contents when it gets around to reading them.

>Another thing being: was there an equivalent of buffer cache?

No.  TOPS-10 definitely had many ad-hoc features.  It's basic design
philosophy was rooted back in the days when it could run in 64 K words.

The KA-10 I grew up on had 192 K words (3 boxes of 64 K by 36-bit core
stacks).  By the time TOPS-10 version 6.01 came around, the Monitor took up
about 92 K words.  The remaining 100 K words were shared by all the
timesharing users.  Since programs had to be completely resident in physical
memory in order to run, CORMAX was set to 40 K.  This allowed up to two
programs at 40 K words each to fit into memory with enough core left over for
several small jobs.  That small system could handle 30 timesharing users in
100 K.  This was back before GUIs, back before curses even.  Editing was
done on hardcopy terminals (LA36 at 300 baud) or glass TTYs (ADM3A
running at 1200 baud) - a vast improvement over Teletypes (at 110 baud)
or punched cards.

>IOW: how many disk reads would happen if two processes had the same file
>opened for read and both requested read on the same block?

Each one went to the disk for its data.

TYMCOM-X, a highly modified descendent of TOPS-10 that Tymshare ran on its
PDP-10s, was page oriented.  It simulated TOPS-10 I/O by mapping file
pages into kernel address space and BLTing data to or from user space.
TYMCOM-X (like TOPS-20) did what you are talking about; the second process
would get shared access to the physical page holding the file data, or it
would get a copy of the data already in core.

TYMCOM-X also elminated the idea of dedicated swap space.  Any unused
page on the disk could be used as backing store for pages in user address
space.

>>people could append to your file (until your disk quota was used up) but
>>could not delete, supersede or update your file.
>	Erm... I hope it does include protection from rename() too.

Permissions for directories were like many other operating systems;
three independent bits for read access (get a list of all files in
the directory), write access (create new files in the directory) and
transit the directory (to read or update an existing file therein).

Permissions for files, however, were numbers from 0 to 7.

    7	Greatest protection, which means no access privileges.  However,
	the owner may LOOKUP the file so that (s)he can rename the
	protection to a less restrictive code via RENAME.

    6	Execute-only.  This disables user meddling and examining
	(DUMP, DCORE, D, E, SAVE, SSAVE, START n, CSTART n,
	DDT, CORE n) with the error message ?ILLEGAL WHEN EXECUTE ONLY.

    5	Append, read, execute.

    4	Read, execute.

    3	Update, append, read, execute.

    2	Write, update, append, read, execute.

    1	Rename, write, update, append, read, execute.

    0	Change protection, rename, write, update, append, read, execute.

Later on there was FILDAE, the file daemon that allowed for user customized
access control lists, but that is a completely different story.
	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 2 May 1999 16:24:58 -0400
Organization: -ENOENT
Lines: 13
Message-ID: <7gicaq$ksn@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7g8sbh$e88@weyl.math.psu.edu> <7gb4n6$795$1@shell3.ba.best.com> <372CA8A3.C1DE2BBD@MA.UltraNet.Com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!howland.erols.net!math.ohio-state.edu!news.cis.ohio-state.edu!news.ems.psu.edu!news3.cac.psu.edu!not-for-mail

In article <372CA8A3.C1DE2BBD@MA.UltraNet.Com>,
Alan H. Martin <AMartin@MA.UltraNet.Com> wrote:
>BTW, I'd love to see the official explanation of how /dev/tty is implemented
>by Unix.  It has always struck me as the only logical name on Unix.

	If you are asking about the way it means different devices for different
processes - trivial. When you open a device with that major:minor pair driver
takes major:minor of your controlling TTY and goes ahead. BTW, take a look at
/dev/fd/0, /dev/fd/1, etc. - /dev/tty is not the only specimen.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 2 May 1999 16:37:57 -0700
Organization: Chez Inwap
Message-ID: <7ginkl$9of$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gc02r$go9@weyl.math.psu.edu> <7gcoig$63f$1@shell3.ba.best.com> <372C9BBC.FDC22D8D@MA.UltraNet.Com>
Lines: 53
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925688282 209 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <372C9BBC.FDC22D8D@MA.UltraNet.Com>,
Alan H. Martin <AMartin@MA.UltraNet.Com> wrote:
>Joe Smith wrote:
>
>> The down side was that, in general, it was not possible to watch the progress
>> of a slowly growing file.  Programs that wanted to make their log files
>> visible in real time would need to open+append+close for each block written.
>
>(Unless you watched the disk's free blocks decrease (and knew no one else was
>writing).
>
>I accidentally discovered in high school that the disk space consumed by the
>prime and spare RIBs, and perhaps the first data block, were debited upon
>ENTER, not OUT (let alone CLOSE).  It made sense to me in retrospect, though.
>				/AHM

When a file grew, it did so by one or more clusters instead of a block at a
time.  One of the file system parameters was the number of clusters to
allocate on initial creation.

In addition to the four-word LOOKUP/ENTER/RENAME block, TOPS-10 had an
extended LOOKUP/ENTER/RENAME block, signified by a zero in the left half
of the first word (where the file name usually goes) and a number greater
than 3 in the right half.  Word 10 of the extended block is the estimated
size of the file.  If nonzero on an ENTER, the Monitor would attempt to
allocate that many contiguous data blocks for the file.  A file that had
become fragmented could be made contiguous by simply copying it using a
program that set .RBEST in the extend ENTER block.  (BACKUP, the program
that dumped files to tape, did this.)

If a file outgrew its initial allocation of blocks, another cluster of
blocks would be allocated.  The SATs actually kept track of clusters, not
individual blocks.  A cluster size of 5 blocks was typical, although small
sites tended to define DSKB with a cluster size of 3.  Three was the
practical minimum since a one-word file needed 3 blocks: 1 for the RIB,
1 for the data, 1 for the redundant copy of the RIB.

The RIB consisted of three types of entries:
1) Change of unit - specifies the disk unit for succeeding group pointers.
2) Group pointer - starting cluster number and repeat count.
3) End of file = word of zero.

A disk structure consisted of 1 to 64 disk units.  The logical size of
the disk structure was the concatination of the disk units, which could
be of varying size.  It meant that the disk structure could be much larger
than a single IBM 3330-I, but a head crash on one disk pack would wipe
out the entire structure.  Some of the hassles of this form of disk
structure were the impetus behind the creation of RAID.
	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 2 May 1999 17:10:43 -0700
Organization: Chez Inwap
Message-ID: <7gipi3$sn6$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7g8sbh$e88@weyl.math.psu.edu> <7gb4n6$795$1@shell3.ba.best.com> <372CA8A3.C1DE2BBD@MA.UltraNet.Com>
Lines: 40
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925690247 218 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feeder.qis.net!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

>>...  There was no concept of
>> STDIN or STDOUT.  ...
>
>Sure there was - every (attached) job had a device TTY: which referred to the
>job's controlling terminal.  Reassign TTY: to point elsewhere, and that's
>where the primary I/O went.

Not true.  Assigning TTY to DSK had no affect on the TTCALLs.  For many
programs, the primary input was via INCHWL and the primary output was via
OUTCHR and OUTSTR.  Those programs did I/O to the controlling terminal
and could not be redirected.  That's why I say there was no concept to
STDIN or STDOUT - no pre-opened I/O channels ala Unix and TOPS-20.

>>... Programs had two ways of doing I/O to the terminal: it could
>> do an OPEN on device 'TTY' and use normal IN, INPUT, OUT, or OUTPUT UUOs, or
>> it do TTCALLs.  ...  If you wanted to
>> turn off echo, or to be able to read all possible characters (including
>> Control-C and/or parity), you had to use the OPEN call.  ...
>
>Nope, you may have enjoyed turning off echo by setting IO.SUP on an opened
>TTY's device status, but I always did it by setting GL.NEC in the line
>characteristics with SETLCH TTCALL.  Other people might have preferred using
>the .TOLCP function of the TRMOP. terminal operation UUO.
>				/AHM

Yeah, but if your program bombed out with ?ILLEGAL MEMORY REFERENCE, the poor
user would be stuck at the Monitor prompt with echo disabled.  This behaviour
was distressing to users that did not know that they had to type "SET TTY
ECHO" to see what they were typing.  SETLCH and TRMOP. were two ways of
semipermenantly changing the terminal's line characteristics.

The temporary echo suppression via the OPEN call was automatically cleared
when the job returned to Monitor mode.  This was a great feature not
duplicated elsewhere.
	-Joe

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 2 May 1999 17:18:01 -0700
Organization: Chez Inwap
Message-ID: <7gipvp$2l8$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gdk1j$ilu@weyl.math.psu.edu> <7gfnpc$3in$1@shell3.ba.best.com> <RLDFmCAC+NL3EwuU@museltd.demon.co.uk>
Lines: 19
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925690687 210 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!newsfeed.stanford.edu!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <RLDFmCAC+NL3EwuU@museltd.demon.co.uk>,
Roy Trubshaw  <Roy@museltd.demon.co.uk> wrote:
>5 levels explicitly, our favourite way of exceeding disk quota was to
>"bury" the offending files in SFDs greater than 5 levels deep! (You have
>to rename the SFDs into another SFD, thus "pushing" the contents of the
>first SFD one level deeper.

Yep.  After 6 levels, LOGOUT couldn't find your files when recalculating
your disk quota.  But neither could BACKUP.  So, if you were at a school
that did a periodic refresh and restore (to defragment the disk), then
all of your hidden files would disappear.

The cure to this was a change in the Monitor which disallowed a RENAME
of an SFD to a level deeper than it currently was.
	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: "Alan H. Martin" <AMartin@MA.UltraNet.Com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 02 May 1999 18:19:58 -0400
Organization: UltraNet Communications , an RCN Company http://www.ultranet.com/
Lines: 19
Message-ID: <372CCF8E.1530FDB@MA.UltraNet.Com>
References: <371be3e4.0@newsfeed.one.net> <7fs7mv$hii$1@ligarius.ultra.net>
		<7fsjbg$91o$1@zingo.tninet.se> <3723DBA5.E5B2FF0@MA.UltraNet.Com>
		<7ggt46$cv3$1@shell3.ba.best.com> <544slv1emr.fsf@flipper.cisco.com>
NNTP-Posting-Host: 209-122-232-243.s497.tnt4.sbo.ma.dialup.rcn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 2 May 1999 22:20:44 GMT
X-Mailer: Mozilla 4.51 [en] (Win95; U)
X-Accept-Language: en,en-US,en-GB,es
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!masternews.telia.net!feed2.news.luth.se!luth.se!cam-news-hub1.bbnplanet.com!news.gtei.net!newsfeed.mathworks.com!newsswitch.lcs.mit.edu!news.ultranet.com!not-for-mail

Bill Westfield wrote:

> Using PMAP always bothered me a bit on some conceptual plane.  It seemed
> that the monitor could/would do some quite sensible "read-ahead"
> behavior if a file was opened for sequential IO, and of course you could
> do this yourself with async IO under tops10 (IIRC.)  However, using
> PMAP, it seemed that you lost the ability for such lookahead.  If you
> didn't touch a page, it may or may not have been in core, and when you
> did touch it you'd have to wait for the paging code to read it from disk
> (probably.)  I didn't see any sort of "start getting me this page, and
> I'll do the actual accesss in a bit, so don't block now." (You could set
> up a whole file to be prefetched, I think, but the calls didn't seem
> applicable to the scenario I describe here.)

Yeah.  VMS has page fault clusters, but I never heard of that, or any prefetch
calls/modes on Tops-20.  We should hear the answer shortly here...
				/AHM
-- 
Alan Howard Martin				AMartin@MA.UltraNet.Com

######

From: "Alan H. Martin" <AMartin@MA.UltraNet.Com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 02 May 1999 18:25:05 -0400
Organization: UltraNet Communications , an RCN Company http://www.ultranet.com/
Lines: 22
Message-ID: <372CD0C1.9027E17@MA.UltraNet.Com>
References: <371be3e4.0@newsfeed.one.net> <7g8sbh$e88@weyl.math.psu.edu> <7gb4n6$795$1@shell3.ba.best.com> <372CA8A3.C1DE2BBD@MA.UltraNet.Com> <7gicaq$ksn@weyl.math.psu.edu>
NNTP-Posting-Host: 209-122-232-243.s497.tnt4.sbo.ma.dialup.rcn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 2 May 1999 22:25:50 GMT
X-Mailer: Mozilla 4.51 [en] (Win95; U)
X-Accept-Language: en,en-US,en-GB,es
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!uio.no!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!not-for-mail

Alexander Viro wrote:
> 
> In article <372CA8A3.C1DE2BBD@MA.UltraNet.Com>,
> Alan H. Martin <AMartin@MA.UltraNet.Com> wrote:
> >BTW, I'd love to see the official explanation of how /dev/tty is implemented
> >by Unix.  It has always struck me as the only logical name on Unix.
> 
>         If you are asking about the way it means different devices for different
> processes - trivial. When you open a device with that major:minor pair driver
> takes major:minor of your controlling TTY and goes ahead.  ...

Well, how and where.  Thanks.


>... BTW, take a look at
> /dev/fd/0, /dev/fd/1, etc. - /dev/tty is not the only specimen.

And some systems have /dev/stdin and /dev/stdout, right?  I suppose they could
just be links to /dev/fd/{0,1}.
				/AHM
-- 
Alan Howard Martin				AMartin@MA.UltraNet.Com

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 2 May 1999 19:00:57 -0400
Organization: -ENOENT
Lines: 17
Message-ID: <7gilf9$l5k@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <372CA8A3.C1DE2BBD@MA.UltraNet.Com> <7gicaq$ksn@weyl.math.psu.edu> <372CD0C1.9027E17@MA.UltraNet.Com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news-fra.pop.de!newsfeed.ecrc.net!newspeer.monmouth.com!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <372CD0C1.9027E17@MA.UltraNet.Com>,
Alan H. Martin <AMartin@MA.UltraNet.Com> wrote:
>Alexander Viro wrote:
>>... BTW, take a look at
>> /dev/fd/0, /dev/fd/1, etc. - /dev/tty is not the only specimen.
>
>And some systems have /dev/stdin and /dev/stdout, right?  I suppose they could
>just be links to /dev/fd/{0,1}.
	Yup. Notice that /dev/tty is much more essential. Controlling TTY is
shared by the whole process group and nothing forces a member of said group to
have it opened. /dev/tty provides an ability to reopen the controlling TTY.
OTOH opening /dev/fd/0 is pretty much equivalent to dup(0), i.e. you already
have that stream opened.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 2 May 1999 20:56:34 -0700
Organization: Chez Inwap
Message-ID: <7gj6pi$12g$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gh5mj$kb3@weyl.math.psu.edu> <7gilg8$pvp$1@shell3.ba.best.com> <372CFA63.DFDEE636@MA.UltraNet.Com>
Lines: 34
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925703798 198 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <372CFA63.DFDEE636@MA.UltraNet.Com>,
Alan H. Martin <AMartin@MA.UltraNet.Com> wrote:
>Joe Smith wrote:
>
>>     5   Append, read, execute.
>> 
>>     4   Read, execute.
>
>(Twiddle).

You're absolutely correct, I made a typo.

I can take this opportunity to mention that TYMCOM-X changed the definition
of the permission modes slightly.  The difference between 0 and 1 on TOPS-10
was of hardly any use, so Tymshare shifted everything down one and defined
a new code 7.

    7	No access.  Can't even do a LOOKUP to read file's attributes.
    6	Attributes only.  LOOKUP returns file's size, etc.  Cannot read.
    5   Execute only.
    4   Read and execute.
    3   Append, read, execute
    2   Update, append, read, execute
    1   Write, update, append, read, execute
    0   Rename, write, update, append, read, execute.

One nice thing about the TYMCOM-X definition is that if the high order
bit is set, the file is read-only.
	-Joe

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: "Alan H. Martin" <AMartin@MA.UltraNet.Com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 02 May 1999 21:09:34 -0400
Organization: UltraNet Communications , an RCN Company http://www.ultranet.com/
Lines: 19
Message-ID: <372CF74E.BFEFD131@MA.UltraNet.Com>
References: <371be3e4.0@newsfeed.one.net> <7gc02r$go9@weyl.math.psu.edu> <7gcoig$63f$1@shell3.ba.best.com> <372C9BBC.FDC22D8D@MA.UltraNet.Com> <7ginkl$9of$1@shell3.ba.best.com>
NNTP-Posting-Host: 209-122-232-243.s497.tnt4.sbo.ma.dialup.rcn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 3 May 1999 01:10:19 GMT
X-Mailer: Mozilla 4.51 [en] (Win95; U)
X-Accept-Language: en,en-US,en-GB,es
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!not-for-mail

Joe Smith wrote:
> 
> In article <372C9BBC.FDC22D8D@MA.UltraNet.Com>,
> Alan H. Martin <AMartin@MA.UltraNet.Com> wrote:
> >Joe Smith wrote:
...
> If a file outgrew its initial allocation of blocks, another cluster of
> blocks would be allocated.  The SATs actually kept track of clusters, not
> individual blocks.  A cluster size of 5 blocks was typical, although small
> sites tended to define DSKB with a cluster size of 3.  Three was the
> practical minimum since a one-word file needed 3 blocks: 1 for the RIB,
> 1 for the data, 1 for the redundant copy of the RIB.

Yep, but I never observed a disk with a cluster size larger than 3 in high
school or college.  At MR, I don't recall how things were sized; we developed
Fortran on the -20's by then.
				/AHM
-- 
Alan Howard Martin				AMartin@MA.UltraNet.Com

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 2 May 1999 21:12:29 -0700
Organization: Chez Inwap
Message-ID: <7gj7nd$c1v$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7g8sbh$e88@weyl.math.psu.edu> <7gb4n6$795$1@shell3.ba.best.com> <372CA8A3.C1DE2BBD@MA.UltraNet.Com>
Lines: 38
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925704755 200 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newspump.monmouth.com!newspeer.monmouth.com!newsfeed.enteract.com!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <372CA8A3.C1DE2BBD@MA.UltraNet.Com>,
Alan H. Martin <AMartin@MA.UltraNet.Com> wrote:
>Joe Smith wrote:
>>...  There was no concept of
>> STDIN or STDOUT.  ...
>
>Sure there was - every (attached) job had a device TTY: which referred to the
>job's controlling terminal.  Reassign TTY: to point elsewhere, and that's
>where the primary I/O went.

Here's a failure mode that I reported (and got a patch for):
	.ASSIGN DSK TTY
	.R BASIC
	^C
	^C

BASIC opened logical device TTY for input and output.  It then began to
immediatly use IN and OUT UUOs without doing a LOOKUP or ENTER, since the
program always talks to TTY:, right?

The IN fails because the unprivileged program had done an OPEN and IN without
an intervening LOOKUP.  It tried to output "I/O error reading from device TTY"
to device TTY.  But because it is doing an OUT without an intervening ENTER,
that UUO fails.  The output routine goes into an infinite loop trying to
output error messages.  In the mean time, the Control-C intercept routine
was doing its job, preventing Control-C from stopping execution.

The patch was to exit quietly if TTY was not assigned to a terminal device.

Given that STDOUT is the concept of "output that usually goes to the
terminal could be redirected to any arbitrary file instead", most DEC
CUSPs did not understand the concept of STDOUT.
	-Joe

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: "Alan H. Martin" <AMartin@MA.UltraNet.Com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 02 May 1999 21:20:11 -0400
Organization: UltraNet Communications , an RCN Company http://www.ultranet.com/
Lines: 30
Message-ID: <372CF9CA.272C4577@MA.UltraNet.Com>
References: <371be3e4.0@newsfeed.one.net> <7g8sbh$e88@weyl.math.psu.edu> <7gb4n6$795$1@shell3.ba.best.com> <372CA8A3.C1DE2BBD@MA.UltraNet.Com> <7gipi3$sn6$1@shell3.ba.best.com>
NNTP-Posting-Host: 209-122-232-243.s497.tnt4.sbo.ma.dialup.rcn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 3 May 1999 01:20:57 GMT
X-Mailer: Mozilla 4.51 [en] (Win95; U)
X-Accept-Language: en,en-US,en-GB,es
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!not-for-mail

Joe Smith wrote:
> 
> >>...  There was no concept of
> >> STDIN or STDOUT.  ...
> >
> >Sure there was - every (attached) job had a device TTY: which referred to the
> >job's controlling terminal.  Reassign TTY: to point elsewhere, and that's
> >where the primary I/O went.
> 
> Not true.  Assigning TTY to DSK had no affect on the TTCALLs.  ...

Never said it did.  To a zeroth approximation, TTCALL goes where /dev/tty does
on Unix, and where stderr does starts out by default.  Of course, people
sometimes redirect stderr too, which just isn't an option for TTCALL.


> Yeah, but if your program bombed out with ?ILLEGAL MEMORY REFERENCE, the poor
> user would be stuck at the Monitor prompt with echo disabled.  This behaviour
> was distressing to users that did not know that they had to type "SET TTY
> ECHO" to see what they were typing.  SETLCH and TRMOP. were two ways of
> semipermenantly changing the terminal's line characteristics.
> 
> The temporary echo suppression via the OPEN call was automatically cleared
> when the job returned to Monitor mode.  This was a great feature not
> duplicated elsewhere.

Where's your sense of adventure?
				/AHM
-- 
Alan Howard Martin				AMartin@MA.UltraNet.Com

######

From: "Alan H. Martin" <AMartin@MA.UltraNet.Com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 02 May 1999 21:22:43 -0400
Organization: UltraNet Communications , an RCN Company http://www.ultranet.com/
Lines: 10
Message-ID: <372CFA63.DFDEE636@MA.UltraNet.Com>
References: <371be3e4.0@newsfeed.one.net> <7gfg5m$jen@weyl.math.psu.edu> <7ggqll$u0$1@shell3.ba.best.com> <7gh5mj$kb3@weyl.math.psu.edu> <7gilg8$pvp$1@shell3.ba.best.com>
NNTP-Posting-Host: 209-122-232-243.s497.tnt4.sbo.ma.dialup.rcn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 3 May 1999 01:23:28 GMT
X-Mailer: Mozilla 4.51 [en] (Win95; U)
X-Accept-Language: en,en-US,en-GB,es
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!not-for-mail

Joe Smith wrote:

>     5   Append, read, execute.
> 
>     4   Read, execute.

(Twiddle).
				/AHM
-- 
Alan Howard Martin				AMartin@MA.UltraNet.Com

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 2 May 1999 21:23:52 -0700
Organization: Chez Inwap
Message-ID: <7gj8co$flt$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7ghaqi$hnv$2@ligarius.ultra.net> <372C95FD.8F899038@MA.UltraNet.Com> <FB4xAy.AoE@world.std.com>
Lines: 21
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925705437 211 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.nyu.edu!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <FB4xAy.AoE@world.std.com>, Sam Weiner <weiner@world.std.com> wrote:
>TOPS-20 monitor builds were done on TOPS-20, I'm not sure about the
>language products.

TOPS-20 monitor builds could be done on TOPS-20.  But some people didn't.

I have heard that following situation was repeatable:

  Person in charge of TOPS-20 Monitor development says the files
  are ready to compile.  Person A starts compiling the Monitor sources
  on a TOPS-20 machine with 10 users.  Person B FTP's the files over
  to a similarly equiped TOPS-10 machine with 10 users.  Person B
  compiles the sources and links the Monitor on the TOPS-10 system,
  then FTP's the EXE file back to the TOPS-20 machine.  Person B wins
  the race; person A's compile and load job is still running.

	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: Roy Trubshaw <Roy@museltd.demon.co.uk>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 3 May 1999 00:28:02 +0100
Organization: Very little
Message-ID: <RLDFmCAC+NL3EwuU@museltd.demon.co.uk>
References: <371be3e4.0@newsfeed.one.net> <7gcr9l$hn5@weyl.math.psu.edu>
 <3729fa95.0@news.wizvax.net> <7gdk1j$ilu@weyl.math.psu.edu>
 <7gfnpc$3in$1@shell3.ba.best.com>
NNTP-Posting-Host: museltd.demon.co.uk
X-NNTP-Posting-Host: museltd.demon.co.uk:158.152.108.40
X-Trace: news.demon.co.uk 925688691 nnrp-13:17862 NO-IDENT museltd.demon.co.uk:158.152.108.40
X-Complaints-To: abuse@demon.net
MIME-Version: 1.0
X-Newsreader: Turnpike Integrated Version 4.02 U <PA+tDYAQAmEUIwT6QWhMCaEU0f>
Lines: 34
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!museltd.demon.co.uk!Roy

In article <7gfnpc$3in$1@shell3.ba.best.com>, Joe Smith
<inwap@best.com> writes
>In article <7gdk1j$ilu@weyl.math.psu.edu>,
>Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>>any case I strongly doubt it was possible to change the PPN of a UFD in-place
>>                                                      ^^^^^
>>      Excuse me? Do you imply that user could have only one directory?
>>AFAICS PPN == UID+GID and it looks like per-user thing, not per-directory one.
>
>Each user could have only one top-level directory per disk.
>
[Snip stuff about UFDs and multiple devices]
>The ersatz device name "MFD" was an alias for "DSK:[1,1]".  The only entries
>in the Master File Directory were the UFDs (User File Directories).
>UFDs were the login directorys, and could contain files or SFDs (Sub File
>Directories).  The limit was 5 levels if SFDs.
5 levels explicitly, our favourite way of exceeding disk quota was to
"bury" the offending files in SFDs greater than 5 levels deep! (You have
to rename the SFDs into another SFD, thus "pushing" the contents of the
first SFD one level deeper. Keep this up to create SFDs nested
arbitrarily deep!) {Actually the accounting system new about the size of
the files 8-{, but the app. that ran to clean up (eventually) couldn't
find them. Oh how we laughed! The CUSP that did the work actually
crashed!?! Neat work for a student in those days.}

[Snipped]
>
>       -Joe

-- 
Roy Trubshaw
"... their fundamental design flaws are completely hidden by their
superficial design flaws."
 - Douglas Adams; So long, and thanks for all the fish

######

From: Roy Trubshaw <Roy@museltd.demon.co.uk>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 3 May 1999 00:43:42 +0100
Organization: Very little
Message-ID: <iI45WEAuMOL3EwZZ@museltd.demon.co.uk>
References: <372c6419$0$496@news.zetnet.co.uk>
NNTP-Posting-Host: museltd.demon.co.uk
X-NNTP-Posting-Host: museltd.demon.co.uk:158.152.108.40
X-Trace: news.demon.co.uk 925688692 nnrp-13:17862 NO-IDENT museltd.demon.co.uk:158.152.108.40
X-Complaints-To: abuse@demon.net
MIME-Version: 1.0
X-Newsreader: Turnpike Integrated Version 4.02 U <PA+tDYAQAmEUIwT6QWhMCaEU0f>
Lines: 33
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!isdnet!newsfeed.icl.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!museltd.demon.co.uk!Roy

In article <372c6419$0$496@news.zetnet.co.uk>, lisard@zetnet.co.uk
writes
>On 1999-05-01 dpeschel@u.washington.edu(DerekPeschel) said:
>   :I've always wanted to know more about BCPL.  Specifically, I've
>   :heard about the "global vector".  Is this related to the dispatch
>   :table you talked about?  Or was that only an OS6 creation?
[Good stuff about the only true programming language]
>
>The global vector doesn't in itself add overhead; the overhead comes
>because *every* function call is indirected through a pointer, and also
>because you have to scale pointers. BCPL's other quirk is that a
>word-addressed machine is an inbuilt assumption. The earlier 68Ks can't
>scale addresses in the way that the 386 (for example) can, which means
>that every memory addressing operation needs a left shift by 2 as well.
>Now that's overhead. :> However, I think a 386 would be a near-perfect
>BCPL machine, if someone fancies porting AmigaDOS...?
Actually the ideal BCPL machine _was_ the DecSystem-10. And of course
Essex BCPL was about the most fun you could have with your clothes on!
Mix in shareable, writable hi-segs ...
(Excuse me I'm just going into the garage to drool over my first MUD
listings?!? 8-})

[Snip]

Toodle pip,
Roy
PS We were always taught BCPL had _two_ types the bitstring and the
vector.
-- 
Roy Trubshaw
"... their fundamental design flaws are completely hidden by their
superficial design flaws."
 - Douglas Adams; So long, and thanks for all the fish

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
From: weiner@world.std.com (Sam Weiner)
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Message-ID: <FB4wwn.5qL@world.std.com>
Date: Mon, 3 May 1999 02:14:47 GMT
References: <372c6419$0$496@news.zetnet.co.uk> <iI45WEAuMOL3EwZZ@museltd.demon.co.uk>
Organization: The World Public Access UNIX, Brookline, MA
Lines: 21
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!dispose.news.demon.net!demon!tank.news.pipex.net!pipex!uunet!ams.uu.net!ffx.uu.net!in1.uu.net!world!weiner

In article <iI45WEAuMOL3EwZZ@museltd.demon.co.uk>,
Roy Trubshaw  <Roy@museltd.demon.co.uk> wrote:

[snip BCPL on Amiga...]

>Actually the ideal BCPL machine _was_ the DecSystem-10. And of course
>Essex BCPL was about the most fun you could have with your clothes on!
>Mix in shareable, writable hi-segs ...

[snip ...]

It should be noted BLISS was descended from BCPL and used for a number
of TOPS-10/-20 programs including FORTRAN-10, RMS-10/-20, Datatrieve-20,
etc.  A few customers used it as well despite the lack of support from
Digital.

Sam

PS	Yes Barb, I've heard your rant before but I still liked it.

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
From: weiner@world.std.com (Sam Weiner)
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Message-ID: <FB4xAy.AoE@world.std.com>
Date: Mon, 3 May 1999 02:23:21 GMT
References: <371be3e4.0@newsfeed.one.net> <372B1D0A.2AB40504@MA.UltraNet.Com> <7ghaqi$hnv$2@ligarius.ultra.net> <372C95FD.8F899038@MA.UltraNet.Com>
Organization: The World Public Access UNIX, Brookline, MA
Lines: 47
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!newsfeed.tli.de!newsfeed.amsterdam.nl.net!sun4nl!uunet!ams.uu.net!ffx.uu.net!in1.uu.net!world!weiner

In article <372C95FD.8F899038@MA.UltraNet.Com>,
Alan H. Martin <AMartin@MA.UltraNet.Com> wrote:
>jmfbahciv@aol.com wrote:
>> 
>> In article <372B1D0A.2AB40504@MA.UltraNet.Com>,
>>    "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
>> >jmfbahciv@aol.com wrote:
>> >>
>> >> In article <3723DBA5.E5B2FF0@MA.UltraNet.Com>,
>> >>    "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
>> >> >"Peter Hendén" wrote:
>> >> >> jmfbahciv@aol.com wrote:
>> >> >....
>> >> >> >The reason I didn't like to use the -20 is because ...; it was
>> >> >> >too slow for my taste ...
>> >> >>
>> >> >> Wasn't this dependent on whether your programs were
>> >> >> TOPS-10 (used PA1050) or TOPS-20 native. As I recall
>> >> >> (1977 - 80) we had a lot of users on our machine running
>> >> >> Interlisp-20 and a bunch of other native programs, and they
>> >> >> didn't get sluggish until long after the TOPS-10 progs did.
>...
>> >>... Just because one user used
>> >> something that did UUOs, should NOT have affected another user's
>> >> performance who is using JSYSes.  ...
>...
>> Reread what I wrote, Alan.  I would use the -20 in the early morning
>> until somebody else logged on (I didn't have to know about the
>> person..I could feel the system degrade).  Then, about that time,
>> TW would give the TOPS10 systems up for timesharing and I would
>> move to the -10.  If the -20 system was so great, how come
>> builds were taken over to the -10 (or didn't you know about that?)
>
>Could you sense whether the person was running native or compatibility apps?

TOPS-10 was definitely able to support more users in general but perhaps
some of what Barb experienced was from early versions of assemblers/linkers
which hadn't been modified to use native TOPS-20 I/O?  I'm not sure of
the timeframe (plus I was mostly a TOPS-10 person before LSBU) but I think
it took a while for TOPS-20 aware tools to be made available.

TOPS-20 monitor builds were done on TOPS-20, I'm not sure about the
language products.

Sam

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 3 May 1999 03:05:21 -0400
Organization: -ENOENT
Lines: 47
Message-ID: <7gjhrh$lrk@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7ggqll$u0$1@shell3.ba.best.com> <7gh5mj$kb3@weyl.math.psu.edu> <7gilg8$pvp$1@shell3.ba.best.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!sunqbc.risq.qc.ca!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7gilg8$pvp$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>It was written to disk.  The bits in the SAT (Storate Allocation Table)
>corresponding to the block containing the RIB and to the data blocks written
>thus far were marked as being in use.  The Monitor was very paranoid at
>making sure the disk structure was not damaged if a crash occurred at any
>time.  To create a new file:

>  *) Allocate blocks by find free bits in the SAT.  Mark 16 consecutive
>     blocks in use for the RIB and the first 15 data blocks.

>  *) Write the modified SAT to disk.  Wait for disk I/O to complete.

>  *) Mark the RIB has having 16 blocks; itself and the contiguous
>     data blocks.  Set the file's size to be 0 words.  Set the file's
>     allocated size to 16 blocks.  Write the RIB to disk, wait for I/O.

>  *) When the program does an OUT UUO, update the RIB and write the user's
>     data to disk.  Do not return to the user program until I/O is done.

Possible failure mode: directory getting full before CLOSE. How do you handle
that?

Grrrm... It's a funny semantics... Open anonymous with rename on close...
Let's see... Say it, we could mark upgrade-waiting records with \0 as a
first character and set ->close() in their default file_operations to
call vfs_rename()... Aha, ->readdir() should be told to skip such entries too.
That would give a reasonable approximation (+ flag for open(), indeed). Seems
to be doable with ext2 and ufs on Linux. Almost the same would work for 4.4BSD. 
OTOH NFS will be very unhappy with such stuff and bunch of userland will need
rewrite (fsck, dump and restore - for sure, maybe something else). That, and
issues with rename() and unlink() - final rename() waiting to happen may
create quite a few interesting holes. Ditto for directories with the sticky
bit set.

How does TOPS-10 implementation handle renaming/removing in presense of
still not closed new version?

>If the system crashed before the file was closed, there would not be any
>pointer in any directory for the RIB.  The RIB and its data blocks would
>become "lost blocks", and not reclaimed until CHKDSK was run with the
>system in single-user mode.

Was there any regular way to determine that block is a RIB?

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
From: Daniel Seagraves <root@bony.umtec.com>
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
In-Reply-To: <7gj8co$flt$1@shell3.ba.best.com>
Message-ID: <Pine.LNX.4.10.9905030748110.19836-100000@bony.umtec.com>
References: <371be3e4.0@newsfeed.one.net> <7ghaqi$hnv$2@ligarius.ultra.net> <372C95FD.8F899038@MA.UltraNet.Com> <FB4xAy.AoE@world.std.com> <7gj8co$flt$1@shell3.ba.best.com>
Approved: Why bother?
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Lines: 27
Date: Mon, 3 May 1999 07:49:07 -0500
NNTP-Posting-Host: 198.199.189.6
X-Trace: newsfeed.slurp.net 925735380 198.199.189.6 (Mon, 03 May 1999 07:43:00 CDT)
NNTP-Posting-Date: Mon, 03 May 1999 07:43:00 CDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.slurp.net!bony.umtec.com!root

On 2 May 1999, Joe Smith wrote:

> In article <FB4xAy.AoE@world.std.com>, Sam Weiner <weiner@world.std.com> wrote:
> >TOPS-20 monitor builds were done on TOPS-20, I'm not sure about the
> >language products.
> 
> TOPS-20 monitor builds could be done on TOPS-20.  But some people didn't.
> 
> I have heard that following situation was repeatable:
> 
>   Person in charge of TOPS-20 Monitor development says the files
>   are ready to compile.  Person A starts compiling the Monitor sources
>   on a TOPS-20 machine with 10 users.  Person B FTP's the files over
>   to a similarly equiped TOPS-10 machine with 10 users.  Person B
>   compiles the sources and links the Monitor on the TOPS-10 system,
>   then FTP's the EXE file back to the TOPS-20 machine.  Person B wins
>   the race; person A's compile and load job is still running.

Did this work the opposite way?  Could you use a TOPS-20 to build a TOPS-10?

Daniel Seagraves | I'm an International Clandestine Arms Dealer!
#!/bin/perl -sp0777i<X+d*lMLa^*lN%0]dsXx++lMlN/dsM0<j]dsj
$/=unpack('H*',$_);$_=`echo 16dio\U$k"SK$/SM$n\EsN0p[lN*1
lK[d2%Sa2/d0$^Ixp"|dc`;s/\W//g;$_=pack('H*',/((..)*)$/)
What is this?  See http://www.dcs.ex.ac.uk/~aba/rsa | 36 BITS 4EVER!

######

From: westin*nospam@graphics.cornell.edu (Stephen H. Westin)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 03 May 1999 08:41:56 -0400
Organization: Cornell University Program of Computer Graphics
Lines: 22
Sender: westin@DIESEL
Message-ID: <uogk21g17.fsf@graphics.cornell.edu>
References: <371be3e4.0@newsfeed.one.net> <7gcoig$63f$1@shell3.ba.best.com> <7gcr9l$hn5@weyl.math.psu.edu> <3729fa95.0@news.wizvax.net> <7gdk1j$ilu@weyl.math.psu.edu>
NNTP-Posting-Host: diesel.graphics.cornell.edu
X-Newsreader: Gnus v5.5/Emacs 20.3
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newspeer.monmouth.com!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news.graphics.cornell.edu!news

viro@weyl.math.psu.edu (Alexander Viro) writes:

> In article <3729fa95.0@news.wizvax.net>, John Wilson <wilson@dbit.com> wrote:

<snip>

> >any case I strongly doubt it was possible to change the PPN of a UFD in-place
> 							^^^^^
> 	Excuse me? Do you imply that user could have only one directory?

No. Just that changing the owner of a given directory was not
supported. If you created it, you own it. Can't give it away.

> AFAICS PPN == UID+GID and it looks like per-user thing, not
> per-directory one.

<snip>

-- 
-Stephen H. Westin
Any information or opinions in this message are mine: they do not
represent the position of Cornell University or any of its sponsors.

######

From: westin*nospam@graphics.cornell.edu (Stephen H. Westin)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 03 May 1999 08:45:31 -0400
Organization: Cornell University Program of Computer Graphics
Sender: westin@DIESEL
Message-ID: <un1zm1fv8.fsf@graphics.cornell.edu>
References: <371be3e4.0@newsfeed.one.net> <7gc02r$go9@weyl.math.psu.edu> <7gcoig$63f$1@shell3.ba.best.com> <7gcr9l$hn5@weyl.math.psu.edu> <3729fa95.0@news.wizvax.net>
NNTP-Posting-Host: diesel.graphics.cornell.edu
X-Newsreader: Gnus v5.5/Emacs 20.3
Lines: 33
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!cpk-news-hub1.bbnplanet.com!news.gtei.net!cliffs.rs.itd.umich.edu!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news.graphics.cornell.edu!news

wilson@dbit.com (John Wilson) writes:

> In article <7gcr9l$hn5@weyl.math.psu.edu>,
> Alexander Viro <viro@weyl.math.psu.edu> wrote:
> >>Here is one command on TOPS-10 that cannot done the same way on Unix:
> >>	COPY RESUME.TXT=HEADER.TXT,RESUME.TXT,FOOTER.TXT
> >
> >[snip]
> >
> >	Can be done almost the same way. rename() over the open file
> >unlinks it, but leaves the thing intact. And it's atomic (if you
> >don't care for atomicity - simpler yet, open();unlink();open(); will
> >do the thing).
> 
> Maybe you can do flips and twists to get the same effect on UNIX if you
> *know* that the input and output files have the same name, but the point is,
> on TOPS-10 (and many other DEC OSes) it will Just Work, even with simple-minded
> programs that plow ahead and open the files the most obvious way.

Another subtlety that seems to have been lost in the discussion is
file locking: that it's possible in many systems (but not Unix) to
open a file for write, preventing others from doing the same. In
Apollo's DOMAIN/OS, for example, this was the default mode; if you're
writing a file, no one else can clobber it by blindly writing it at
the same time. My impression is that TOPS-10/TOPS-20 had similar
design.

<snip>

-- 
-Stephen H. Westin
Any information or opinions in this message are mine: they do not
represent the position of Cornell University or any of its sponsors.

######

From: westin*nospam@graphics.cornell.edu (Stephen H. Westin)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 03 May 1999 08:54:21 -0400
Organization: Cornell University Program of Computer Graphics
Lines: 23
Sender: westin@DIESEL
Message-ID: <ulnf61fgi.fsf@graphics.cornell.edu>
References: <371be3e4.0@newsfeed.one.net> <372b2c32.0@news.wizvax.net> <7gfg5m$jen@weyl.math.psu.edu> <7ggqll$u0$1@shell3.ba.best.com> <7gh5mj$kb3@weyl.math.psu.edu>
NNTP-Posting-Host: diesel.graphics.cornell.edu
X-Newsreader: Gnus v5.5/Emacs 20.3
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newspump.monmouth.com!newspeer.monmouth.com!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news.graphics.cornell.edu!news

viro@weyl.math.psu.edu (Alexander Viro) writes:

<snip>

> Another thing being: was there an equivalent of buffer cache? Darn useful
> when you are doing a big compile... Since you had a big lock over write
> access it shouldn't give any problems with cache consistency. IOW: how many
> disk reads would happen if two processes had the same file opened for read
> and both requested read on the same block? OTOH I got used to situation when
> CPU is much faster than any IO, so relative costs might be completely
> different...

CPU has always been faster than I/O. At least since, say, about
1955. Disks still have access times measured in milliseconds; if you
connected a disk drive of today to a KA10, there would still be about
3 orders of magnitude speed advantage to the CPU/core memory.

<snip>

-- 
-Stephen H. Westin
Any information or opinions in this message are mine: they do not
represent the position of Cornell University or any of its sponsors.

######

From: blarsen@infostream.no (Bjørn Hell Larsen)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 03 May 1999 09:15:47 +0200
Lines: 18
Message-ID: <u90b6mxng.fsf@infostream.no>
References: <371be3e4.0@newsfeed.one.net> <7geldj$j3t$2@antiochus.ultra.net> <372b2c32.0@news.wizvax.net> <7gfg5m$jen@weyl.math.psu.edu> <7ggqll$u0$1@shell3.ba.best.com>
NNTP-Posting-Host: aubert.svg.infostream.no
X-Newsreader: Gnus v5.5/Emacs 20.3
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!newsfeed1.online.no!newsfeed.online.no!Norway.EU.net!not-for-mail


inwap@best.com (Joe Smith) writes:

> One of the file protection codes was for append-only.  Set properly,
> other people could append to your file (until your disk quota was
> used up) but could not delete, supersede or update your file.

Darn. I seem to have forgotten the semantics of the file protection codes.
Was this something that you could set by regular protection, or did
you have to start messing about with FILDAE in order to achieve it?

(I once wrote a game that maintained a high-score list, and had to
use FILDAE protection to enable my program to write the list
while still not making it world-writable, but I cannot recall if
this entailed append or update access to the file.)


Bjørn

######

From: jmfbah@ma.ultranet.com (Barbara A. Huizenga)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 03 May 99 09:56:54 GMT
Organization: UltraNet Communications, Inc.
Lines: 39
Message-ID: <7gk23r$tu1$1@ligarius.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu><371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net><Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7fs7mv$hii$1@ligarius.ultra.net> <7fsjbg$91o$1@zingo.tninet.se> <3723DBA5.E5B2FF0@MA.UltraNet.Com> <7g1gvb$bid$1@ligarius.ultra.net> <372B1D0A.2AB40504@MA.UltraNet.Com> <7ghaqi$hnv$2@ligarius.ultra.net> <372C95FD.8F899038@MA.UltraNet.Com>
NNTP-Posting-Host: d9.dial-12.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 3 May 1999 11:42:51 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!newsfeed.mathworks.com!newsswitch.lcs.mit.edu!news.ultranet.com!d9

In article <372C95FD.8F899038@MA.UltraNet.Com>,
   "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
>jmfbahciv@aol.com wrote:
>> 
>> In article <372B1D0A.2AB40504@MA.UltraNet.Com>,
>>    "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
>> >jmfbahciv@aol.com wrote:
>> >>
>> >> In article <3723DBA5.E5B2FF0@MA.UltraNet.Com>,
>> >>    "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
>> >> >"Peter Hendén" wrote:
>> >> >> jmfbahciv@aol.com wrote:
>> >> >....
>> >> >> >The reason I didn't like to use the -20 is because ...; it was
>> >> >> >too slow for my taste ...
>> >> >>
>> >> >> Wasn't this dependent on whether your programs were
>> >> >> TOPS-10 (used PA1050) or TOPS-20 native. As I recall
>> >> >> (1977 - 80) we had a lot of users on our machine running
>> >> >> Interlisp-20 and a bunch of other native programs, and they
>> >> >> didn't get sluggish until long after the TOPS-10 progs did.
>....
>> >>... Just because one user used
>> >> something that did UUOs, should NOT have affected another user's
>> >> performance who is using JSYSes.  ...
>....
>> Reread what I wrote, Alan.  I would use the -20 in the early morning
>> until somebody else logged on (I didn't have to know about the
>> person..I could feel the system degrade).  Then, about that time,
>> TW would give the TOPS10 systems up for timesharing and I would
>> move to the -10.  If the -20 system was so great, how come
>> builds were taken over to the -10 (or didn't you know about that?)
>
>Could you sense whether the person was running native or compatibility apps?
>				/AHM
You still didn't read what I wrote.  The "program" that the user was
running was called LOGIN.  You can't get more native than that.

/BAH

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
From: lisard@zetnet.co.uk
X-Everything: Net-Tamer V 1.08X
Date: 03 May 1999 11:31:03 GMT
Lines: 13
Message-ID: <372d88f7$0$490@news.zetnet.co.uk>
NNTP-Posting-Host: man-122.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 925731063 490 194.247.40.156
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!newsfeed1.uni2.dk!newsfeed.tli.de!newscore.gigabell.net!newscore.ipf.de!news0.de.colt.net!colt.net!newspeer.clara.net!news.clara.net!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail



On 1999-05-03 weiner@world.std.com(SamWeiner) said:
   :It should be noted BLISS was descended from BCPL and used for a
   :number of TOPS-10/-20 programs including FORTRAN-10, RMS-10/-20,
   :Datatrieve-20, etc.  A few customers used it as well despite the
   :lack of support from Digital.
   :Sam
   :PS Yes Barb, I've heard your rant before but I still liked it.

I haven't. Please, do tell...
--
Communa (lisard@zetnet.co.uk) -- you know soft spoken changes nothing

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 03 May 99 13:38:09 GMT
Organization: UltraNet Communications, Inc.
Lines: 59
Message-ID: <7gkf2m$pj9$1@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <372B1D0A.2AB40504@MA.UltraNet.Com> <7ghaqi$hnv$2@ligarius.ultra.net> <372C95FD.8F899038@MA.UltraNet.Com> <FB4xAy.AoE@world.std.com>
NNTP-Posting-Host: d4.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 3 May 1999 15:24:06 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!netnews.com!newspeer.monmouth.com!news.ultranet.com!d4

In article <FB4xAy.AoE@world.std.com>,
   weiner@world.std.com (Sam Weiner) wrote:
>In article <372C95FD.8F899038@MA.UltraNet.Com>,
>Alan H. Martin <AMartin@MA.UltraNet.Com> wrote:
>>jmfbahciv@aol.com wrote:
>>> 
>>> In article <372B1D0A.2AB40504@MA.UltraNet.Com>,
>>>    "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
>>> >jmfbahciv@aol.com wrote:
>>> >>
>>> >> In article <3723DBA5.E5B2FF0@MA.UltraNet.Com>,
>>> >>    "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
>>> >> >"Peter Hendén" wrote:
>>> >> >> jmfbahciv@aol.com wrote:
>>> >> >....
>>> >> >> >The reason I didn't like to use the -20 is because ...; it was
>>> >> >> >too slow for my taste ...
>>> >> >>
>>> >> >> Wasn't this dependent on whether your programs were
>>> >> >> TOPS-10 (used PA1050) or TOPS-20 native. As I recall
>>> >> >> (1977 - 80) we had a lot of users on our machine running
>>> >> >> Interlisp-20 and a bunch of other native programs, and they
>>> >> >> didn't get sluggish until long after the TOPS-10 progs did.
>>...
>>> >>... Just because one user used
>>> >> something that did UUOs, should NOT have affected another user's
>>> >> performance who is using JSYSes.  ...
>>...
>>> Reread what I wrote, Alan.  I would use the -20 in the early morning
>>> until somebody else logged on (I didn't have to know about the
>>> person..I could feel the system degrade).  Then, about that time,
>>> TW would give the TOPS10 systems up for timesharing and I would
>>> move to the -10.  If the -20 system was so great, how come
>>> builds were taken over to the -10 (or didn't you know about that?)
>>
>>Could you sense whether the person was running native or compatibility 
apps?
>
>TOPS-10 was definitely able to support more users in general but perhaps
>some of what Barb experienced was from early versions of 
assemblers/linkers
>which hadn't been modified to use native TOPS-20 I/O?  I'm not sure of
>the timeframe (plus I was mostly a TOPS-10 person before LSBU) but I think
>it took a while for TOPS-20 aware tools to be made available.
>
>TOPS-20 monitor builds were done on TOPS-20,

Officially, they were done on the -20.  It was with the
advent of V6 LINK that the -20 guys had to build on the -20
and not sneak over to the -10 (but some of them still 
assembled over on the -10).  I'm not talking about tasks
done by Release Engineering, Sam.

> I'm not sure about the language products.

After they got their own machine, the builds were done on the -20
for the -20.

/BAH

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 03 May 99 13:40:24 GMT
Organization: UltraNet Communications, Inc.
Lines: 24
Message-ID: <7gkf6t$pj9$2@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7ghaqi$hnv$2@ligarius.ultra.net> <372C95FD.8F899038@MA.UltraNet.Com> <FB4xAy.AoE@world.std.com> <7gj8co$flt$1@shell3.ba.best.com>
NNTP-Posting-Host: d4.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 3 May 1999 15:26:21 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d4

In article <7gj8co$flt$1@shell3.ba.best.com>,
   inwap@best.com (Joe Smith) wrote:
>In article <FB4xAy.AoE@world.std.com>, Sam Weiner <weiner@world.std.com> 
wrote:
>>TOPS-20 monitor builds were done on TOPS-20, I'm not sure about the
>>language products.
>
>TOPS-20 monitor builds could be done on TOPS-20.  But some people didn't.
>
>I have heard that following situation was repeatable:
>
>  Person in charge of TOPS-20 Monitor development says the files
>  are ready to compile.  Person A starts compiling the Monitor sources
>  on a TOPS-20 machine with 10 users.  Person B FTP's the files over
>  to a similarly equiped TOPS-10 machine with 10 users.  Person B
>  compiles the sources and links the Monitor on the TOPS-10 system,
>  then FTP's the EXE file back to the TOPS-20 machine.  Person B wins
>  the race; person A's compile and load job is still running.

Even when sneaker FTP (I think that's what it's called) was used.
And it was done on the sly and never during prime time since -20 
work on the -10 (back then) was frowned upon.

/BAH

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 03 May 99 13:44:26 GMT
Organization: UltraNet Communications, Inc.
Lines: 42
Message-ID: <7gkfef$pj9$3@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu><371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net><Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7fs7mv$hii$1@ligarius.ultra.net> <7fsjbg$91o$1@zingo.tninet.se> <3723DBA5.E5B2FF0@MA.UltraNet.Com> <7g1gvb$bid$1@ligarius.ultra.net> <372B1D0A.2AB40504@MA.UltraNet.Com> <7ghaqi$hnv$2@ligarius.ultra.net> <372C95FD.8F899038@MA.UltraNet.Com>
NNTP-Posting-Host: d4.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 3 May 1999 15:30:23 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d4

In article <372C95FD.8F899038@MA.UltraNet.Com>,
   "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
>jmfbahciv@aol.com wrote:
>> 
>> In article <372B1D0A.2AB40504@MA.UltraNet.Com>,
>>    "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
>> >jmfbahciv@aol.com wrote:
>> >>
>> >> In article <3723DBA5.E5B2FF0@MA.UltraNet.Com>,
>> >>    "Alan H. Martin" <AMartin@MA.UltraNet.Com> wrote:
>> >> >"Peter Hendén" wrote:
>> >> >> jmfbahciv@aol.com wrote:
>> >> >....
>> >> >> >The reason I didn't like to use the -20 is because ...; it was
>> >> >> >too slow for my taste ...
>> >> >>
>> >> >> Wasn't this dependent on whether your programs were
>> >> >> TOPS-10 (used PA1050) or TOPS-20 native. As I recall
>> >> >> (1977 - 80) we had a lot of users on our machine running
>> >> >> Interlisp-20 and a bunch of other native programs, and they
>> >> >> didn't get sluggish until long after the TOPS-10 progs did.
>....
>> >>... Just because one user used
>> >> something that did UUOs, should NOT have affected another user's
>> >> performance who is using JSYSes.  ...
>....
>> Reread what I wrote, Alan.  I would use the -20 in the early morning
>> until somebody else logged on (I didn't have to know about the
>> person..I could feel the system degrade).  Then, about that time,
>> TW would give the TOPS10 systems up for timesharing and I would
>> move to the -10.  If the -20 system was so great, how come
>> builds were taken over to the -10 (or didn't you know about that?)
>
>Could you sense whether the person was running native or compatibility 
apps?

It was when the guy logged in (you didn't read what I wrote).  You 
can't get more native than that.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 03 May 99 13:48:53 GMT
Organization: UltraNet Communications, Inc.
Lines: 22
Message-ID: <7gkfmq$pj9$4@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu><371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net><Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7fs7mv$hii$1@ligarius.ultra.net> <7fsjbg$91o$1@zingo.tninet.se> <3723DBA5.E5B2FF0@MA.UltraNet.Com> <7g1gvb$bid$1@ligarius.ultra.net> <372B1D0A.2AB40504@MA.UltraNet.Com> <7ghaqi$hnv$2@ligarius.ultra.net> <372C95FD.8F899038@MA.UltraNet.Com>
NNTP-Posting-Host: d4.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 3 May 1999 15:34:50 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d4


Due to trying to fix something that wasn't broken, all of my
defaults, settings, and who-knows-what got creamed w.r.t.
this ISP's software.  I am in the process of trying to
figure out and restore my year-old save...sigh!!!!!  It
doesn't help that windows has been multiplying config
files into more than one directory.  I have been mentally
attaching Gates' balls to the Bruins' hockey puck....and
I think I'll be attaching his head to the stick.

AAAARRRRRRRRGGGGGGGGG!!!!!!!!!!!

If you do notice something strange..like my full name getting
printed or other things that I've been trying to circumvent,
please let me know.  And if things are set, mail notifications
to jmfbah@aol.com and not here at Ultranet.

I considering practicing Ludditeism at the moment.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 03 May 99 14:01:10 GMT
Organization: UltraNet Communications, Inc.
Lines: 20
Message-ID: <7gkgdq$pj9$5@antiochus.ultra.net>
References: <372d88f7$0$490@news.zetnet.co.uk>
NNTP-Posting-Host: d4.dial-15.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 3 May 1999 15:47:06 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.wirehub.nl!netnews.com!newsswitch.lcs.mit.edu!news.ultranet.com!d4

In article <372d88f7$0$490@news.zetnet.co.uk>, lisard@zetnet.co.uk wrote:
>
>
>On 1999-05-03 weiner@world.std.com(SamWeiner) said:
>   :It should be noted BLISS was descended from BCPL and used for a
>   :number of TOPS-10/-20 programs including FORTRAN-10, RMS-10/-20,
>   :Datatrieve-20, etc.  A few customers used it as well despite the
>   :lack of support from Digital.
>   :Sam
>   :PS Yes Barb, I've heard your rant before but I still liked it.
>
>I haven't. Please, do tell...

<grin>  Can't....unless you allow swears.  BLISS of any flavor
rapidly became a swear in one of my jobs.  Of course, BLISS
was a swear to anybody who had to support it, too.

/BAH

Subtract a hundred and four for e-mail.

######

From: nospam-gagner@clark.net (Philip Gagner)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 04 May 1999 01:10:05 GMT
Organization: TWLG
Reply-To: nospam-gagner@clark.net
Message-ID: <37324689.4539727@news.kivex.com>
References: <371be3e4.0@newsfeed.one.net> <7g0usf$4lg@weyl.math.psu.edu> <H20V2.1086$WA4.216881@ptah.visi.com> <p98mccabe-2604991226580001@mepa2pp48.alltel.net> <7g26qv$55r@weyl.math.psu.edu> <7g9jj5$88h$2@antiochus.ultra.net>
X-Newsreader: Forte Agent 1.5/32.451
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
NNTP-Posting-Host: pc162.svg-law.com
Lines: 35
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.cwix.com!129.250.35.146!iad-peer.news.verio.net!uunet!ffx.uu.net!in3.uu.net!newsfeed.one.net!news!pc162.svg-law.com

Yes, this was the theory. In practice, if what one was debugging
wasn't really VERY likely to halt the system, then you a user with
appropriate privileges could load FILDDT and watch (as well as modify)
memory directly, including monitor tables, etc.

On very many more than one occassions, people in the monitor group
wrote code in a patch space, then loaded a jump to that code, thereby
debugging the monitor while it wasn't stand alone. On very few
occassions this resulted in the system crashing, which resulted in
sheepish explanations, lectures from management, annoyed operators,
and half-hearted assurances it would never happen again. Of course
what the managers meant would never happen was this sort of
test-on-the-fly during daylight timesharing hours. What WE meant was
that the system wouldn't crash next time, because we'd figured out a
fix.

Oh, and FILDDT was, of course, a user-level program. So although BAH
is correct as a matter of management theory, the facility existed to
do it differently.

>When one wanted to debug the TOPS10 monitor, one took the whole
>system stand-alone.  The monitor could not be debugged as a user
>mode program.  A very special DDT, called EDDT, was loaded with
>the monitor for symbolic debugging purposes.  On our systems,
>one could go to the system console, type <ctrl-D>, and stop
>the system by hitting that special break point.  Since our
>shops were run based on something we called "gentleman's 
>timesharing" this was not done unless there existed a great
>need.  There a few times that the command was typed by
>mistake but an <altmode>P would continue system operation.
>
>/BAH
>
>Subtract a hundred and four for e-mail.

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <372d88f7$0$490@news.zetnet.co.uk> <7gkgdq$pj9$5@antiochus.ultra.net>
X-Newsreader: NN version 6.5.0 CURRENT #119
From: werme@werme.ne.mediaone.net (Ric Werme)
Lines: 29
Message-ID: <2IsX2.28538$tY1.18718@wbnws01.ne.mediaone.net>
Date: Tue, 04 May 1999 02:12:14 GMT
NNTP-Posting-Host: 24.128.109.10
X-Trace: wbnws01.ne.mediaone.net 925783934 24.128.109.10 (Mon, 03 May 1999 22:12:14 EDT)
NNTP-Posting-Date: Mon, 03 May 1999 22:12:14 EDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!chnws02.mediaone.net!24.128.1.101!chnws05.ne.mediaone.net!24.128.44.7!wbnws01.ne.mediaone.net.POSTED!not-for-mail

jmfbahciv@aol.com writes:

>In article <372d88f7$0$490@news.zetnet.co.uk>, lisard@zetnet.co.uk wrote:
>>
>>
>>On 1999-05-03 weiner@world.std.com(SamWeiner) said:
>>   :It should be noted BLISS was descended from BCPL and used for a
>>   :number of TOPS-10/-20 programs including FORTRAN-10, RMS-10/-20,
>>   :Datatrieve-20, etc.  A few customers used it as well despite the
>>   :lack of support from Digital.
>>   :Sam
>>   :PS Yes Barb, I've heard your rant before but I still liked it.
>>
>>I haven't. Please, do tell...

><grin>  Can't....unless you allow swears.  BLISS of any flavor
>rapidly became a swear in one of my jobs.  Of course, BLISS
>was a swear to anybody who had to support it, too.

After I found Ed Taft's PDP-10 I/O package he wrote at Harvard, I started
rewriting some of my BLISS code in MACRO-10.  With I/O hassles solved, fast
compile times, and a wonder macro language, why use BLISS?

I later named the package Tulip, it's on a DECUS tape.  Need to write a WWW
page for it someday.
--
Ric Werme                  | http://people.ne.mediaone.net/werme
werme@nospam.mediaone.net  | http://www.cyberportal.net/werme
      ^^^^^^^ delete

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 03:29:37 -0700
Organization: Chez Inwap
Message-ID: <7gmi6h$cl3$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <FB4xAy.AoE@world.std.com> <7gj8co$flt$1@shell3.ba.best.com> <Pine.LNX.4.10.9905030748110.19836-100000@bony.umtec.com>
Lines: 12
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925813781 202 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newspump.monmouth.com!newspeer.monmouth.com!newsfeed.enteract.com!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <Pine.LNX.4.10.9905030748110.19836-100000@bony.umtec.com>,
Daniel Seagraves  <root@bony.umtec.com> wrote:
>Did this work the opposite way?  Could you use a TOPS-20 to build a TOPS-10?

Should have.  The goal was that the TOPS-10 build process was self-contained.
All the source files and UNV files (similar to "include files") came from
the current directory; no system library files were included in the Monitor.
	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 03:55:33 -0700
Organization: Chez Inwap
Message-ID: <7gmjn5$lo2$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gh5mj$kb3@weyl.math.psu.edu> <7gilg8$pvp$1@shell3.ba.best.com> <7gjhrh$lrk@weyl.math.psu.edu>
Lines: 46
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925815338 210 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!howland.erols.net!newsfeed.berkeley.edu!news2.best.com!news.idiom.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gjhrh$lrk@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>Possible failure mode: directory getting full before CLOSE. How do you handle
>that?

Directories don't get full.  There was no limit to the number of file names
that could be stored in a directory.

>How does TOPS-10 implementation handle renaming/removing in presense of
>still not closed new version?

The RENAME UUO would fail with file-being-modified.
A RENAME on a file that had no writers was allowed - jobs that already
had the file open for reading would continue unaffected.

To rename a file, the program had to do a LOOKUP first, then do a RENAME
on the same channel.  If the file name portion of the LOOKUP/ENTER/RENAME
block was zero, the file would be deleted.

Directories were created by doing a LOOKUP and CLOSE with no OUTPUT.
A common method of creating SFDs was
	.MAKE FOOBAR.SFD
	*EX$$
which used the TECO text editor to create the directory.  Removing
a directory was usually done by using the regular DELETE command on
the empty directory.

>>If the system crashed before the file was closed, there would not be any
>>pointer in any directory for the RIB.  The RIB and its data blocks would
>>become "lost blocks", and not reclaimed until CHKDSK was run with the
>>system in single-user mode.
>
>Was there any regular way to determine that block is a RIB?

RIBs had a special code in the right half of the first word of the
first block of a cluster.  The left half was zero, something that
does not happen in ASCII files or system binary files.  CHKDSK did
several consistency checks, including looking at the block after the
end-of-file.  In addition to the block pointers, the RIB had the file's
name, extension, protection, and PPN.  There was no SFD info in the
RIB, so recovered files were written to the owner's UFD.
	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 04:02:59 -0700
Organization: Chez Inwap
Message-ID: <7gmk53$nnv$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gfg5m$jen@weyl.math.psu.edu> <7ggqll$u0$1@shell3.ba.best.com> <u90b6mxng.fsf@infostream.no>
Lines: 30
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925815782 225 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newspump.monmouth.com!newspeer.monmouth.com!newsfeed.wli.net!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <u90b6mxng.fsf@infostream.no>,
Bjørn Hell Larsen <blarsen@infostream.no> wrote:
>
>inwap@best.com (Joe Smith) writes:
>
>> One of the file protection codes was for append-only.  Set properly,
>> other people could append to your file (until your disk quota was
>> used up) but could not delete, supersede or update your file.
>
>Darn. I seem to have forgotten the semantics of the file protection codes.
>Was this something that you could set by regular protection, or did
>you have to start messing about with FILDAE in order to achieve it?

That was a standard feature, available pre-FILDAE.
"PROTECT <244> FILE.DAT" made it append-only and protected yourself
from an accidental "DELETE *.DAT".

>(I once wrote a game that maintained a high-score list, and had to
>use FILDAE protection to enable my program to write the list
>while still not making it world-writable, but I cannot recall if
>this entailed append or update access to the file.)

Something like putting "FILE.DAT/APPEND/PROGRAM:[12,14]GAME.EXE"
inside of ACCESS.INI to allow any PPN to append to the file, but only
if they were running GAME.EXE from directory [12,14].
	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 04:15:04 -0700
Organization: Chez Inwap
Message-ID: <7gmkro$qvn$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <3729fa95.0@news.wizvax.net> <7gdk1j$ilu@weyl.math.psu.edu> <uogk21g17.fsf@graphics.cornell.edu>
Lines: 44
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925816508 226 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!newsfeed.berkeley.edu!news2.best.com!news.idiom.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <uogk21g17.fsf@graphics.cornell.edu>,
Stephen H. Westin <westin*nospam@graphics.cornell.edu> wrote:
>viro@weyl.math.psu.edu (Alexander Viro) writes:
>
>> In article <3729fa95.0@news.wizvax.net>, John Wilson <wilson@dbit.com> wrote:
>
><snip>
>
>> >any case I strongly doubt it was possible to change the PPN of a UFD in-place
>> 							^^^^^
>> 	Excuse me? Do you imply that user could have only one directory?
>
>No. Just that changing the owner of a given directory was not
>supported. If you created it, you own it. Can't give it away.

UFDs were created by LOGIN and MOUNT.
UFDs were deleted by LOGOUT and DISMOUNT if empty.
UFDs belonged to [1,1] but the files inside them belonged to the PPN
matching the UFD's name.
UFDs could not be renamed.

Files could be renamed from one UFD to another, but ownership was
transfered to the destination PPN.

As far as I remember, SFDs could not be renamed from one UFD to another.

>> AFAICS PPN == UID+GID and it looks like per-user thing, not
>> per-directory one.

It's both.

TOPS-10 people being introduced to Unix had to learn a new concept.  With
Unix, even though you renamed a file out of your directory and into another
user's directory, you still had ownership of the file and it was still counted
against your disk quota.  With TOPS-10, files belonged to you only if they
were in your directory tree.  If you created a directory protected <777>
(writable by world), then any files created in it belonged to you and were
charged against your disk quota regardless of who created them.
	-Joe

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 04:19:57 -0700
Organization: Chez Inwap
Message-ID: <7gml4t$s3v$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gcr9l$hn5@weyl.math.psu.edu> <3729fa95.0@news.wizvax.net> <un1zm1fv8.fsf@graphics.cornell.edu>
Lines: 23
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925816800 226 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.berkeley.edu!news2.best.com!news.idiom.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <un1zm1fv8.fsf@graphics.cornell.edu>,
Stephen H. Westin <westin*nospam@graphics.cornell.edu> wrote:
>Another subtlety that seems to have been lost in the discussion is
>file locking: that it's possible in many systems (but not Unix) to
>open a file for write, preventing others from doing the same. In
>Apollo's DOMAIN/OS, for example, this was the default mode; if you're
>writing a file, no one else can clobber it by blindly writing it at
>the same time. My impression is that TOPS-10/TOPS-20 had similar
>design.

Yep.  Mandatory file locking.  No one, not even [1,2] (the equivalent of
'root') could write to a file that was currently open for writing.  The only
exception was when owner of the write lock performed a special UUO to set the
file to simultaneous update mode.

One paper I saw on the development of Unix stated that they chose not
to implement mandatory file locking because then 'root' would not be
all powerful.  Hooey.
	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 05:18:25 -0700
Organization: Chez Inwap
Message-ID: <7gmoih$mgg$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7g9jj5$88h$2@antiochus.ultra.net> <37324689.4539727@news.kivex.com> <7gmjv5$aq5$1@antiochus.ultra.net>
Lines: 50
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925820308 226 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gmjv5$aq5$1@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
>In article <37324689.4539727@news.kivex.com>,
>   nospam-gagner@clark.net (Philip Gagner) wrote:
>>Yes, this was the theory. In practice, if what one was debugging
>>wasn't really VERY likely to halt the system, then you a user with
>>appropriate privileges could load FILDDT and watch (as well as modify)
>>memory directly, including monitor tables, etc.
>
>Nitpik here.  FILDDT was the program to patch the monitor file
>that was one disk...not the code currently running--that's what
>EDDT was for and could be invoked by that nefarious command
><ctrl-D>.  Or am I misremembering?

Incomplete memory.

FILDDT could peek at the running monitor.  If write-enabled,
FILDDT could poke at the running monitor.  You could not set
breakpoints with FILDDT, but you could load the symbol table
from SYSTEM.EXE and view locations and values symbolically.

I would use EDDT to set a breakpoint on a dummy routine when the
Monitor was loaded.  While the system was running, I would use
FILDDT to patch some instruction to jump to the dummy routine under
certain conditions.  When the system hit that condition, the Monitor
would come to a screeching halt, with EDDT running on the console.

>I should probaby point out that DDT.EXE, FILDDT.EXE and EDDT.REL
>were all built from the same DDT.MAC source file with the 
>appropriate feature test switches.

Don't forget DDT.REL (statically loaded with user's program), VMDDT.EXE
(dynamically merged in when DDT.REL was not loaded), and that FILDDT came in
two flavors (TOPS-10 and TOPS-20) all from the same source.

EDDT.REL, while designed to run in kernel mode, was capable of running in user
mode as well, using UUOs or JSYS for I/O to the terminal.  This allowed
for patching the Monitor by using
	.GET SYS:SYSTEM
	.DDT
	* manually patch things or read in a file with DDT commands
	^C
	.SAVE SYS:SYSTEM


		-Joe

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 05:27:15 -0700
Organization: Chez Inwap
Message-ID: <7gmp33$p17$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gjhrh$lrk@weyl.math.psu.edu> <7gmjn5$lo2$1@shell3.ba.best.com> <7gmko2$aq5$3@antiochus.ultra.net>
Lines: 20
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925820839 225 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gmko2$aq5$3@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
>In article <7gmjn5$lo2$1@shell3.ba.best.com>,
>But this is from the user point of view, Joe.  Tony would just 
>modify the UFD entry WITHOUT moving any bits of the file to
>be renamed.  The addresses changed...not the position.

One thing we forgot to mention about the features of FILDDT:
In addition to patching a file or modifying the running Monitor,
FILDDT was capable of modifying the disk structure directly.

You could use it to fix bad values in the HOME block, undo a
bad retrieval pointer, or obliterate an undeletable SFD.

The only thing I've seen vaguely like this function is 'ext2ed' on Linux.
	-Joe

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 05:31:26 -0700
Organization: Chez Inwap
Message-ID: <7gmpau$qfg$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gjhrh$lrk@weyl.math.psu.edu> <7gmjn5$lo2$1@shell3.ba.best.com> <7gmknh$nol@weyl.math.psu.edu>
Lines: 19
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925821092 198 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.wirehub.nl!newsfeed.tli.de!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gmknh$nol@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>Directories don't get full.  There was no limit to the number of file names
>>that could be stored in a directory.
>
>Where did you get infinite disks?

I was referring to there not being a hard limit, such as 256 entries
in C:\ on MS-DOS.

>What happened if the target of rename already existed?

Error code 4 (ERAEF%) = Already Existing Filename.
	-Joe

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 07:12:49 -0400
Organization: -ENOENT
Lines: 27
Message-ID: <7gmknh$nol@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7gilg8$pvp$1@shell3.ba.best.com> <7gjhrh$lrk@weyl.math.psu.edu> <7gmjn5$lo2$1@shell3.ba.best.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!newsxfer.itd.umich.edu.MISMATCH!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7gmjn5$lo2$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>In article <7gjhrh$lrk@weyl.math.psu.edu>,
>Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>Possible failure mode: directory getting full before CLOSE. How do you handle
>>that?
>
>Directories don't get full.  There was no limit to the number of file names
>that could be stored in a directory.

Where did you get infinite disks?

>>How does TOPS-10 implementation handle renaming/removing in presense of
>>still not closed new version?
>
>The RENAME UUO would fail with file-being-modified.
>A RENAME on a file that had no writers was allowed - jobs that already
>had the file open for reading would continue unaffected.
>
>To rename a file, the program had to do a LOOKUP first, then do a RENAME
>on the same channel.  If the file name portion of the LOOKUP/ENTER/RENAME
>block was zero, the file would be deleted.

What happened if the target of rename already existed?

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 04 May 99 09:13:45 GMT
Organization: UltraNet Communications, Inc.
Lines: 36
Message-ID: <7gmjv5$aq5$1@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7g0usf$4lg@weyl.math.psu.edu> <H20V2.1086$WA4.216881@ptah.visi.com> <p98mccabe-2604991226580001@mepa2pp48.alltel.net> <7g26qv$55r@weyl.math.psu.edu> <7g9jj5$88h$2@antiochus.ultra.net> <37324689.4539727@news.kivex.com>
NNTP-Posting-Host: d6.dial-12.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 4 May 1999 10:59:49 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d6

In article <37324689.4539727@news.kivex.com>,
   nospam-gagner@clark.net (Philip Gagner) wrote:
>Yes, this was the theory. In practice, if what one was debugging
>wasn't really VERY likely to halt the system, then you a user with
>appropriate privileges could load FILDDT and watch (as well as modify)
>memory directly, including monitor tables, etc.

Nitpik here.  FILDDT was the program to patch the monitor file
that was one disk...not the code currently running--that's what
EDDT was for and could be invoked by that nefarious command
<ctrl-D>.  Or am I misremembering?
>
>On very many more than one occassions, people in the monitor group
>wrote code in a patch space, then loaded a jump to that code, thereby
>debugging the monitor while it wasn't stand alone. On very few
>occassions this resulted in the system crashing, which resulted in
>sheepish explanations, lectures from management, annoyed operators,
>and half-hearted assurances it would never happen again. Of course
>what the managers meant would never happen was this sort of
>test-on-the-fly during daylight timesharing hours. What WE meant was
>that the system wouldn't crash next time, because we'd figured out a
>fix.

Yup [broadly grinning emoticon here].  That was also a test of
a manager.  The one who understood that, passed.  The others
were put into the clueless category and essentially barred from
ever getting hired into the group.
<snip>

I should probaby point out that DDT.EXE, FILDDT.EXE and EDDT.REL
were all built from the same DDT.MAC source file with the 
appropriate feature test switches.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 04 May 99 09:18:13 GMT
Organization: UltraNet Communications, Inc.
Lines: 27
Message-ID: <7gmk7g$aq5$2@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <FB4xAy.AoE@world.std.com> <7gj8co$flt$1@shell3.ba.best.com> <Pine.LNX.4.10.9905030748110.19836-100000@bony.umtec.com> <7gmi6h$cl3$1@shell3.ba.best.com>
NNTP-Posting-Host: d6.dial-12.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 4 May 1999 11:04:16 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d6

In article <7gmi6h$cl3$1@shell3.ba.best.com>,
   inwap@best.com (Joe Smith) wrote:
>In article <Pine.LNX.4.10.9905030748110.19836-100000@bony.umtec.com>,
>Daniel Seagraves  <root@bony.umtec.com> wrote:
>>Did this work the opposite way?  Could you use a TOPS-20 to build a 
TOPS-10?
>
>Should have.  The goal was that the TOPS-10 build process 
>was self-contained. All the source files and UNV files 
>(similar to "include files") came from the current directory; 
>no system library files were included in the Monitor.

On our systems, it just wasn't practical.  Besides the user
response time being abyssmal (I'm comparing it to an SMP TOPS10),
the setup so that the universals and RELs, etc. was almost
beyond source management techniques.  It also didn't help that
the -20 group wouldn't take our updated sources (e.g., UUOSYM,
MACTEN, SCAN) and ship them.  They always stayed with some
stuff that was so old, nothing would have worked.

I managed to solve that problem but not with a guarantee that
the process would continue to work.  Another little project
that I never could get off the ground.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 04 May 99 09:27:02 GMT
Organization: UltraNet Communications, Inc.
Lines: 45
Message-ID: <7gmko2$aq5$3@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7gh5mj$kb3@weyl.math.psu.edu> <7gilg8$pvp$1@shell3.ba.best.com> <7gjhrh$lrk@weyl.math.psu.edu> <7gmjn5$lo2$1@shell3.ba.best.com>
NNTP-Posting-Host: d6.dial-12.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 4 May 1999 11:13:06 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d6

In article <7gmjn5$lo2$1@shell3.ba.best.com>,
   inwap@best.com (Joe Smith) wrote:
>In article <7gjhrh$lrk@weyl.math.psu.edu>,
>Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>Possible failure mode: directory getting full before CLOSE. How do you 
handle
>>that?
>
>Directories don't get full.  There was no limit to the 
>number of file names that could be stored in a directory.

Well, there were practical limits.  One couldn't have a disk full
of one RIB...hmm...I wonder what would have happened if we had
tried to fill the disk with just a RIB and then modified a file?
To make the file organization extensible there was the notion of
a RIB...and then there was the spare RIB.

>
>>How does TOPS-10 implementation handle renaming/removing in presense of
>>still not closed new version?
>
>The RENAME UUO would fail with file-being-modified.
>A RENAME on a file that had no writers was allowed - jobs that already
>had the file open for reading would continue unaffected.
>
>To rename a file, the program had to do a LOOKUP first, then do a RENAME
>on the same channel.  If the file name portion of the LOOKUP/ENTER/RENAME
>block was zero, the file would be deleted.
>
>Directories were created by doing a LOOKUP and CLOSE with no OUTPUT.
>A common method of creating SFDs was
>	.MAKE FOOBAR.SFD
>	*EX$$
>which used the TECO text editor to create the directory.  Removing
>a directory was usually done by using the regular DELETE command on
>the empty directory.
<snip>

But this is from the user point of view, Joe.  Tony would just 
modify the UFD entry WITHOUT moving any bits of the file to
be renamed.  The addresses changed...not the position.

/BAH

Subtract a hundred and four for e-mail.

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 10:32:05 -0400
Organization: -ENOENT
Lines: 31
Message-ID: <7gn0d5$ntt@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7gmjn5$lo2$1@shell3.ba.best.com> <7gmknh$nol@weyl.math.psu.edu> <7gmpau$qfg$1@shell3.ba.best.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!sunqbc.risq.qc.ca!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7gmpau$qfg$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>In article <7gmknh$nol@weyl.math.psu.edu>,
>Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>>Directories don't get full.  There was no limit to the number of file names
>>>that could be stored in a directory.
>>
>>Where did you get infinite disks?
>
>I was referring to there not being a hard limit, such as 256 entries
>in C:\ on MS-DOS.

OK, so what would happen in the following scenario: we have a file FOO.
We do ENTER on it. New version (still nameless) is created. Something
else creates several files in the same directory, driving it to the state
when next record creation will require allocation of new block. We are
writing to the opened file, eating the quota up (or simply filling the
remaining space on disk). We are closing the file. Directory can't be
extended since there is no free blocks (or no more blocks are available
to us due to quota). Actions of the system?

>>What happened if the target of rename already existed?
>
>Error code 4 (ERAEF%) = Already Existing Filename.
>	-Joe

I.e. temporary immutable set on the old copy would more or less reporduce
the semantics... Aha...

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: Bill Westfield <billw@flipper.cisco.com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 04 May 1999 10:39:55 -0700
Organization: Ye 'Ol Disorganized NNTPCache groupie
Lines: 12
Message-ID: <544slswx78.fsf@flipper.cisco.com>
References: <371be3e4.0@newsfeed.one.net> <7ghaqi$hnv$2@ligarius.ultra.net>
	<372C95FD.8F899038@MA.UltraNet.Com> <FB4xAy.AoE@world.std.com>
	<7gj8co$flt$1@shell3.ba.best.com>
NNTP-Posting-Host: usenet.cisco.com
X-Newsreader: Gnus v5.3/Emacs 19.34
Cache-Post-Path: newsproxy.cisco.com!unknown@flipper.cisco.com
X-Cache: nntpcache 2.3.3 (see http://www.nntpcache.org/)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newspeer.monmouth.com!uunet!nyc.uu.net!ffx.uu.net!in2.uu.net!news-master.cisco.com!not-for-mail

    Person in charge of TOPS-20 Monitor development says the files
    are ready to compile.  Person A starts compiling the Monitor sources
    on a TOPS-20 machine with 10 users.  Person B FTP's the files over
    to a similarly equiped TOPS-10 machine with 10 users.

Well, the FTP probably used up all the 20's cycles, in addition to causing
other forms of thrashing there!

 :-)
BillW
-- 
(remove spam food from return address)

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 04 May 99 10:50:11 GMT
Organization: UltraNet Communications, Inc.
Lines: 38
Message-ID: <7gmpju$9gl$1@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7g9jj5$88h$2@antiochus.ultra.net> <37324689.4539727@news.kivex.com> <7gmjv5$aq5$1@antiochus.ultra.net> <7gmoih$mgg$1@shell3.ba.best.com>
NNTP-Posting-Host: d6.dial-12.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 4 May 1999 12:36:14 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d6

In article <7gmoih$mgg$1@shell3.ba.best.com>,
   inwap@best.com (Joe Smith) wrote:
>In article <7gmjv5$aq5$1@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
>>In article <37324689.4539727@news.kivex.com>,
>>   nospam-gagner@clark.net (Philip Gagner) wrote:
>>>Yes, this was the theory. In practice, if what one was debugging
>>>wasn't really VERY likely to halt the system, then you a user with
>>>appropriate privileges could load FILDDT and watch (as well as modify)
>>>memory directly, including monitor tables, etc.
>>
>>Nitpik here.  FILDDT was the program to patch the monitor file
>>that was one disk...not the code currently running--that's what
>>EDDT was for and could be invoked by that nefarious command
>><ctrl-D>.  Or am I misremembering?
>
>Incomplete memory.

Oh, of course!  /M/P  How could I forget that?   It has been
a bad week.  Thanks, Joe :-).
>
>FILDDT could peek at the running monitor.  If write-enabled,
>FILDDT could poke at the running monitor.  You could not set
>breakpoints with FILDDT, but you could load the symbol table
>from SYSTEM.EXE and view locations and values symbolically.
>
>I would use EDDT to set a breakpoint on a dummy routine when the
>Monitor was loaded.  While the system was running, I would use
>FILDDT to patch some instruction to jump to the dummy routine under
>certain conditions.  When the system hit that condition, the Monitor
>would come to a screeching halt, with EDDT running on the console.

A screeching halt or an abrupt halt?  :-)))

<snip further DDT marvelous design comments>

/BAH

Subtract a hundred and four for e-mail.

######

From: blarsen@infostream.no (Bjørn Hell Larsen)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 04 May 1999 11:04:13 +0200
Lines: 20
Message-ID: <u6769mcj6.fsf@infostream.no>
References: <372d88f7$0$490@news.zetnet.co.uk> <7gkgdq$pj9$5@antiochus.ultra.net> <2IsX2.28538$tY1.18718@wbnws01.ne.mediaone.net>
NNTP-Posting-Host: aubert.svg.infostream.no
X-Newsreader: Gnus v5.5/Emacs 20.3
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!newsfeed1.online.no!newsfeed.online.no!Norway.EU.net!not-for-mail


werme@werme.ne.mediaone.net (Ric Werme) writes:

> After I found Ed Taft's PDP-10 I/O package he wrote at Harvard, I started
> rewriting some of my BLISS code in MACRO-10.  With I/O hassles solved, fast
> compile times, and a wonder macro language, why use BLISS?
> 
> I later named the package Tulip, it's on a DECUS tape.  Need to write a WWW
> page for it someday.

You wrote Tulip? In that case, here's a fan letter. I used stuff in Tulip
regularly for several years, it was a great library.

The documentation was great, especially the section that explained the
coroutine support in Tulip. One of the more hilarious documents I've
ever read.


Bjorn

######

From: benh@lsl.co.uk (Ben Hutchings)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Followup-To: alt.folklore.computers,alt.sys.pdp10
Date: 4 May 1999 11:30:48 GMT
Organization: Laser-Scan Ltd.
Message-ID: <7gmlp8$h6j@relay.lsl.co.uk>
References: <372c6419$0$496@news.zetnet.co.uk>
NNTP-Posting-Host: 192.9.200.110
X-Newsreader: TIN [version 1.2 PL2]
Lines: 22
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!dispose.news.demon.net!demon!newspeer.clara.net!news.clara.net!peernews!tank.news.pipex.net!pipex!join.news.pipex.net!pipex!warm.news.pipex.net!pipex!news.lsl.co.uk!benh

lisard@zetnet.co.uk wrote:
<snip>
: And this is what the global vector is; it's a long vector in which each
: element is associated with a name. You make the association yourself, in
: a header file somewhere, and the onus is on you to avoid scribbling over
: other functions. It's an interesting way of avoiding the requirement to
: have a linker. From reading about Amiga Exec, it seems that the address
: of the Amiga's global vector is held at absolute address 4.

That's a pointer to ExecBase, not the global vector.  The global vector
is held in DOSBase->dl_GV and is passed into each new process in a
register (possibly d2).  The Amiga was never really meant to run BCPL;
all that stuff (DOS, file-system, console-handler, CLI and command
programs) was hurriedly fitted at a late stage when it became clear
that the original Unix-like high-level system software was not going to
be finished.  That is why Amiga programmers hate BCPL.  All BCPL code
was replaced by release 2.0 (1991) of AmigaOS, but BCPL conventions are
set in stone in much of the workings of AmigaDOS - and BCPL support
remains so that old versions of command programs will still run.

--
Any opinions expressed are my own and not necessarily those of Laser-Scan.

######

From: "Alan T. Bowler" <atbowler@thinkage.on.ca>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 04 May 1999 11:45:52 -0400
Organization: UUNET Canada News Transport
Lines: 27
Message-ID: <372F1630.31A00BAD@thinkage.on.ca>
References: <371be3e4.0@newsfeed.one.net> <7gcr9l$hn5@weyl.math.psu.edu> <3729fa95.0@news.wizvax.net> <un1zm1fv8.fsf@graphics.cornell.edu> <7gml4t$s3v$1@shell3.ba.best.com>
NNTP-Posting-Host: 192.102.11.4
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.5 [en] (WinNT; U)
X-Accept-Language: en
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!sunqbc.risq.qc.ca!news.uunet.ca!not-for-mail

Joe Smith wrote:

> Yep.  Mandatory file locking.  No one, not even [1,2] (the equivalent of
> 'root') could write to a file that was currently open for writing.  The only
> exception was when owner of the write lock performed a special UUO to set the
> file to simultaneous update mode.
> 
> One paper I saw on the development of Unix stated that they chose not
> to implement mandatory file locking because then 'root' would not be
> all powerful.  Hooey.

The explaination I saw did not make any such claim.  Just that
the designers had experience with such locks and did not feel
they were needed for the style of system they were building.
The locks get in the way sometimes, and don't always prevent
the common problems they are advertised to.  There known techniques
for optional advisory locking that work adequately in the type
of co-operative atmosphere envisioned at the time for Unis system
design.

The designers certainly had experience with such locks in Gcos.
I have occasionally wondered if the Unix design decision was not
partially motivated by the tendency of some stock Gcos software
to lock inappropriately.  For example, at the time Unix was evolving
a Gcos timesharing list file command would write lock the file
and leave it locked after the command terminated until the user
issued another command to unbusy the file.

######

From: Mark Crispin <mrc@CAC.Washington.EDU>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 4 May 1999 12:28:40 -0700
Organization: Networks & Distributed Computing
Lines: 17
Message-ID: <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU>
References: <371be3e4.0@newsfeed.one.net> <7gcr9l$hn5@weyl.math.psu.edu> <3729fa95.0@news.wizvax.net> <un1zm1fv8.fsf@graphics.cornell.edu> <7gml4t$s3v$1@shell3.ba.best.com>
NNTP-Posting-Host: tomobiki-cho.cac.washington.edu
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Trace: nntp1.u.washington.edu 925846123 31308 (None) 140.142.17.40
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: brit
In-Reply-To: <7gml4t$s3v$1@shell3.ba.best.com>
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news-fra.pop.de!newsfeed.tli.de!newsfeed.berkeley.edu!logbridge.uoregon.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc

On 4 May 1999, Joe Smith wrote:
> One paper I saw on the development of Unix stated that they chose not
> to implement mandatory file locking because then 'root' would not be
> all powerful.  Hooey.

Yup, and it's one of the biggest reasons that UNIX sucks.  Because of
this, it is much more complicated to do database semantics on UNIX.
Sheesh, even NT gets this right.

By the way, TOPS-20 had both modes available to the programmer.

-- Mark --

* RCW 19.190 notice: This email address is located in Washington State.	*
* Unsolicited commercial email may be billed $500 per message.		*
Science does not emerge from voting, party politics, or public debate.

######

From: David Scheidt <dscheidt@enteract.com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 14:28:36 GMT
Organization: EnterAct L.L.C. Turbo-Elite News Server
Lines: 17
Message-ID: <7gn06k$7hn$1@eve.enteract.com>
References: <371be3e4.0@newsfeed.one.net> <7gjhrh$lrk@weyl.math.psu.edu> <7gmjn5$lo2$1@shell3.ba.best.com> <7gmko2$aq5$3@antiochus.ultra.net> <7gmp33$p17$1@shell3.ba.best.com>
NNTP-Posting-Host: 207.229.143.40
User-Agent: tin/pre-1.4-19990216 ("Styrofoam") (UNIX) (FreeBSD/3.1-STABLE (i386))
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.enteract.com!news.enteract.com!not-for-mail

In alt.folklore.computers Joe Smith <inwap@best.com> wrote:

: The only thing I've seen vaguely like this function is 'ext2ed' on Linux.
: 	-Joe

That's because the linux "filesystem" doesn't guarantee filesystem consistancy
following a crash.  Lots of things that you would like to consider
invariants aren't.  Little things like metadata writes being done before
data writes; your files containing your data, and not some random other
processes; your data not being in someone else's file, are not necessarily
true.  You need something like that.  

David
-- 
dscheidt@enteract.com
Ketchup, therefore, shows both thixotropic and pseudoplastic rheological
properties.   -- John Schmitt, in AFU

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 15:45:45 -0400
Organization: -ENOENT
Lines: 18
Message-ID: <7gnip9$ok2@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <un1zm1fv8.fsf@graphics.cornell.edu> <7gml4t$s3v$1@shell3.ba.best.com> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!howland.erols.net!newsxfer.itd.umich.edu.MISMATCH!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU>,
Mark Crispin  <mrc@CAC.Washington.EDU> wrote:
>On 4 May 1999, Joe Smith wrote:
>> One paper I saw on the development of Unix stated that they chose not
>> to implement mandatory file locking because then 'root' would not be
>> all powerful.  Hooey.
>
>Yup, and it's one of the biggest reasons that UNIX sucks.  Because of
>this, it is much more complicated to do database semantics on UNIX.
>Sheesh, even NT gets this right.
>
>By the way, TOPS-20 had both modes available to the programmer.

So does BSD (and derivatives). And Linux. And Solaris. Your point being? v7?

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 16:48:42 -0700
Organization: Chez Inwap
Message-ID: <7go10q$pq6$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gml4t$s3v$1@shell3.ba.best.com> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu>
Lines: 27
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925861731 200 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!masternews.telia.net!newsfeed1.swip.net!swipnet!feed1.news.rcn.net!rcn!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gnip9$ok2@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>In article <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU>,
>Mark Crispin  <mrc@CAC.Washington.EDU> wrote:
>>On 4 May 1999, Joe Smith wrote:
>>> One paper I saw on the development of Unix stated that they chose not
>>> to implement mandatory file locking because then 'root' would not be
>>> all powerful.  Hooey.
>>
>>Yup, and it's one of the biggest reasons that UNIX sucks.  Because of
>>this, it is much more complicated to do database semantics on UNIX.
>>Sheesh, even NT gets this right.
>>
>>By the way, TOPS-20 had both modes available to the programmer.
>
>So does BSD (and derivatives). And Linux. And Solaris. Your point being? v7?

The two modes we are discussing here are "mandatory file locks" (aka "single
process update") and "simultaneous update".  BSD has advisory file locks, not
mandatory ones.  Any process that has write permissions and choses not
to call flock()/lockf()/fcntl() can merrily override an exclusive lock.
	-Joe

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: westin*nospam@graphics.cornell.edu (Stephen H. Westin)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 04 May 1999 17:22:01 -0400
Organization: Cornell University Program of Computer Graphics
Lines: 27
Sender: westin@DIESEL
Message-ID: <u90b41qfa.fsf@graphics.cornell.edu>
References: <371be3e4.0@newsfeed.one.net> <un1zm1fv8.fsf@graphics.cornell.edu> <7gml4t$s3v$1@shell3.ba.best.com> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu>
NNTP-Posting-Host: diesel.graphics.cornell.edu
X-Newsreader: Gnus v5.5/Emacs 20.3
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newspump.monmouth.com!newspeer.monmouth.com!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news.graphics.cornell.edu!news

viro@weyl.math.psu.edu (Alexander Viro) writes:

> In article <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU>,
> Mark Crispin  <mrc@CAC.Washington.EDU> wrote:
> >On 4 May 1999, Joe Smith wrote:
> >> One paper I saw on the development of Unix stated that they chose not
> >> to implement mandatory file locking because then 'root' would not be
> >> all powerful.  Hooey.

> >Yup, and it's one of the biggest reasons that UNIX sucks.  Because of
> >this, it is much more complicated to do database semantics on UNIX.
> >Sheesh, even NT gets this right.

> >By the way, TOPS-20 had both modes available to the programmer.

> So does BSD (and derivatives). And Linux. And Solaris. Your point
> being? v7?

I'm afraid I'm naive about this. Are you saying that if I fopen a file
for write, other processes are blocked from doing the same? I'm pretty
sure that's not true in HP-UX or IRIX. The equivalent in Apollo AEGIS
did that by default; I suspect that TOPS-20 did something similar.

-- 
-Stephen H. Westin
Any information or opinions in this message are mine: they do not
represent the position of Cornell University or any of its sponsors.

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 17:43:38 -0400
Organization: -ENOENT
Lines: 16
Message-ID: <7gnpma$oqf@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu> <u90b41qfa.fsf@graphics.cornell.edu>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!sunqbc.risq.qc.ca!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <u90b41qfa.fsf@graphics.cornell.edu>,
Stephen H. Westin <westin*nospam@graphics.cornell.edu> wrote:
>> So does BSD (and derivatives). And Linux. And Solaris. Your point
>> being? v7?
>
>I'm afraid I'm naive about this. Are you saying that if I fopen a file
>for write, other processes are blocked from doing the same? I'm pretty
>sure that's not true in HP-UX or IRIX. The equivalent in Apollo AEGIS
>did that by default; I suspect that TOPS-20 did something similar.

man flock. man 2 chmod. Choice between advisory and mandatory locking is
determined by permissions on a file in question.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 18:14:22 -0700
Organization: Chez Inwap
Message-ID: <7go61e$l6r$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gml4t$s3v$1@shell3.ba.best.com> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu>
Lines: 31
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925866867 219 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news.algonet.se!algonet!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gnip9$ok2@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>In article <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU>,
>Mark Crispin  <mrc@CAC.Washington.EDU> wrote:
>>On 4 May 1999, Joe Smith wrote:
>>> One paper I saw on the development of Unix stated that they chose not
>>> to implement mandatory file locking because then 'root' would not be
>>> all powerful.  Hooey.
>>
>>Yup, and it's one of the biggest reasons that UNIX sucks.  Because of
>>this, it is much more complicated to do database semantics on UNIX.
>>Sheesh, even NT gets this right.
>>
>>By the way, TOPS-20 had both modes available to the programmer.
>
>So does BSD (and derivatives). And Linux. And Solaris. Your point being? v7?

Oh, really?  Let me look....

Found in man pages for Solaris for chmod and lockf.  (I did not know that.)
Poorly documented in SunOS-4.1.x under lockf but not chmod.
Not in the man pages for FreeBSD 2.2.8-STABLE.
Not in the man pages for Linux 2.0.36 (just a comment that says to see
     /usr/src/linux-2.0.36/Documentation/mandatory.txt).

Rhetorical question: what good is a new feature if it is not documented?
	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <BlpT2.3$MF5.1852@srvr1.engin.umich.edu> <7fmv2b$591$2@antiochus.ultra.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu> <371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net> <Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu> <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net> <7fri4v$3eq@weyl.math.psu.edu> <8LoU2.21349$tY1.13078@wbnws01.ne.mediaone.net> <87hfq5dd5j.fsf@banet.net> <Pine.LNX.4.10.9904242321320.23119-100000@solo.hib.no> <1dqubne.nwt4wdxd4cgN@n242-114.berlin.snafu.de>
From: greg andruk <meowing@banet.net>
Date: 04 May 1999 18:25:31 -0400
Message-ID: <87emkwwjz8.fsf@slip-32-100-56-204.ma.us.ibm.net>
Lines: 10
X-Newsreader: Gnus v5.7/XEmacs 20.4 - "Emerald"
Cancel-Lock: sha1:VqFRz2T8wteSfPzrEkIM7k75g9w=
NNTP-Posting-Host: 32.100.56.77
X-Trace: 4 May 1999 23:28:38 GMT, 32.100.56.77
Organization: IBM Global Services - Remote Access Mail & News Services
X-Notice: Items posted that violate the IBM.NET Acceptable Use Policy
X-Notice: should be reported to postmaster@ibm.net
X-Complaints-To: postmaster@ibm.net
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.us.ibm.net!ibm.net!news3.us.ibm.net!32.100.56.77

Juergen Nickelsen <jnickelsen@acm.org> writes:

> Screen has originally been written 1987 by Oliver Laumann at the
> Technical Universiy of Berlin (Germany) under 4.2 or 4.3 BSD.

> At some point (after the 2.1 release from 1991, I think), someone in .ca
> or the US took over the maintenace of Screen;

.ca.  It was Wayne Davison, who still handles trn's care and feeding.

-- 
Misc. Meowing: <URL:http://members.xoom.com/meowing/> **meow**

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 4 May 1999 20:15:33 -0400
Organization: -ENOENT
Lines: 16
Message-ID: <7go2j5$p52@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu> <7go10q$pq6$1@shell3.ba.best.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newspeer.clara.net!news.clara.net!newsfeed.nacamar.de!news-peer.gip.net!news.gsl.net!gip.net!cliffs.rs.itd.umich.edu!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7go10q$pq6$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>In article <7gnip9$ok2@weyl.math.psu.edu>,
>Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>So does BSD (and derivatives). And Linux. And Solaris. Your point being? v7?
>
>The two modes we are discussing here are "mandatory file locks" (aka "single
>process update") and "simultaneous update".  BSD has advisory file locks, not
>mandatory ones.  Any process that has write permissions and choses not
>to call flock()/lockf()/fcntl() can merrily override an exclusive lock.

Oops. Should not post when low on caffeine... On Solaris and Linux it is done
with setting SGID and *not* setting executable for group. Sorry.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: <root@mauve.demon.co.uk>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 04 May 1999 21:55:44 GMT
Message-ID: <925854944.29414.0.nnrp-13.9e98d142@news.demon.co.uk>
References: <371be3e4.0@newsfeed.one.net> <7g47sv$rg5$2@antiochus.ultra.net> <7g4fkm$j34@bonkers.taronga.com> <3726BC51.E80F0C3D@tnglwood.demon.co.uk> <3729c2b2.0@news.wizvax.net> <372AB10D.3F5A3E51@tnglwood.demon.co.uk>
NNTP-Posting-Host: mauve.demon.co.uk
X-NNTP-Posting-Host: mauve.demon.co.uk:158.152.209.66
X-Trace: news.demon.co.uk 925854944 nnrp-13:29414 NO-IDENT mauve.demon.co.uk:158.152.209.66
X-Complaints-To: abuse@demon.net
User-Agent: tin/pre-1.4-980818 ("Laura") (UNIX) (Linux/2.2.5 (i586))
Originator: root@mauve.demon.co.uk
Lines: 14
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!mauve.demon.co.uk!root

Robert Billing <unclebob@tnglwood.demon.co.uk> wrote:
: John Wilson wrote:

:> Of course, in this case the kernel has to have the RAM disk driver and
:> its file system compiled in.  Both can be pretty primitive, but still...

:  Yes. The ramdisk driver is of course minimal. The file system is the
: ext2 file system that it will need anyway when it is up and running.

Not always.
I've for example done romfs, which is much smaller than the ext2 module.
This way, a floppy can contain stuff to probe for a network interface,
then can pull other things (NFS, ext2 filesystem, fat filesystem, ...)
over that network device, so you can do what you want with it later.

######

From: Mark Crispin <MRC@CAC.Washington.EDU>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 4 May 1999 23:27:56 -0700
Organization: University of Washington
Lines: 17
Sender: MRC@shivams.cac.washington.edu
Message-ID: <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC>
References: <371be3e4.0@newsfeed.one.net> <un1zm1fv8.fsf@graphics.cornell.edu> <7gml4t$s3v$1@shell3.ba.best.com> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu>
NNTP-Posting-Host: cs60-6.cacmodem.washington.edu
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Trace: nntp1.u.washington.edu 925885672 32964 (None) 140.142.17.39
X-Complaints-To: help@cac.washington.edu
To: Alexander Viro <viro@weyl.math.psu.edu>
In-Reply-To: <7gnip9$ok2@weyl.math.psu.edu>
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!nuq-peer.news.verio.net!sea-feed.news.verio.net!feed.news.verio.net!netnews1.nw.verio.net!netnews.nwnet.net!news.verio.net!news.u.washington.edu!cs60-6.cacmodem.washington.edu!MRC

On 4 May 1999, Alexander Viro wrote:
> >By the way, TOPS-20 had both modes available to the programmer.
> So does BSD (and derivatives). And Linux. And Solaris. Your point being? v7?

Wrong.  BSD, Linux, and Solaris do *NOT* have mandatory locking.

Mandatory locking is an open mode that prevents any other programs from
opening the file.  It also prevents opening if any other program has the
file open.

This is different from advisory locking via flock() or fcntl().

You need both forms of locking.

You also need both forms of open ("frozen" using mandatory locking, and
"thawed" which is UNIX-style).

######

From: jnickelsen@acm.org (Juergen Nickelsen)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 4 May 1999 23:37:48 +0200
Organization: [Posted via] Interactive Networx
Lines: 11
Message-ID: <1dquae3.115a04dottw2aN@n242-114.berlin.snafu.de>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <BlpT2.3$MF5.1852@srvr1.engin.umich.edu> <7fmv2b$591$2@antiochus.ultra.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu> <371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net> <Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu> <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net> <7fri4v$3eq@weyl.math.psu.edu> <8LoU2.21349$tY1.13078@wbnws01.ne.mediaone.net>
NNTP-Posting-Host: n242-114.berlin.snafu.de
User-Agent: MacSOUP/2.4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!unlisys!news.snafu.de!jnickelsen

Ric Werme <werme@werme.ne.mediaone.net> wrote:

>    $ man screen
>    No manual entry for screen

"screen" has always been an add-on with Unix. While there may be
Unix(-like) system shipping with screen, this is not the norm. And, BTW,
screen is older than Linux.

-- 
Juergen Nickelsen

######

From: jnickelsen@acm.org (Juergen Nickelsen)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 4 May 1999 23:37:50 +0200
Organization: [Posted via] Interactive Networx
Lines: 23
Message-ID: <1dqubne.nwt4wdxd4cgN@n242-114.berlin.snafu.de>
References: <371be3e4.0@newsfeed.one.net> <8A3T2.192$W4.62532@srvr1.engin.umich.edu> <FAI5rt.72p@midway.uchicago.edu> <Rf5T2.194$W4.62887@srvr1.engin.umich.edu> <7fk9i4$spj$1@ligarius.ultra.net> <BlpT2.3$MF5.1852@srvr1.engin.umich.edu> <7fmv2b$591$2@antiochus.ultra.net> <ZSIT2.14$MF5.5706@srvr1.engin.umich.edu> <371F79F1.69EB3101@stoneweb.com> <7fpjgn$v7t$4@antiochus.ultra.net> <Ci0U2.24$MF5.9024@srvr1.engin.umich.edu> <ALDERSON.99Apr23113531@netcom.netcom.com> <7frcok$3ck@weyl.math.psu.edu> <V4bU2.21165$tY1.12930@wbnws01.ne.mediaone.net> <7fri4v$3eq@weyl.math.psu.edu> <8LoU2.21349$tY1.13078@wbnws01.ne.mediaone.net> <87hfq5dd5j.fsf@banet.net> <Pine.LNX.4.10.9904242321320.23119-100000@solo.hib.no>
NNTP-Posting-Host: n242-114.berlin.snafu.de
User-Agent: MacSOUP/2.4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!unlisys!news.snafu.de!jnickelsen

Eivind <db96ek@oliven.hib.no> wrote:

[screen]
> Does anyone happen to know the origin of this nifty little program ? It
> occurs to me that a terminal multiplexer must've been a really useful
> thing to have around for a rather long time.

Screen has originally been written 1987 by Oliver Laumann at the
Technical Universiy of Berlin (Germany) under 4.2 or 4.3 BSD.

At some point (after the 2.1 release from 1991, I think), someone in .ca
or the US took over the maintenace of Screen; and some time later
Juergen Weigert (Univ. Erlangen, Germany) took over. I know Oliver a
bit, and I guess he just didn't want to support the loads of features
that Screen's users demanded. His 2.1 release consisted of seven files
and was 133 KB in size; the current version 3.7.6 has 83 files and
nearly 1.5 MB.

Oliver Lauman also wrote Elk ("Extension Language Kit"), an embeddable
and extendable Scheme interpreter.

-- 
Juergen Nickelsen

######

From: tsmurphy@students.uiuc.edu (Terry Murphy)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 5 May 1999 00:41:06 GMT
Organization: University of Illinois at Urbana-Champaign
Lines: 16
Message-ID: <7go432$ekt$1@vixen.cso.uiuc.edu>
References: <371be3e4.0@newsfeed.one.net> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu> <u90b41qfa.fsf@graphics.cornell.edu>
NNTP-Posting-Host: ux8.cso.uiuc.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!vixen.cso.uiuc.edu!tsmurphy

In article <u90b41qfa.fsf@graphics.cornell.edu>,
Stephen H. Westin <westin*nospam@graphics.cornell.edu> wrote:

>I'm afraid I'm naive about this. Are you saying that if I fopen a file
>for write, other processes are blocked from doing the same? I'm pretty
>sure that's not true in HP-UX or IRIX. The equivalent in Apollo AEGIS
>did that by default; I suspect that TOPS-20 did something similar.

Another problem related to this is that it seems that it is customary
under Unix to open the file only for the actual reading of the file,
rather than for the duration that the file is open (from the users'
standpoint).  For example, in text editors under Unix, it seems that
the file is read and then immediately closed, and when it is saved it
is open/written/closed again.  Very annoying, of course.

-- Terry

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
From: ehrice@his.com (Edward Rice)
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Message-ID: <B35550DE9668E4D05@pm2m-90.his.com>
Organization: NDS
Date: Wed, 05 May 1999 01:35:26 -0400
References: <371be3e4.0@newsfeed.one.net> <7gcr9l$hn5@weyl.math.psu.edu> <3729fa95.0@news.wizvax.net> <un1zm1fv8.fsf@graphics.cornell.edu> <7gml4t$s3v$1@shell3.ba.best.com> <372F1630.31A00BAD@thinkage.on.ca>
NNTP-Posting-Host: pm2m-90.his.com
Lines: 38
X-Authenticated-User: ehrice
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feeder.qis.net!news-xfer.newsread.com!netaxs.com!newsread.com!news4.his.com!user

In article <372F1630.31A00BAD@thinkage.on.ca>,
"Alan T. Bowler" <atbowler@thinkage.on.ca> wrote:

 > The [UNIX] designers certainly had experience with such locks in Gcos.

I'd appreciate feedback from Dennis Ritchie as to whether the above
statement is really accurate.  The experience with GCOS existed, but I'm
not sure how much of that experience was in the UNIX development "group,"
such as it initially was.  I would expect there to be a great deal of
experience, but more with CTSS and IBSYS than with GECOS, at that
particular time.

 > I have occasionally wondered if the Unix design decision was not
 > partially motivated by the tendency of some stock Gcos software
 > to lock inappropriately.  For example, at the time Unix was evolving
 > a Gcos timesharing list file command would write lock the file
 > and leave it locked after the command terminated until the user
 > issued another command to unbusy the file.

This was meaningful and not totally unreasonable, though.  And, before you
get upset with this defense, it bit me in the ass at least as often as it
did you, Alan -- I can't count how many times I swore at that "feature." 
But, when you listed (in the sense of printing-to-terminal) a file under
GCOS TSS, you had an image in front of you of what that file looked like. 
The system was making sure that so long as you had that image, and took no
other action, the file would continue to look like that.  So, if you did an
OLD on a file, you owned read and write permissions on it until you did a
REMOVE (or started a fresh session), and if you did a LIST, you owned the
write permission on it so that nobody else could update it while you looked
at it.  As I recall, there was a little-used feature that permitted a user
to decline acquisition of write permission on the file, by doing a "LIST
foo,R" or even "OLD foo,R" -- that gave you read permission, and it became
a question of whether the file system would permit a write at that point. 
(In early versions, it would not; later on, with more complex file
permissions, it could.)  It was an irritating feature, but it wasn't a
meaningless decision to do it that way.

######

Sender: eric@ruckus.brouhaha.com
From: Eric Smith <eric-no-spam-for-me@brouhaha.com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7gml4t$s3v$1@shell3.ba.best.com> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu> <7go61e$l6r$1@shell3.ba.best.com>
X-Disclaimer: Everything I write is false.
Organization: Eric Conspiracy Secret Labs
X-Eric-Conspiracy: There is no conspiracy.
Date: 05 May 1999 02:56:42 -0700
Message-ID: <qhpv4f4z6t.fsf@ruckus.brouhaha.com>
Lines: 16
X-Newsreader: Gnus v5.5/Emacs 20.2
NNTP-Posting-Host: ruckus.brouhaha.com
X-Trace: 5 May 1999 02:01:15 -0700, ruckus.brouhaha.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!news.sgi.com!news.spies.com!ruckus.brouhaha.com

inwap@best.com (Joe Smith) writes about the apparently lack of mandatory
file locks in Unix variants:
> Oh, really?  Let me look....
> Not in the man pages for Linux 2.0.36 (just a comment that says to see
>      /usr/src/linux-2.0.36/Documentation/mandatory.txt).

And that comment wasn't sufficient to suggest to you that Linux may in
fact support mandatory file locks?

The man page for fcntl(2) is admittedly rather cryptic, but it does
document the mandatory lock facilities.  In particular, note the return
value "EACCES  Operation is prohibited by locks held by other processes."

Although POSIX.1 doesn't define mandatory locking, it is part of the SVID,
so any Unix variants derived from System V (such as Solaris) presumably
support it.

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <372d88f7$0$490@news.zetnet.co.uk> <7gkgdq$pj9$5@antiochus.ultra.net>
     <2IsX2.28538$tY1.18718@wbnws01.ne.mediaone.net> <u6769mcj6.fsf@infostream.no>
X-Newsreader: NN version 6.5.0 CURRENT #119
From: werme@werme.ne.mediaone.net (Ric Werme)
Lines: 54
Message-ID: <vvOX2.1222$kE6.916@wbnws01.ne.mediaone.net>
Date: Wed, 05 May 1999 03:00:43 GMT
NNTP-Posting-Host: 24.128.109.10
X-Complaints-To: abuse@rr.com
X-Trace: wbnws01.ne.mediaone.net 925873243 24.128.109.10 (Tue, 04 May 1999 23:00:43 EDT)
NNTP-Posting-Date: Tue, 04 May 1999 23:00:43 EDT
Organization: Road Runner
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!masternews.telia.net!news.algonet.se!algonet!news.maxwell.syr.edu!newsfeed.cwix.com!24.128.1.125!chnws03.mediaone.net!24.128.1.101!chnws05.ne.mediaone.net!24.128.44.7!wbnws01.ne.mediaone.net.POSTED!not-for-mail

blarsen@infostream.no writes:


>werme@werme.ne.mediaone.net (Ric Werme) writes:

>> After I found Ed Taft's PDP-10 I/O package he wrote at Harvard, I started
>> rewriting some of my BLISS code in MACRO-10.  With I/O hassles solved, fast
>> compile times, and a wonderful macro language, why use BLISS?

>> I later named the package Tulip, it's on a DECUS tape.  Need to write a WWW
>> page for it someday.

>You wrote Tulip? In that case, here's a fan letter. I used stuff in Tulip
>regularly for several years, it was a great library.

I wrote just the documentation!  Well, I think I added four
instructions to the parser.  Ed didn't write any documentation, so I
figured I should, if only to repay him for the time it saved me.

>The documentation was great, especially the section that explained the
>coroutine support in Tulip. One of the more hilarious documents I've
>ever read.

Also one of the first pieces of prose I _wanted_ to write, or at least
to write well.  One thing I learned from it was a) boy, do a lot of
typos slip in, and b) after fixing everything I could find I was now a
great proofreader.  The latter can be more hindrance than help -
beyond a certain % of misspellings in a USENET post, I hit the n key,
no matter how interesting the post appeared.  I can spot typos in
things I'm not reading.  When I started typing this I noticed I had
written "wonder macro language".  A few months ago I was reading one
of Bob Metcalfe's columns in the back of Infoworld about networking in
schools.  I was the first to point out to him that he had written
"pubic education".  I don't remember anything else about the column,
it may not have been on networking, but the school system itself.

Yeah, I had fun writing it.  Let's see, I had a couple challenges in it.
One was what was the item in the index that conveyed no useful information.
That was the start of the index.  There was also something about that
coroutine code.  I think the second coroutine did a jump to the JSP
in the first to save an instruction.  (You really have to see that on
paper for it to make any sense.)

I do have the documentation and sample programs.  I will get them on
the WWW, but not for a while.

I do have a new page on the WWW you might enjoy -
http://people.ne.mediaone.net/werme/deimos.html

	-Ric Werme
--
Ric Werme                  | http://people.ne.mediaone.net/werme
werme@nospam.mediaone.net  | http://www.cyberportal.net/werme
      ^^^^^^^ delete

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <un1zm1fv8.fsf@graphics.cornell.edu> <7gml4t$s3v$1@shell3.ba.best.com> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC>
From: greg andruk <meowing@banet.net>
Date: 05 May 1999 04:09:19 -0400
Message-ID: <87pv4g9bv4.fsf@slip-32-101-160-211.ma.us.ibm.net>
Lines: 16
X-Newsreader: Gnus v5.7/XEmacs 20.4 - "Emerald"
Cancel-Lock: sha1:zjtLB/c10nfwqYu5jWXhMCXjP6A=
NNTP-Posting-Host: 32.101.27.54
X-Trace: 5 May 1999 08:09:30 GMT, 32.101.27.54
Organization: IBM Global Services - Remote Access Mail & News Services
X-Notice: Items posted that violate the IBM.NET Acceptable Use Policy
X-Notice: should be reported to postmaster@ibm.net
X-Complaints-To: postmaster@ibm.net
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!news-peer.gip.net!news.gsl.net!gip.net!newsfeed2.us.ibm.net!ibm.net!news3.us.ibm.net!32.101.27.54

In alt.folklore.computers, Mark Crispin <MRC@CAC.Washington.EDU> writes:

> On 4 May 1999, Alexander Viro wrote:
>> >By the way, TOPS-20 had both modes available to the programmer.
>> So does BSD (and derivatives). And Linux. And Solaris. Your point being? v7?

> Wrong.  BSD, Linux, and Solaris do *NOT* have mandatory locking.

Your info is out of date.  Solaris, HPUX and linux do support
mandatory locks.  Not sure about the state of the various BSDen.

In the linux kernel sources, see mandatory.txt and locks.txt under
Documentation/ for details (it also touches on how the other Unices
implement it).


-- 
Misc. Meowing: <URL:http://members.xoom.com/meowing/> **meow**

######

From: westin*nospam@graphics.cornell.edu (Stephen H. Westin)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 05 May 1999 08:20:52 -0400
Organization: Cornell University Program of Computer Graphics
Lines: 51
Sender: westin@DIESEL
Message-ID: <u7lqn1zdn.fsf@graphics.cornell.edu>
References: <371be3e4.0@newsfeed.one.net> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu> <u90b41qfa.fsf@graphics.cornell.edu> <7gnpma$oqf@weyl.math.psu.edu>
NNTP-Posting-Host: diesel.graphics.cornell.edu
X-Newsreader: Gnus v5.5/Emacs 20.3
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cliffs.rs.itd.umich.edu!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news.graphics.cornell.edu!news

viro@weyl.math.psu.edu (Alexander Viro) writes:

> In article <u90b41qfa.fsf@graphics.cornell.edu>,
> Stephen H. Westin <westin*nospam@graphics.cornell.edu> wrote:
> >> So does BSD (and derivatives). And Linux. And Solaris. Your point
> >> being? v7?
> >
> >I'm afraid I'm naive about this. Are you saying that if I fopen a file
> >for write, other processes are blocked from doing the same? I'm pretty
> >sure that's not true in HP-UX or IRIX. The equivalent in Apollo AEGIS
> >did that by default; I suspect that TOPS-20 did something similar.
> 
> man flock. man 2 chmod.

dick-westin % uname -a
HP-UX dick B.10.20 A 9000/770 2003270344 two-user license
dick-westin % man flock
No manual entry for flock.

So HP-UX, the Unix from what I believe is the largest Unix vendor (in
dollar volume), doesn't have it.

IRIX (6.4) seems to have it. Except "Unlike BSD, child processes
created by fork(2) do not inherit references to locks acquired by
their parents through flock(3B) calls.  This bug results from flock's
implementation atop System V file and record locks."

So one of the two has *advisory* locks, i.e. something that's not
enforced by the system. As for mandatory locks, how can two
cooperating processes lock the same file for simultaneous write
access? I don't think there's any syntax in fopen() to allow
overriding the default locking.

> Choice between advisory and mandatory locking is
> determined by permissions on a file in question.

And how many programs will be broken when it's switched on? One of the
aggravating things about Unix is that even if a feature is added,
using it generally breaks legacy code. Not a problem with a system
that was built that way from the beginning.

Here's what happens in DOMAIN/OS: default behavior is one writer/many
readers: many processes can open the file to read, but only one to
write. It's perfectly possible for programs to override this behavior
explicitly, allowing many cooperating processes to write to the same
file.

-- 
-Stephen H. Westin
Any information or opinions in this message are mine: they do not
represent the position of Cornell University or any of its sponsors.

######

From: westin*nospam@graphics.cornell.edu (Stephen H. Westin)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 05 May 1999 08:29:32 -0400
Organization: Cornell University Program of Computer Graphics
Lines: 24
Sender: westin@DIESEL
Message-ID: <u67671yz7.fsf@graphics.cornell.edu>
References: <371be3e4.0@newsfeed.one.net> <un1zm1fv8.fsf@graphics.cornell.edu> <7gml4t$s3v$1@shell3.ba.best.com> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <87pv4g9bv4.fsf@slip-32-101-160-211.ma.us.ibm.net>
NNTP-Posting-Host: diesel.graphics.cornell.edu
X-Newsreader: Gnus v5.5/Emacs 20.3
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!cliffs.rs.itd.umich.edu!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news.graphics.cornell.edu!news

greg andruk <meowing@banet.net> writes:

> In alt.folklore.computers, Mark Crispin <MRC@CAC.Washington.EDU> writes:
> 
> > On 4 May 1999, Alexander Viro wrote:
> >> >By the way, TOPS-20 had both modes available to the programmer.
> >> So does BSD (and derivatives). And Linux. And Solaris. Your point being? v7?
> 
> > Wrong.  BSD, Linux, and Solaris do *NOT* have mandatory locking.
> 
> Your info is out of date.  Solaris, HPUX and linux do support
> mandatory locks.  Not sure about the state of the various BSDen.

This is a correction to an earlier post of mine. HP-UX seems to
support file locking as lockf() rather than flock(). This is an
example of one of the drawbacks to grafted-on capabilities: creeping
incompatibility. It's why ports between Unix systems are not trivial.

And mandatory locking seems to be optional on a per-file basis...

-- 
-Stephen H. Westin
Any information or opinions in this message are mine: they do not
represent the position of Cornell University or any of its sponsors.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 05 May 99 09:18:45 GMT
Organization: UltraNet Communications, Inc.
Lines: 37
Message-ID: <7gp8kq$5rc$1@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <7gml4t$s3v$1@shell3.ba.best.com> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu> <7go61e$l6r$1@shell3.ba.best.com>
NNTP-Posting-Host: d3.dial-14.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 5 May 1999 11:04:58 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!195.211.211.32.MISMATCH!newscore.gigabell.net!newscore.ipf.de!news0.de.colt.net!colt.net!news-lond.gip.net!news-peer.gip.net!news.gsl.net!gip.net!feeder.qis.net!feed1.news.rcn.net!rcn!news.ultranet.com!d3

In article <7go61e$l6r$1@shell3.ba.best.com>,
   inwap@best.com (Joe Smith) wrote:
>In article <7gnip9$ok2@weyl.math.psu.edu>,
>Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>In article 
<Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU>,
>>Mark Crispin  <mrc@CAC.Washington.EDU> wrote:
>>>On 4 May 1999, Joe Smith wrote:
>>>> One paper I saw on the development of Unix stated that they chose not
>>>> to implement mandatory file locking because then 'root' would not be
>>>> all powerful.  Hooey.
>>>
>>>Yup, and it's one of the biggest reasons that UNIX sucks.  Because of
>>>this, it is much more complicated to do database semantics on UNIX.
>>>Sheesh, even NT gets this right.
>>>
>>>By the way, TOPS-20 had both modes available to the programmer.
>>
>>So does BSD (and derivatives). And Linux. And Solaris. Your point being? 
v7?
>
>Oh, really?  Let me look....
>
>Found in man pages for Solaris for chmod and lockf.  (I did not know 
that.)
>Poorly documented in SunOS-4.1.x under lockf but not chmod.
>Not in the man pages for FreeBSD 2.2.8-STABLE.
>Not in the man pages for Linux 2.0.36 (just a comment that says to see
>     /usr/src/linux-2.0.36/Documentation/mandatory.txt).
>
>Rhetorical question: what good is a new feature if it is not documented?

Chuckle.  That was our opinion, too.  

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 05 May 99 09:25:37 GMT
Organization: UltraNet Communications, Inc.
Lines: 21
Message-ID: <7gp91m$5rc$2@antiochus.ultra.net>
References: <372d88f7$0$490@news.zetnet.co.uk> <7gkgdq$pj9$5@antiochus.ultra.net>
 <2IsX2.28538$tY1.18718@wbnws01.ne.mediaone.net> <u6769mcj6.fsf@infostream.no> <vvOX2.1222$kE6.916@wbnws01.ne.mediaone.net>
NNTP-Posting-Host: d3.dial-14.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 5 May 1999 11:11:50 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!news.ultranet.com!d3

In article <vvOX2.1222$kE6.916@wbnws01.ne.mediaone.net>,
   werme@werme.ne.mediaone.net (Ric Werme) wrote:

<snip Tulip story>

>I do have the documentation and sample programs.  I will get them on
>the WWW, but not for a while.

<snip>

I had that package shipped on the Customer Supported Tape, Ric.
Once Jim and I got that tape past the idiots who had to approve
everything, I put everything I could on that tape that I thought
would be useful to the customers and all the "old stuff" that
noone seemed to be interested in anymore.  That was one of our
ways to skin the "we have to charge for everything!" cat that
afflicted us beginning in the early 80s.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 05 May 99 09:26:57 GMT
Organization: UltraNet Communications, Inc.
Lines: 30
Message-ID: <7gp946$5rc$3@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <372b2c32.0@news.wizvax.net> <7gfg5m$jen@weyl.math.psu.edu> <7ggqll$u0$1@shell3.ba.best.com> <7gh5mj$kb3@weyl.math.psu.edu> <ulnf61fgi.fsf@graphics.cornell.edu>
NNTP-Posting-Host: d3.dial-14.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 5 May 1999 11:13:10 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.enteract.com!newspeer.monmouth.com!news.ultranet.com!d3

In article <ulnf61fgi.fsf@graphics.cornell.edu>,
   westin*nospam@graphics.cornell.edu (Stephen H. Westin) wrote:
>viro@weyl.math.psu.edu (Alexander Viro) writes:
>
><snip>
>
>> Another thing being: was there an equivalent of buffer cache? Darn 
useful
>> when you are doing a big compile... Since you had a big lock over write
>> access it shouldn't give any problems with cache consistency. IOW: how 
many
>> disk reads would happen if two processes had the same file opened for 
read
>> and both requested read on the same block? OTOH I got used to situation 
when
>> CPU is much faster than any IO, so relative costs might be completely
>> different...
>
>CPU has always been faster than I/O. At least since, say, about
>1955. Disks still have access times measured in milliseconds; if you
>connected a disk drive of today to a KA10, there would still be about
>3 orders of magnitude speed advantage to the CPU/core memory.

That wasn't our experience.  Sure they may have been faster on
paper, but the real test was when one hooked one up and ran
software.  

/BAH

Subtract a hundred and four for e-mail.

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
From: hoh@approve.se.NO_JUNK_EMAIL (Goran Larsson)
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Message-ID: <1999May5.163308.22256@lorelei.approve.se>
Originator: hoh@approve.se.NO_JUNK_EMAIL (Goran Larsson)
Sender: hoh@lorelei.approve.se.NO_JUNK_EMAIL (Goran Larsson)
Organization: [1]  + 5934 done        /bin/rm -rf ~/ &
X-No-Archive: yes
X-Newsreader: trn 4.0-test66 (4 June 1998)
References: <371be3e4.0@newsfeed.one.net> <7gml4t$s3v$1@shell3.ba.best.com> <372F1630.31A00BAD@thinkage.on.ca> <B35550DE9668E4D05@pm2m-90.his.com>
Lines: 17
Date: Wed, 5 May 1999 16:33:08 GMT
NNTP-Posting-Host: 130.244.169.6
X-Complaints-To: news-abuse@swip.net
X-Trace: nntpserver.swip.net 925922323 130.244.169.6 (Wed, 05 May 1999 18:38:43 MET DST)
NNTP-Posting-Date: Wed, 05 May 1999 18:38:43 MET DST
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!newsfeed1.swip.net!swipnet!nntpserver.swip.net!not-for-mail

In article <B35550DE9668E4D05@pm2m-90.his.com>,
Edward Rice <ehrice@his.com> wrote:

> But, when you listed (in the sense of printing-to-terminal) a file under
> GCOS TSS, you had an image in front of you of what that file looked like. 
> The system was making sure that so long as you had that image, and took no
> other action, the file would continue to look like that.

Does that extend to printed versions of the file as well?
Print the file on paper and the file will be locked forever,
or at least until you promise GCOS that you have destroyed
every copy of the printout?    :-)

-- 
 Goran Larsson            hoh AT approve DOT se
 I was an atheist,        http://home DOT swipnet DOT se/hoh/
 until I found out I was God.

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
From: hoh@approve.se.NO_JUNK_EMAIL (Goran Larsson)
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Message-ID: <1999May5.164445.22696@lorelei.approve.se>
Originator: hoh@approve.se.NO_JUNK_EMAIL (Goran Larsson)
Sender: hoh@lorelei.approve.se.NO_JUNK_EMAIL (Goran Larsson)
Organization: [1]  + 5934 done        /bin/rm -rf ~/ &
X-No-Archive: yes
X-Newsreader: trn 4.0-test66 (4 June 1998)
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <87pv4g9bv4.fsf@slip-32-101-160-211.ma.us.ibm.net> <u67671yz7.fsf@graphics.cornell.edu>
Lines: 21
Date: Wed, 5 May 1999 16:44:45 GMT
NNTP-Posting-Host: 130.244.76.26
X-Complaints-To: news-abuse@swip.net
X-Trace: nntpserver.swip.net 925922922 130.244.76.26 (Wed, 05 May 1999 18:48:42 MET DST)
NNTP-Posting-Date: Wed, 05 May 1999 18:48:42 MET DST
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news.sprintlink.net!newsfeed1.swip.net!swipnet!nntpserver.swip.net!not-for-mail

In article <u67671yz7.fsf@graphics.cornell.edu>,
Stephen H. Westin <westin*nospam@graphics.cornell.edu> wrote:

> This is a correction to an earlier post of mine. HP-UX seems to
> support file locking as lockf() rather than flock().

The standard (SVID2, SVID3 and XPG2) is lockf().
The old ancient, do not use, BSD function is flock().
I am not surprised at all that HP choose to implement
lockf() and not flock().

>                                                       This is an
> example of one of the drawbacks to grafted-on capabilities: creeping
> incompatibility. It's why ports between Unix systems are not trivial.

Use the standard lockf() and forget about flock().

-- 
 Goran Larsson            hoh AT approve DOT se
 I was an atheist,        http://home DOT swipnet DOT se/hoh/
 until I found out I was God.

######

From: Mark Crispin <mrc@CAC.Washington.EDU>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 5 May 1999 14:08:37 -0700
Organization: Networks & Distributed Computing
Lines: 29
Message-ID: <Pine.NXT.4.20.9905051403480.26122-100000@Tomobiki-Cho.CAC.Washington.EDU>
References: <371be3e4.0@newsfeed.one.net> <un1zm1fv8.fsf@graphics.cornell.edu> <7gml4t$s3v$1@shell3.ba.best.com> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <87pv4g9bv4.fsf@slip-32-101-160-211.ma.us.ibm.net>
NNTP-Posting-Host: tomobiki-cho.cac.washington.edu
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Trace: nntp1.u.washington.edu 925938519 31404 (None) 140.142.17.39
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: smj
To: greg andruk <meowing@banet.net>
In-Reply-To: <87pv4g9bv4.fsf@slip-32-101-160-211.ma.us.ibm.net>
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!news.maxwell.syr.edu!newsfeed.wli.net!newsfeed.berkeley.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc

On 5 May 1999, greg andruk wrote:
> > Wrong.  BSD, Linux, and Solaris do *NOT* have mandatory locking.
> Your info is out of date.  Solaris, HPUX and linux do support
> mandatory locks.  Not sure about the state of the various BSDen.

No, it does not.

Solaris, HPUX, and Linux have something that is called "mandatory
locking", but is not the mandatory locking in the context of this
discussion.

"Mandatory locking" in the context of this discussion is an open mode
which specifies that the file is to be opened in a "frozen" (vs.
UNIX-style "thawed") mode, and whether the "frozen" mode is shared or
exclusive.  This is *separate* and *independent* of advisory locking (e.g.
via flock() or fcntl() system calls).

Thawed opens preclude any frozen opens, and vice-versa.

Shared frozen opens preclude exclusive frozen opens.

Exclusive frozen opens preclude other frozen opens.

-- Mark --

* RCW 19.190 notice: This email address is located in Washington State.	*
* Unsolicited commercial email may be billed $500 per message.		*
Science does not emerge from voting, party politics, or public debate.

######

From: Mark Crispin <mrc@CAC.Washington.EDU>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 5 May 1999 14:25:31 -0700
Organization: Networks & Distributed Computing
Lines: 13
Message-ID: <Pine.NXT.4.20.9905051424160.26122-100000@Tomobiki-Cho.CAC.Washington.EDU>
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <87pv4g9bv4.fsf@slip-32-101-160-211.ma.us.ibm.net> <u67671yz7.fsf@graphics.cornell.edu> <1999May5.164445.22696@lorelei.approve.se>
NNTP-Posting-Host: tomobiki-cho.cac.washington.edu
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Trace: nntp1.u.washington.edu 925939536 27034 (None) 140.142.17.40
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: vsl
To: Goran Larsson <hoh@approve.se>
In-Reply-To: <1999May5.164445.22696@lorelei.approve.se>
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!isdnet!netnews.com!newsfeed.berkeley.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc

On Wed, 5 May 1999, Goran Larsson wrote:
> Use the standard lockf() and forget about flock().

You mean fcntl().

lockf() does not support shared locking.

-- Mark --

* RCW 19.190 notice: This email address is located in Washington State.	*
* Unsolicited commercial email may be billed $500 per message.		*
Science does not emerge from voting, party politics, or public debate.

######

From: Mark Crispin <mrc@CAC.Washington.EDU>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 5 May 1999 14:27:23 -0700
Organization: Networks & Distributed Computing
Lines: 22
Message-ID: <Pine.NXT.4.20.9905051426050.26122-100000@Tomobiki-Cho.CAC.Washington.EDU>
References: <371be3e4.0@newsfeed.one.net> <un1zm1fv8.fsf@graphics.cornell.edu> <7gml4t$s3v$1@shell3.ba.best.com> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <87pv4g9bv4.fsf@slip-32-101-160-211.ma.us.ibm.net> <Pine.NXT.4.20.9905051403480.26122-100000@Tomobiki-Cho.CAC.Washington.EDU>
NNTP-Posting-Host: tomobiki-cho.cac.washington.edu
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Trace: nntp1.u.washington.edu 925939646 31422 (None) 140.142.17.35
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: yuki
To: greg andruk <meowing@banet.net>
In-Reply-To: <Pine.NXT.4.20.9905051403480.26122-100000@Tomobiki-Cho.CAC.Washington.EDU>
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!isdnet!newspump.monmouth.com!newspeer.monmouth.com!newsfeed.berkeley.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc

I almost forgot.  There's another important mode; exclusive opens which
only interact against other exclusive opens but not against shared opens.

On Wed, 5 May 1999, Mark Crispin wrote:
> "Mandatory locking" in the context of this discussion is an open mode
> which specifies that the file is to be opened in a "frozen" (vs.
> UNIX-style "thawed") mode, and whether the "frozen" mode is shared or
> exclusive.  This is *separate* and *independent* of advisory locking (e.g.
> via flock() or fcntl() system calls).
> 
> Thawed opens preclude any frozen opens, and vice-versa.
> 
> Shared frozen opens preclude exclusive frozen opens.
> 
> Exclusive frozen opens preclude other frozen opens.

-- Mark --

* RCW 19.190 notice: This email address is located in Washington State.	*
* Unsolicited commercial email may be billed $500 per message.		*
Science does not emerge from voting, party politics, or public debate.

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 5 May 1999 16:49:27 -0700
Organization: Chez Inwap
Message-ID: <7gqle7$h6j$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gmko2$aq5$3@antiochus.ultra.net> <7gmp33$p17$1@shell3.ba.best.com> <7gn06k$7hn$1@eve.enteract.com>
Lines: 34
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925948181 198 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-raspail.gip.net!news-peer.gip.net!news.gsl.net!gip.net!feeder.qis.net!feed1.news.rcn.net!rcn!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gn06k$7hn$1@eve.enteract.com>,
David Scheidt  <dscheidt@enteract.com> wrote:
>In alt.folklore.computers Joe Smith <inwap@best.com> wrote:
>
>: The only thing I've seen vaguely like this function is 'ext2ed' on Linux.
>: 	-Joe
>
>That's because the linux "filesystem" doesn't guarantee filesystem consistancy

You missed my point.  I was talking about ext2ed, not ext2fs.

Both ext2ed and FILDDT allow an expert bit twiddler to go in a patch
bytes and block on the disk.  To recover from file system damage that
fsck or CHKDSK could not handle.

For instance, one time a block holding a SAT got an unrecoverable write
error.  TOPS-10 handled bad blocks by making an entry in the BAT, and
when CHKDSK rebuilt the SATs by looking up every single file on the disk,
it would also mark bad blocks as being permanently in use.  But it could
not write the updated SAT block.  I used FILDDT to read in the RIB for
SAT.SYS, locate the pointer to the cluster with the bad block, and poke
a different value into that location.

Until I came across ext2ed, I had not seen a similarly functional disk
editing tool for Unix.  (I'm sure they were out there, I just had not
come across any others.)

	-Joe


-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: michael.wojcik@merant.com (Michael Wojcik)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 5 May 1999 18:04:59 GMT
Organization: MERANT Inc.
Lines: 68
Message-ID: <7gq18b$13lf@news2.newsguy.com>
References: <371be3e4.0@newsfeed.one.net> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu> <u90b41qfa.fsf@graphics.cornell.edu> <7go432$ekt$1@vixen.cso.uiuc.edu>
NNTP-Posting-Host: p-335.newsdawg.com
X-Newsreader: xrn 9.00
Originator: mww@raederle.microfocus.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!pln-e!spln!extra.newsguy.com!newsp.newsguy.com!michael.wojcik


In article <7go432$ekt$1@vixen.cso.uiuc.edu>, tsmurphy@students.uiuc.edu (Terry Murphy) writes:
> In article <u90b41qfa.fsf@graphics.cornell.edu>,
> Stephen H. Westin <westin*nospam@graphics.cornell.edu> wrote:
> 
> >I'm afraid I'm naive about this. Are you saying that if I fopen a file
> >for write, other processes are blocked from doing the same? I'm pretty
> >sure that's not true in HP-UX or IRIX. The equivalent in Apollo AEGIS
> >did that by default; I suspect that TOPS-20 did something similar.

Besides the SVID mandatory locking already mentioned in other posts,
some Unix variants (eg. AIX) have open() flags that implement
mandatory-locking-on-open.  They actually implement SMB-style sharing
modes: open a file with O_RSHARE, and the open only succeeds if no
process has it open for writing, and subsequent opening for writing
fails until the O_RSHARE descriptor is closed.  Similarly, O_NSHARE
prohibits all other opens, not just opens for writing.

Yes, it would be nice if this were standard in all Unix implementations,
and had been since Day One (January 2, 1970).  However, claiming that
mandatory locking is simply unavailable in Unix is inaccurate.  It's
not difficult to write mandatory-locking code and wrap it so it builds
on all the popular Unix variants that support it (which I believe is
all of them, for sufficiently sensible definitions of "popular";
besides those already mentioned, SCO OSR 5 and UnixWare 2 support SVID
locking).

(Forget about fopen, by the way.  It's a general-purpose interface
that was never supposed to support everything.  If you want to fopen
and establish a sharing mode, use open and fdopen.  Use of the stdio
FILE functions isn't recommended with locking anyway, since their
buffering may produce confusing results.  Note that fopens from other
processes will respect sharing modes, even though they can't specify
them.)

> Another problem related to this is that it seems that it is customary
> under Unix to open the file only for the actual reading of the file,
> rather than for the duration that the file is open (from the users'
> standpoint).  For example, in text editors under Unix, it seems that
> the file is read and then immediately closed, and when it is saved it
> is open/written/closed again.  Very annoying, of course.

There's no "of course" about it.  I've done extensive work on systems
that do mandatory-locking when files (or the equivalent) are opened
for editing, and on systems that don't.  I prefer the latter.

If the editor is holding a mandatory exclusive lock on the file, then
I have to stop editing the file to use it (for compilation, for
example).  I also can't browse a file that someone else is editing.

If an editable file may be updated by more than one person, it should
be under revision control, and only the person who checked it out can
edit it.  The editor doesn't need to lock it.

A locking editor does make it a bit harder for people to step on
others' feet (or their own), true.  It's not overwhelmingly clear
that the advantages outweigh the disadvantages.


Michael Wojcik          	        michael.wojcik@merant.com
AAI Development, MERANT                 (block capitals are a company mandate)
Department of English, Miami University

However, we maintain that our mission is more than creating high-tech
amusement--rather, we must endeavor to provide high-tech, high-touch
entertainment with an emphasis on enkindling human warmth.
-- "The Ultimate in Entertainment", from the president of video game
   producer Namco

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 5 May 1999 18:38:41 -0700
Organization: Chez Inwap
Message-ID: <7gqrr1$b0o$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC>
Lines: 78
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925954726 216 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!howland.erols.net!outgoing.news.rcn.net.MISMATCH!feed1.news.rcn.net!rcn!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC>,
Mark Crispin  <MRC@CAC.Washington.EDU> wrote:
>On 4 May 1999, Alexander Viro wrote:
>> >By the way, TOPS-20 had both modes available to the programmer.
>> So does BSD (and derivatives). And Linux. And Solaris. Your point being? v7?
>
>Wrong.  BSD, Linux, and Solaris do *NOT* have mandatory locking.

That's what I thought until yesterday.  Solaris does claim mandatory locking.
As does any Unix conforming to version 3 of the SVID that AT&T wrote.

>Mandatory locking is an open mode that prevents any other programs from
>opening the file.  It also prevents opening if any other program has the
>file open.

According to the man pages for Solaris, that is what is implemented.
	-Joe

=========================

chmod(1)                  User Commands                  chmod(1)
SunOS 5.5.1          Last change: 1 Feb 1995                    1

NAME
     chmod - change the permissions mode of a file

        4000      Set user ID on execution.
        20#0      Set group ID on execution if # is 7, 5, 3, or 1.
                  Enable mandatory locking  if # is 6, 4, 2, or 0.
        1000      Turn on sticky bit. See chmod(2).
        0400      Allow read by owner.
        0200      Allow write by owner.
        0100      Allow execute (search in directory) by owner.
        0700      Allow  read,  write,  and   execute (search) by owner.
        0040      Allow read by group.
        0020      Allow write by group.
        0010      Allow execute (search in directory) by group.
        0070      Allow  read,  write,  and   execute (search) by group.
        0004      Allow read by others.
        0002      Allow write by others.
        0001      Allow execute (search in directory) by others.
        0007      Allow  read,  write,  and   execute (search) by others.

...
            permission
                    any compatible combination of  the  following letters:

                    r         read permission
                    w         write permission
                    x         execute permission
                    l         mandatory locking
                    s         user or group set-ID
                    t         sticky bit
...
                    Mandatory file and record locking (l)  refers
                    to  a  file's  ability to have its reading or
                    writing permissions locked while a program is
                    accessing that file.

=========================

lockf(3C)              C Library Functions              lockf(3C)
SunOS 5.5.1          Last change: 3 May 1994                    1

NAME
     lockf - record locking on files

DESCRIPTION
     lockf() allows sections of a file to be locked; advisory  or
     mandatory write locks depending on the mode bits of the file
     (see chmod(2)).  Locking calls  from  other  processes  that
     attempt  to  lock the locked file section will either return
     an error value or be put to sleep until the resource becomes
     unlocked.
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 5 May 1999 19:59:39 -0700
Organization: Chez Inwap
Message-ID: <7gr0ir$bvf$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <87pv4g9bv4.fsf@slip-32-101-160-211.ma.us.ibm.net> <Pine.NXT.4.20.9905051403480.26122-100000@Tomobiki-Cho.CAC.Washington.EDU> <Pine.NXT.4.20.9905051426050.26122-100000@Tomobiki-Cho.CAC.Washington.EDU>
Lines: 92
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 925959585 208 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!uunet!ams.uu.net!newsfeed.amsterdam.nl.net!sun4nl!tank.news.pipex.net!pipex!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <Pine.NXT.4.20.9905051426050.26122-100000@Tomobiki-Cho.CAC.Washington.EDU>,
Mark Crispin  <mrc@CAC.Washington.EDU> wrote:
> "Mandatory locking" in the context of this discussion is an open mode
> which specifies that the file is to be opened in a "frozen" (vs.
> UNIX-style "thawed") mode, and whether the "frozen" mode is shared or
> exclusive.  This is *separate* and *independent* of advisory locking (e.g.
> via flock() or fcntl() system calls).
> 
> Thawed opens preclude any frozen opens, and vice-versa.
> 
> Shared frozen opens preclude exclusive frozen opens.
> 
> Exclusive frozen opens preclude other frozen opens.
>
>I almost forgot.  There's another important mode; exclusive opens which
>only interact against other exclusive opens but not against shared opens.

You lost me there.  Thawed and frozen are part of the TOPS-20 vocabulary,
not TOPS-10.

What is the practical difference between thawed and frozen?
What is the difference between a mandatory thawed lock and an exclusive
frozen lock?  Does frozen imply a lock against reading?
How do those modes compare with what TOPS-10 had?  

TOPS-10:
  unlocked, one generation: Multiple readers, all reading same file.

  create: One job is creating a new file.  There are no readers because
	the file does not exist yet.

  supersede: On job is in the process of superseding an existing file.
	There can be any number of readers; they can read the old
	(existing) generation of the file, including readers that started
	after the single writer did its open.  No other job can open
	the file for writing.

  unlocked, multiple generations: Any job that opened the old file before
	it was superseded can continue to read the old data.  Any reader
	that opened the file after the supersede was finished sees the
	new data.  There can be any number of old superseded files still
	open for input - the RIB and file data hangs around until the
	use count goes to zero.

  single-writer update: One job has the write lock and can do random access
	I/O to the file.  Other jobs can read the file, but it is very
	easy for them to read inconsistent data.  Typically software
	would use some sort of semaphore in the first block of the file
	to indicate "update in progress; do not read".

  simultaneous update: Any job that wants to update the file has to open the
	file with FILOP. function .FOMAU and no other job can have the file
	open via ENTER or via FILOP. function .FOSAU.  Any number of jobs can
	be reading the file before and after the file is put into
	multiple-access update mode.

The Monitor imposed no restrictions or interlocks when a file was being
simultaneously updated.  It was up to the individual programs to use a common
method of working out advisory locks between themselves.  The ENQ/DEQ facility
allowed a single job to request an advisory lock on multiple blocks in
multiple files using a single ENQ request.  That was the proper way to avoid
deadlocks when dealing with multi-file databases.

Note that there was no such thing as a mandatory read lock under TOPS-10.
If the file permissions allowed reading, then the file could be read
regardless of any mandatory write lock.

This contrasts with the behavior documented in the man page for lockf for
SunOS-4.1.x:

     The lock is mandatory if the set-GID bit (S_ISGID) is set and the
     group execute bit (S_IXGRP) is clear.

     If a process holds a mandatory exclusive lock on  a  segment
     of  a  file,  both read and write operations block until the
     lock is removed (see WARNINGS).

   WARNINGS
     Mandatory record locks are dangerous.  If a runaway or  oth-
     erwise  out-of-control  process should hold a mandatory lock
     on a file critical to the system and fail  to  release  that
     lock,  the entire system could hang or crash.  For this rea-
     son, mandatory record locks may be removed in a future SunOS
     release.  Use advisory record locking whenever possible.

Mandatory write locks are good.  Mandatory read locks are bad.
	-Joe

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: Mark Crispin <mrc@CAC.Washington.EDU>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 5 May 1999 21:07:02 -0700
Organization: Networks & Distributed Computing
Lines: 21
Message-ID: <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU>
References: <371be3e4.0@newsfeed.one.net> <Pine.NXT.4.20.9905041226160.25285-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gnip9$ok2@weyl.math.psu.edu> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com>
NNTP-Posting-Host: tomobiki-cho.cac.washington.edu
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Trace: nntp1.u.washington.edu 925963625 34164 (None) 140.142.17.39
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: lampin
To: Joe Smith <inwap@best.com>
In-Reply-To: <7gqrr1$b0o$1@shell3.ba.best.com>
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.enteract.com!newsfeed.berkeley.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc

On 5 May 1999, Joe Smith wrote:
> >Mandatory locking is an open mode that prevents any other programs from
> >opening the file.  It also prevents opening if any other program has the
> >file open.
> 
> According to the man pages for Solaris, that is what is implemented.

No it's not.  Solaris has a file protection mode that changes advisory
locking to mandatory locking.

That is not the same as mandatory locking in PDP-10 operating systems.

Joe, you should have known better.  Think of it as a file protection bit
that changes how the ENQ. UUO works.

-- Mark --

* RCW 19.190 notice: This email address is located in Washington State.	*
* Unsolicited commercial email may be billed $500 per message.		*
Science does not emerge from voting, party politics, or public debate.

######

From: Mark Crispin <mrc@CAC.Washington.EDU>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 5 May 1999 21:26:06 -0700
Organization: Networks & Distributed Computing
Lines: 67
Message-ID: <Pine.NXT.4.20.9905052109000.26847-100000@Tomobiki-Cho.CAC.Washington.EDU>
References: <371be3e4.0@newsfeed.one.net> <87pv4g9bv4.fsf@slip-32-101-160-211.ma.us.ibm.net> <Pine.NXT.4.20.9905051403480.26122-100000@Tomobiki-Cho.CAC.Washington.EDU> <Pine.NXT.4.20.9905051426050.26122-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gr0ir$bvf$1@shell3.ba.best.com>
NNTP-Posting-Host: tomobiki-cho.cac.washington.edu
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Trace: nntp1.u.washington.edu 925964770 21520 (None) 140.142.17.37
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: mturn
To: Joe Smith <inwap@best.com>
In-Reply-To: <7gr0ir$bvf$1@shell3.ba.best.com>
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.berkeley.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc

On 5 May 1999, Joe Smith wrote:
> What is the practical difference between thawed and frozen?

Thawed and frozen are open modes.

In frozen mode, there can be only one writer of the file.  In thawed mode,
there can be many writers of the file.

The first open of the file sets the precedent for future opens (until
everybody closes the file): if the first open is thawed, then all
subsequent opens must be thawed, regardless if read or write.  The same
holds true fro frozen access.

There is also an open mode called "unrestricted read", which allows you to
open a file readonly regardless of its thawed or frozen status.

TOPS-10 has only frozen open mode.

> What is the difference between a mandatory thawed lock and an exclusive
> frozen lock?

"Mandatory thawed lock" is a misnomer; but thawed mode precludes frozen
mode.

"Exclusive frozen lock" is also a misnomer; a file open frozen can only be
opened by a single writer ala TOPS-10.

This is separate from advisory locking via the ENQ% JSYS (or the
equivalent TOPS-10 ENQ. UUO).

> Does frozen imply a lock against reading?

No, just against thawed.

The typical trick if you need to lock against reading (generally, to defer
parsing of appended data while the data is being appended) is for the
sensitive reading code to open a second frozen designator (I/O channel in
TOPS-10 terms) read/write.  This will fail while the appending is going
on.  The neat thing about this is that non-sensitive reading can still go
on.

> How do those modes compare with what TOPS-10 had?  

TOPS-10 only has frozen.  UNIX only has thawed.  TOPS-20 has both.

You really need both, although IMHO frozen is much more commonly useful
than thawed and UNIX sucks for only having thawed.

Actually, for mail files, TOPS-10's capabilities are good enough.  The
problem that exists with mail on UNIX is that UNIX only has thawed mode,
when you really want frozen mode to control new mail read vs. new mail
delivery.  You can use advisory locks on UNIX, but then you have the
problem of how do you lock the resource which governs the mailbox state --
as in being able to "expunge" (purge messages marked for deletion) while
other processes have the mailbox open?  You end up needing another file to
hold that lock status.

In TOPS-20 (and on TOPS-10), you use ENQ% (ENQ.) advisory locking to lock
the mailbox state, and normal frozen access mandatory locking to govern
new mail read vs. new mail delivery.

-- Mark --

* RCW 19.190 notice: This email address is located in Washington State.	*
* Unsolicited commercial email may be billed $500 per message.		*
Science does not emerge from voting, party politics, or public debate.

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 5 May 1999 22:14:29 -0400
Organization: -ENOENT
Lines: 20
Message-ID: <7gqtu5$s3j@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7gn06k$7hn$1@eve.enteract.com> <7gqle7$h6j$1@shell3.ba.best.com> <7gqmbp$r3s$1@eve.enteract.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newspump.monmouth.com!newspeer.monmouth.com!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7gqmbp$r3s$1@eve.enteract.com>,
David Scheidt  <dscheidt@enteract.com> wrote:
>
>No.  You clearly missed mine.  I have never needed an ext2ed like utility on 
>a FFS-like filesystem.  

Then do your friggin' homework and find the papers describing the design of
ext2. It is derived from FFS. Or do the same homework by getting *any*
*BSD and comparing UFS and EXT2 drivers. Ferchrissake, they share 3/4 of
code. Differences between EXT2 and UFS being:
	a) EXT2 doesn't use fragments
	b) EXT2 keeps names in directory entries without '\0'
	c) EXT2 has slightly different order of fields in on-disk inodes
and superblock.
Sheesh... If you want to start old sync vs. async flamewar - at least take
care to RTFDejaNews and learn WTF it is about.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: David Scheidt <dscheidt@enteract.com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 6 May 1999 00:05:13 GMT
Organization: EnterAct L.L.C. Turbo-Elite News Server
Lines: 22
Message-ID: <7gqmbp$r3s$1@eve.enteract.com>
References: <371be3e4.0@newsfeed.one.net> <7gmko2$aq5$3@antiochus.ultra.net> <7gmp33$p17$1@shell3.ba.best.com> <7gn06k$7hn$1@eve.enteract.com> <7gqle7$h6j$1@shell3.ba.best.com>
NNTP-Posting-Host: 207.229.143.42
User-Agent: tin/pre-1.4-19990216 ("Styrofoam") (UNIX) (FreeBSD/3.1-STABLE (i386))
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.enteract.com!news.enteract.com!not-for-mail

In alt.folklore.computers Joe Smith <inwap@best.com> wrote:
: In article <7gn06k$7hn$1@eve.enteract.com>,
: David Scheidt  <dscheidt@enteract.com> wrote:
:>In alt.folklore.computers Joe Smith <inwap@best.com> wrote:
:>
:>: The only thing I've seen vaguely like this function is 'ext2ed' on Linux.
:>: 	-Joe
:>
:>That's because the linux "filesystem" doesn't guarantee filesystem consistancy

: You missed my point.  I was talking about ext2ed, not ext2fs.

No.  You clearly missed mine.  I have never needed an ext2ed like utility on 
a FFS-like filesystem.  

David Scheidt

-- 
dscheidt@enteract.com
Folklore has a long memory, and not much concern for the facts.  --John West

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7gn06k$7hn$1@eve.enteract.com> <7gqle7$h6j$1@shell3.ba.best.com> <7gqmbp$r3s$1@eve.enteract.com>
Organization: Plethora . Net - More net, less spam!
X-Newsreader: trn 4.0-test70 (17 January 1999)
From: seebs@plethora.net (Peter Seebach)
Lines: 24
Message-ID: <Oy6Y2.3136$WA4.552050@ptah.visi.com>
Date: Thu, 06 May 1999 01:49:34 GMT
NNTP-Posting-Host: 205.166.146.8
X-Complaints-To: abuse@visi.com
X-Trace: ptah.visi.com 925955374 205.166.146.8 (Wed, 05 May 1999 20:49:34 CDT)
NNTP-Posting-Date: Wed, 05 May 1999 20:49:34 CDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!chippy.visi.com!news-out.visi.com!ptah.visi.com!not-for-mail

In article <7gqmbp$r3s$1@eve.enteract.com>,
David Scheidt  <dscheidt@enteract.com> wrote:
>No.  You clearly missed mine.  I have never needed an ext2ed like utility on 
>a FFS-like filesystem.  

I have.

Once.

A machine was being rebooted periodically without fsck, sometimes just
crashing and coming back up; it finally died because someone cleverly shoved
the exhaust fan up against about six inches of velvet curtains.

Anyway, had to go in with fsdb and zero out an inode that was making fsck dump
core.

Once.

-s
-- 
Copyright 1999, All rights reserved.  Peter Seebach / seebs@plethora.net
C/Unix wizard, Pro-commerce radical, Spam fighter.  Boycott Spamazon!
Will work for interesting hardware.  http://www.plethora.net/~seebs/
Visit my new ISP <URL:http://www.plethora.net/> --- More Net, Less Spam!

######

From: David Scheidt <dscheidt@enteract.com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 6 May 1999 02:52:37 GMT
Organization: EnterAct L.L.C. Turbo-Elite News Server
Lines: 42
Message-ID: <7gr05l$6o1$1@eve.enteract.com>
References: <371be3e4.0@newsfeed.one.net> <7gn06k$7hn$1@eve.enteract.com> <7gqle7$h6j$1@shell3.ba.best.com> <7gqmbp$r3s$1@eve.enteract.com> <7gqtu5$s3j@weyl.math.psu.edu>
NNTP-Posting-Host: 207.229.143.42
User-Agent: tin/pre-1.4-19990216 ("Styrofoam") (UNIX) (FreeBSD/3.1-STABLE (i386))
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.enteract.com!news.enteract.com!not-for-mail

In alt.folklore.computers Alexander Viro <viro@weyl.math.psu.edu> wrote:
: In article <7gqmbp$r3s$1@eve.enteract.com>,
: David Scheidt  <dscheidt@enteract.com> wrote:
:>
:>No.  You clearly missed mine.  I have never needed an ext2ed like utility on 
:>a FFS-like filesystem.  

: Then do your friggin' homework and find the papers describing the design of
: ext2. It is derived from FFS. Or do the same homework by getting *any*
: *BSD and comparing UFS and EXT2 drivers. Ferchrissake, they share 3/4 of
: code. Differences between EXT2 and UFS being:
: 	a) EXT2 doesn't use fragments
: 	b) EXT2 keeps names in directory entries without '\0'
: 	c) EXT2 has slightly different order of fields in on-disk inodes
: and superblock.

I didn't think EXT2 used cylinder groups, either.
`
: Sheesh... If you want to start old sync vs. async flamewar - at least take
: care to RTFDejaNews and learn WTF it is about.

The syncronous metadata writing are a very important part of FFS stability.[1]
I have spent much too much time fixing ext2fs filesystems that have died 
while doing huge amounts of file creation and deletion to be convinced 
otherwise.  I have never seen the behavior on FFS FSes mounted sync, and have 
with ext2fs, and ffs mounted async.

David

[1] Gregory R. Ganger and Yale N. Patt wrote a paper a couple of years ago
which explained that you needn't actually do the metadata syncronously, but
merely in dependent order.  Their paper is available at 
http://www.ece.cmu.edu/~ganger/papers/CSE-TR-254-95/.  An implementation of 
this by Kirk McKusick is available in FreeBSD 3.0 and later (and probably 
elsewhere).  I haven't had any spectacular system failures since I started 
to use it, haven't heard of many problems with it. 

-- 
David Scheidt <dscheidt@enteract.com>
Large fibreglass fruits are much the same the world over.
                                                    -- Vicki Parslow Stafford

######

From: viro@weyl.math.psu.edu (Alexander Viro)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 6 May 1999 09:08:26 -0400
Organization: -ENOENT
Lines: 68
Message-ID: <7gs48a$smp@weyl.math.psu.edu>
References: <371be3e4.0@newsfeed.one.net> <7gqmbp$r3s$1@eve.enteract.com> <7gqtu5$s3j@weyl.math.psu.edu> <7gr05l$6o1$1@eve.enteract.com>
NNTP-Posting-Host: weyl.math.psu.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!newsfeed.berkeley.edu!cliffs.rs.itd.umich.edu!cloudbreak.rs.itd.umich.edu!srvr1.engin.umich.edu!news.tc.cornell.edu!news3.cac.psu.edu!not-for-mail

In article <7gr05l$6o1$1@eve.enteract.com>,
David Scheidt  <dscheidt@enteract.com> wrote:
>In alt.folklore.computers Alexander Viro <viro@weyl.math.psu.edu> wrote:
>: In article <7gqmbp$r3s$1@eve.enteract.com>,
>: David Scheidt  <dscheidt@enteract.com> wrote:
>:>
>:>No.  You clearly missed mine.  I have never needed an ext2ed like utility on 
>:>a FFS-like filesystem.  
>
>: Then do your friggin' homework and find the papers describing the design of
>: ext2. It is derived from FFS. Or do the same homework by getting *any*
>: *BSD and comparing UFS and EXT2 drivers. Ferchrissake, they share 3/4 of
>: code. Differences between EXT2 and UFS being:
>: 	a) EXT2 doesn't use fragments
>: 	b) EXT2 keeps names in directory entries without '\0'
>: 	c) EXT2 has slightly different order of fields in on-disk inodes
>: and superblock.
>
>I didn't think EXT2 used cylinder groups, either.

RTFSource. Yes, it does, yes, those cylinder groups are the same thing as FFS
cylinder groups, yes, inode table and bitmaps are spread across the groups,
yes, it tries to keep locality (both *BSD and Linux drivers) using the same
strategy as 4.4BSD implementation of UFS does.

>: Sheesh... If you want to start old sync vs. async flamewar - at least take
>: care to RTFDejaNews and learn WTF it is about.
>
>The syncronous metadata writing are a very important part of FFS stability.[1]
>I have spent much too much time fixing ext2fs filesystems that have died 
>while doing huge amounts of file creation and deletion to be convinced 
>otherwise.  I have never seen the behavior on FFS FSes mounted sync, and have 
>with ext2fs, and ffs mounted async.

	So mount ext2 sync and stop worrying.

>[1] Gregory R. Ganger and Yale N. Patt wrote a paper a couple of years ago
>which explained that you needn't actually do the metadata syncronously, but
>merely in dependent order.

	S-U is *not* about ordering. Basically it's a different model of
IO. Normal UNIX approach being that buffer cache implements a large paged
VM with software pager and persistent 1-1 storage. You have the block
device as an array of blocks. You can request a block and get it dragged
in-core. Since you have no hardware-generated page misses for such pointers
you have to use bread()/brelse() pair to inform the pager (==buffer cache)
when the block may be released. You also have to mark them dirty by hands.
Other than that, it's a fairly standard software paging VM. You also have an
IO queue and can send the in-core block to it and be informed when it will
hit the disk. Ordering within the queue is out of your control. IOW, you have
data on disk, data in core shadowing them and have to keep those two
relatively in sync.
	S-U considers differences between the on-disk and (on-disk+in-core)
not as a monolitic diff but as a bunch of separate patches. The strategy
being: pick a block, unroll some patches if needed, send the result to queue.
It's the same stratefy that you would normally use to sync two versions of
source tree - there you have modules instead of blocks, but picture is the
same; you are picking a module, looking for patches that had affected it in
the development tree, if some of them depend on a patch that didn't make
its way into the stable tree yet - unroll and submit the result.
From that POV full-sync simply guarantees that you have at most 1 patch. 
Full-async means that you submit changes in random order and hope that
nobody will shoot you in the midway, leaving a mess in the stable tree.
S-U allows you to safely work on many patches at once.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

######

From: Paul Grayson <paul.grayson@virgin.net>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Thu, 06 May 1999 09:17:56 +0100
Organization: Sirius Cybernetics Corporation
Lines: 30
Message-ID: <37315034.7D0DC294@virgin.net>
References: <371be3e4.0@newsfeed.one.net> <7gn06k$7hn$1@eve.enteract.com> <7gqle7$h6j$1@shell3.ba.best.com> <7gqmbp$r3s$1@eve.enteract.com> <Oy6Y2.3136$WA4.552050@ptah.visi.com>
NNTP-Posting-Host: 194.168.62.237
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.51 [en] (X11; I; Linux 2.2.7 i586)
X-Accept-Language: en
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!news-lond.gip.net!news.gsl.net!gip.net!tank.news.pipex.net!pipex!newsfeed.amsterdam.nl.net!sun4nl!uunet!ams.uu.net!ffx.uu.net!in3.uu.net!news7-gui.server.ntli.net!news-feed.ntli.net!shippo.virgin.net!nobody

> Once.
> 
> A machine was being rebooted periodically without fsck, sometimes just
> crashing and coming back up; it finally died because someone cleverly shoved
> the exhaust fan up against about six inches of velvet curtains.
> 
> Anyway, had to go in with fsdb and zero out an inode that was making fsck dump
> core.
> 

I can beat that. Last year I had to fly out to a customer site to get a
custom mail system up and running again. The mail service was core
dumping during startup when checking its directories. Visited the site
and found that the hardware was the only non-IBM Microchannel machine I
had ever seen and the software was around 5 years old and had not been
upgraded due to inadequate diskspace. This version of the OS had almost
zero tools. The existing second hard disk was way past its use-by date,
and appeared to have been in constant use for around 8 years.

I had to repeatedly run fsdb to clean up broken filesystem, running
strings on the core dumps to find where the service was when it failed.

Four hours and many coffees later it was fixed. I needed to reboot to
force fsck finish the disk tidy-up. At this stage the hardware platform
decided to not want to restart. It was not possible to get a hardware
engineer - this being an off-shore UK island on a Saturday afternoon. It
was unlikely that there would be anyone in the UK still converant with
this platform anyway.

I immediately flew back home.

######

From: don@news.daedalus.co.nz (Don Stokes)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 6 May 1999 10:39:38 GMT
Organization: Daedalus Consulting
Lines: 10
Message-ID: <925987176.988360@estelle.paradise.net.nz>
References: <371be3e4.0@newsfeed.one.net> <7gqle7$h6j$1@shell3.ba.best.com> <7gqmbp$r3s$1@eve.enteract.com> <Oy6Y2.3136$WA4.552050@ptah.visi.com>
NNTP-Posting-Host: estelle.paradise.net.nz
X-Trace: titan.xtra.co.nz 925987178 1878385 203.96.152.5 (6 May 1999 10:39:38 GMT)
X-Complaints-To: abuse@xtra.co.nz
NNTP-Posting-Date: 6 May 1999 10:39:38 GMT
Cache-Post-Path: estelle.paradise.net.nz!unknown@p16-cable.paradise.net.nz
X-Cache: nntpcache 2.3.3b4 (see http://www.nntpcache.org/)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.wli.net!su-news-hub1.bbnplanet.com!lsanca1-snf1!news.gtei.net!news.netgate.net.nz!news.xtra.co.nz!don

In article <Oy6Y2.3136$WA4.552050@ptah.visi.com>,
Peter Seebach <seebs@plethora.net> wrote:
>Anyway, had to go in with fsdb and zero out an inode that was making fsck dump
>core.

Erm, isn't that what clri is for?  That's certainly what I've used when 
faced with a similar situation.

-- 
Don Stokes, Networking Consultant  http://www.daedalus.co.nz  +64 25 739 724

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 6 May 1999 11:09:33 -0700
Organization: Chez Inwap
Message-ID: <7gslst$r0f$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <Pine.NXT.4.20.9905051426050.26122-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gr0ir$bvf$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052109000.26847-100000@Tomobiki-Cho.CAC.Washington.EDU>
Lines: 18
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 926014177 209 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newspump.monmouth.com!newspeer.monmouth.com!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <Pine.NXT.4.20.9905052109000.26847-100000@Tomobiki-Cho.CAC.Washington.EDU>,
Mark Crispin  <mrc@CAC.Washington.EDU> wrote:
>TOPS-10 only has frozen.  UNIX only has thawed.  TOPS-20 has both.

Well, the doc's I've seen show that some versions of Unix have frozen.

When lockf() is called on a file that has the appropriate permissions (the
SGID bit is on but the group-executable bit is off), it becomes a mandatory
lock, not an advisory one.  In this case, anyone else trying to open() the
file will be put to sleep until the original process releases the lockf()
lock.  Only one writer, no readers.

This is a frozen open, it is not?
	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 6 May 1999 11:15:17 -0700
Organization: Chez Inwap
Message-ID: <7gsm7l$dd$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gqle7$h6j$1@shell3.ba.best.com> <7gqmbp$r3s$1@eve.enteract.com> <7gqtu5$s3j@weyl.math.psu.edu>
Lines: 18
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 926014523 218 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!netnews.com!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gqtu5$s3j@weyl.math.psu.edu>,
Alexander Viro <viro@weyl.math.psu.edu> wrote:
>Sheesh... If you want to start old sync vs. async flamewar - at least take
>care to RTFDejaNews and learn WTF it is about.

Hey!  Your comment is completely on-target if this were a mailing list
for OS developers.  

But since this is alt.folklore.computers, it is perfectly legal for people to
spout as gospel things based on limited personal experience.

It would take all the fun out of it if everyone had to do research just
to say "my OS is better than your OS".  :-)
	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 6 May 1999 11:27:42 -0700
Organization: Chez Inwap
Message-ID: <7gsmuu$635$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU>
Lines: 38
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 926015267 216 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!news-peer-europe.sprintlink.net!news-peer1.sprintlink.net!news.sprintlink.net!nntp.abs.net!newshub2.home.com!newshub1.home.com!news.home.com!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU>,
Mark Crispin  <mrc@CAC.Washington.EDU> wrote:
>On 5 May 1999, Joe Smith wrote:
>> >Mandatory locking is an open mode that prevents any other programs from
>> >opening the file.  It also prevents opening if any other program has the
>> >file open.
>> 
>> According to the man pages for Solaris, that is what is implemented.
>
>No it's not.  Solaris has a file protection mode that changes advisory
>locking to mandatory locking.
>
>That is not the same as mandatory locking in PDP-10 operating systems.

There must be some subtle difference that I am just not getting.

TOPS10: Doing an open for write automatically prevents any other
	process from doing an open for write.

Unix: Doing an open for write and lockf() on a file with correct
	permissions prevents any other process from doing an open for write.

>Joe, you should have known better.  Think of it as a file protection bit
>that changes how the ENQ. UUO works.

As I see it, "think of a file protection bit that changes how open() works".

The second process does not get a chance to do a lockf(), flock() or fcntl().
Instead, the open() call goes to sleep until the first process is done.
TOPS-10 returns an immediate error code if a mandatory lock is in force,
Unix sleeps in the middle of open() if the mandatory lock is in force.

What is your interpretation of how the Unix mandatory lock works?
	-Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: Mark Crispin <mrc@CAC.Washington.EDU>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Thu, 6 May 1999 14:34:51 -0700
Organization: Networks & Distributed Computing
Lines: 65
Message-ID: <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU>
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gsmuu$635$1@shell3.ba.best.com>
NNTP-Posting-Host: tomobiki-cho.cac.washington.edu
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Trace: nntp1.u.washington.edu 926026499 26990 (None) 140.142.17.39
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: pao
To: Joe Smith <inwap@best.com>
In-Reply-To: <7gsmuu$635$1@shell3.ba.best.com>
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.ecrc.net!newshub.northeast.verio.net!iad-peer.news.verio.net!netnews1.nw.verio.net!netnews.nwnet.net!news.verio.net!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc

On 6 May 1999, Joe Smith wrote:
> There must be some subtle difference that I am just not getting.

What you're not getting is there is a big difference between:
 1) all opens are frozen (TOPS-10).  An independent advisory
    locking mechanism exists.  (ENQ. UUO)
 2) the programmer can decide whether to do a frozen or thawed
    open (TOPS-20).  An independent advisory locking mechanism
    exists.  (ENQ% JSYS)
 3) all opens are thawed, except on some variants of the OS in
    which, by setting the file protection code properly, the
    advisory lock system call on an open file will cause the
    file open to become frozen (UNIX).  Advisory locking is
    unavailable except in thawed mode.

In particular, note the following characteristics of (3):
 a) the default is thawed, and on many UNIX variants (including
    all traditional UNIX) that is the only mode.
 b) only the owner of the file can decide thawed vs. frozen; an
    application which wishes a frozen open must be the owner (as
    opposed to merely having write permission) if the file
    protection was not set in advance.  The same is true for an
    application which wishes a thawed open with advisory locking.
 c) freezing is not atomic; it requires a separate system call.
 d) a frozen-out open blocks, instead of failing.
 e) use of frozen mode precludes the use of advisory locking.

These are all undesirable characteristics.  (a) and (d) would be
acceptable as defaults if there was an always-available (none of this
"lock yourself into my super-duper variant" crap) way to choose the
non-default behavior.  (b), (c), and (e) are design bugs, plain and
simple.

Then there's the other bug in UNIX locking, which is that there is no such
thing as lock promotion and demotion!  Don't believe me?  Real the man
page, really carefully.  Have a barf bag handy.

Then there's the designers of SVR4, who think that locks are a global
resource in a process; specifically if a process touches a lock in one
open file descriptor, obviously any other open file descriptors on that
file in that process should be changed.

The less said about NFS and locking, the better.

The designers of UNIX apparently never talked to any database guys.  This
is typical of OS developers who think that the kernel is the only project
for a Real Programmer.

As much as I liked to badmouth TOPS-10's design, the fact remains that the
TOPS-10 guys did pay attention to applications.  It's unfortunate that the
ENQ. UUO entered TOPS-10 so late in its life (6.01?) but the fact is that
ENQ. worked and worked well.  TOPS-20's ENQ% was almost identical.  It
would be horrendous to try to emulate ENQ on UNIX.  Even the simple cases
are horrendous.

NT gets locking right.  I predict that, if the UNIX community continues to
bury its head in the sand about locking, this will start becoming an
issue.  The Evil Empire is certainly doing its best to make it one.

-- Mark --

* RCW 19.190 notice: This email address is located in Washington State.	*
* Unsolicited commercial email may be billed $500 per message.		*
Science does not emerge from voting, party politics, or public debate.

######

From: tsmurphy@students.uiuc.edu (Terry Murphy)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 6 May 1999 21:16:27 GMT
Organization: University of Illinois at Urbana-Champaign
Lines: 40
Message-ID: <7gt0rb$ab0$1@vixen.cso.uiuc.edu>
References: <371be3e4.0@newsfeed.one.net> <7gnip9$ok2@weyl.math.psu.edu> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com>
NNTP-Posting-Host: ux5.cso.uiuc.edu
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.cwix.com!128.174.5.49!vixen.cso.uiuc.edu!tsmurphy

In article <7gqrr1$b0o$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:

>...
>            permission
>                    any compatible combination of  the  following letters:
>
>                    r         read permission
>                    w         write permission
>                    x         execute permission
>                    l         mandatory locking
>                    s         user or group set-ID
>                    t         sticky bit
>...
>                    Mandatory file and record locking (l)  refers
>                    to  a  file's  ability to have its reading or
>                    writing permissions locked while a program is
>                    accessing that file.

Can you please give us a demo of how to use this feature? When I
try it just locks me completely out of the file:

ux8:[pts/14]/[16:11:46]:~% uname -a
SunOS ux8.cso.uiuc.edu 5.5.1 Generic_103640-24 sun4u sparc SUNW,Ultra-Enterprise
ux8:[pts/14]/[16:11:48]:~% cat > test_mandatory_lock
this file test the mandatory lock
ux8:[pts/14]/[16:12:10]:~% ls -l test_mandatory_lock
-rw-r--r--   1 tsmurphy       34 May  6 16:12 test_mandatory_lock
ux8:[pts/14]/[16:12:16]:~% cat test_mandatory_lock
this file test the mandatory lock
ux8:[pts/14]/[16:12:23]:~% chmod +l test_mandatory_lock
ux8:[pts/14]/[16:12:31]:~% ls -l test_mandatory_lock
-rw-r-Sr--   1 tsmurphy       34 May  6 16:12 test_mandatory_lock
ux8:[pts/14]/[16:12:37]:~% cat test_mandatory_lock
cat: cannot open test_mandatory_lock
ux8:[pts/14]/[16:12:43]:~%  

????

-- Terry

######

From: dpeschel@u.washington.edu (Derek Peschel)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 7 May 1999 02:28:55 GMT
Organization: University of Washington, Seattle
Lines: 28
Message-ID: <7gtj57$pae$1@nntp1.u.washington.edu>
References: <371be3e4.0@newsfeed.one.net> <7gr0ir$bvf$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052109000.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gslst$r0f$1@shell3.ba.best.com>
NNTP-Posting-Host: saul10.u.washington.edu
X-Trace: nntp1.u.washington.edu 926044135 25934 (None) 140.142.17.35
X-Complaints-To: help@cac.washington.edu
NNTP-Posting-User: dpeschel
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newspeer.clara.net!news.clara.net!news-lond.gip.net!news-peer.gip.net!news.gsl.net!gip.net!newsfeed.cwix.com!192.220.250.21!netnews1.nw.verio.net!netnews.nwnet.net!news.verio.net!news.u.washington.edu!dpeschel

In article <7gslst$r0f$1@shell3.ba.best.com>, Joe Smith <inwap@best.com> wrote:
>In article <Pine.NXT.4.20.9905052109000.26847-100000@Tomobiki-Cho.CAC.Washington.EDU>,
>Mark Crispin  <mrc@CAC.Washington.EDU> wrote:

>>TOPS-10 only has frozen.  UNIX only has thawed.  TOPS-20 has both.

>Well, the doc's I've seen show that some versions of Unix have frozen.
>
>When lockf() is called on a file that has the appropriate permissions (the
>SGID bit is on but the group-executable bit is off), it becomes a mandatory
>lock, not an advisory one.  In this case, anyone else trying to open() the
>file will be put to sleep until the original process releases the lockf()
>lock.  Only one writer, no readers.

I just wanted to point out that your phrase "has the appropriate
permissions" is what started the current round of the argument.  Mark will
probably reply and say the same thing.  He seems to think that if you have
to set a bit in a _directory entry_ for a file, in order to get a certain
behavior (frozen opens) relating to the _file itself_, then the OS as a
whole can't be said to provide the frozen-open service in the same way that
TOPS-10/20 does.

I don't want to prolong the argument, I want to shorten it (by making an
explicit point that you seem to have missed).  If I totally screw things up
because of this post, I'll apologize... that's sort of what happened the
last time I tried to jump in like this (on another forum).

-- Derek

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7gqmbp$r3s$1@eve.enteract.com> <Oy6Y2.3136$WA4.552050@ptah.visi.com> <925987176.988360@estelle.paradise.net.nz>
Organization: Plethora . Net - More net, less spam!
X-Newsreader: trn 4.0-test70 (17 January 1999)
From: seebs@plethora.net (Peter Seebach)
Lines: 18
Message-ID: <chGY2.4013$WA4.587291@ptah.visi.com>
Date: Fri, 07 May 1999 18:28:24 GMT
NNTP-Posting-Host: 205.166.146.8
X-Complaints-To: abuse@visi.com
X-Trace: ptah.visi.com 926101704 205.166.146.8 (Fri, 07 May 1999 13:28:24 CDT)
NNTP-Posting-Date: Fri, 07 May 1999 13:28:24 CDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!bignews.mediaways.net!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!chippy.visi.com!news-out.visi.com!ptah.visi.com!not-for-mail

In article <925987176.988360@estelle.paradise.net.nz>,
Don Stokes <don@news.daedalus.co.nz> wrote:
>In article <Oy6Y2.3136$WA4.552050@ptah.visi.com>,
>Peter Seebach <seebs@plethora.net> wrote:
>>Anyway, had to go in with fsdb and zero out an inode that was making fsck dump
>>core.

>Erm, isn't that what clri is for?  That's certainly what I've used when 
>faced with a similar situation.

fsdb let me figure out which inode.  ;)

-s
-- 
Copyright 1999, All rights reserved.  Peter Seebach / seebs@plethora.net
C/Unix wizard, Pro-commerce radical, Spam fighter.  Boycott Spamazon!
Will work for interesting hardware.  http://www.plethora.net/~seebs/
Visit my new ISP <URL:http://www.plethora.net/> --- More Net, Less Spam!

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <7gqmbp$r3s$1@eve.enteract.com> <7gqtu5$s3j@weyl.math.psu.edu> <7gr05l$6o1$1@eve.enteract.com>
Organization: Plethora . Net - More net, less spam!
X-Newsreader: trn 4.0-test70 (17 January 1999)
From: seebs@plethora.net (Peter Seebach)
Lines: 16
Message-ID: <cjGY2.4014$WA4.587504@ptah.visi.com>
Date: Fri, 07 May 1999 18:30:32 GMT
NNTP-Posting-Host: 205.166.146.8
X-Complaints-To: abuse@visi.com
X-Trace: ptah.visi.com 926101832 205.166.146.8 (Fri, 07 May 1999 13:30:32 CDT)
NNTP-Posting-Date: Fri, 07 May 1999 13:30:32 CDT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!news.maxwell.syr.edu!chippy.visi.com!news-out.visi.com!ptah.visi.com!not-for-mail

In article <7gr05l$6o1$1@eve.enteract.com>,
David Scheidt  <dscheidt@enteract.com> wrote:
>http://www.ece.cmu.edu/~ganger/papers/CSE-TR-254-95/.  An implementation of 
>this by Kirk McKusick is available in FreeBSD 3.0 and later (and probably 
>elsewhere).  I haven't had any spectacular system failures since I started 
>to use it, haven't heard of many problems with it. 

It's also in BSD/OS 4.x.  I know there were some early problem reports, and
I think there's at least one known failure mode - but it's *very* nice.

-s
-- 
Copyright 1999, All rights reserved.  Peter Seebach / seebs@plethora.net
C/Unix wizard, Pro-commerce radical, Spam fighter.  Boycott Spamazon!
Will work for interesting hardware.  http://www.plethora.net/~seebs/
Visit my new ISP <URL:http://www.plethora.net/> --- More Net, Less Spam!

######

Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
From: lisard@zetnet.co.uk
X-Everything: Net-Tamer V 1.08X
Date: 07 May 1999 21:01:26 GMT
Message-ID: <373354a6$0$494@news.zetnet.co.uk>
NNTP-Posting-Host: man-138.dialup.zetnet.co.uk
X-Trace: news.zetnet.co.uk 926110886 494 194.247.40.176
Lines: 17
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!dispose.news.demon.net!demon!colt.net!newspeer.clara.net!news.clara.net!peer.news.zetnet.net!master.news.zetnet.net!not-for-mail



On 1999-05-05 viro@weyl.math.psu.edu(AlexanderViro) said:
   :Then do your friggin' homework and find the papers describing the
   :design of ext2. It is derived from FFS. Or do the same homework by
   :getting *any* *BSD and comparing UFS and EXT2 drivers. Ferchrissake,
   :they share 3/4 of code. Differences between EXT2 and UFS being:
   :a) EXT2 doesn't use fragments
   :b) EXT2 keeps names in directory entries without '\0'
   :c) EXT2 has slightly different order of fields in on-disk inodes
   :and superblock.
   :Sheesh... If you want to start old sync vs. async flamewar - at
   :least take care to RTFDejaNews and learn WTF it is about.

A Unix vs Unix jihad. I'm awestruck.
--
Communa (lisard@zetnet.co.uk) -- you know soft spoken changes nothing

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sat, 08 May 99 13:19:41 GMT
Organization: UltraNet Communications, Inc.
Lines: 72
Message-ID: <7h1jtf$jqt$2@ligarius.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gsmuu$635$1@shell3.ba.best.com> <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU>
NNTP-Posting-Host: d5.dial-13.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 8 May 1999 15:06:23 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!fu-berlin.de!feeder.qis.net!newsfeed.usit.net!newsswitch.lcs.mit.edu!news.ultranet.com!d5

In article 
<Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU>,
   Mark Crispin <mrc@CAC.Washington.EDU> wrote:
>On 6 May 1999, Joe Smith wrote:
>> There must be some subtle difference that I am just not getting.
>
>What you're not getting is there is a big difference between:
> 1) all opens are frozen (TOPS-10).  An independent advisory
>    locking mechanism exists.  (ENQ. UUO)
> 2) the programmer can decide whether to do a frozen or thawed
>    open (TOPS-20).  An independent advisory locking mechanism
>    exists.  (ENQ% JSYS)
> 3) all opens are thawed, except on some variants of the OS in
>    which, by setting the file protection code properly, the
>    advisory lock system call on an open file will cause the
>    file open to become frozen (UNIX).  Advisory locking is
>    unavailable except in thawed mode.
>
>In particular, note the following characteristics of (3):
> a) the default is thawed, and on many UNIX variants (including
>    all traditional UNIX) that is the only mode.
> b) only the owner of the file can decide thawed vs. frozen; an
>    application which wishes a frozen open must be the owner (as
>    opposed to merely having write permission) if the file
>    protection was not set in advance.  The same is true for an
>    application which wishes a thawed open with advisory locking.
> c) freezing is not atomic; it requires a separate system call.
> d) a frozen-out open blocks, instead of failing.
> e) use of frozen mode precludes the use of advisory locking.
>
>These are all undesirable characteristics.  (a) and (d) would be
>acceptable as defaults if there was an always-available (none of this
>"lock yourself into my super-duper variant" crap) way to choose the
>non-default behavior.  (b), (c), and (e) are design bugs, plain and
>simple.
>
>Then there's the other bug in UNIX locking, which is that there is no such
>thing as lock promotion and demotion!  Don't believe me?  Real the man
>page, really carefully.  Have a barf bag handy.
>
>Then there's the designers of SVR4, who think that locks are a global
>resource in a process; specifically if a process touches a lock in one
>open file descriptor, obviously any other open file descriptors on that
>file in that process should be changed.
>
>The less said about NFS and locking, the better.
>
>The designers of UNIX apparently never talked to any database guys.  This
>is typical of OS developers who think that the kernel is the only project
>for a Real Programmer.
>
>As much as I liked to badmouth TOPS-10's design, the fact remains that the
>TOPS-10 guys did pay attention to applications.  It's unfortunate that the
>ENQ. UUO entered TOPS-10 so late in its life (6.01?) but the fact is that
>ENQ. worked and worked well.  TOPS-20's ENQ% was almost identical.  It
>would be horrendous to try to emulate ENQ on UNIX.  Even the simple cases
>are horrendous.
>
>NT gets locking right.  I predict that, if the UNIX community continues to
>bury its head in the sand about locking, this will start becoming an
>issue.  The Evil Empire is certainly doing its best to make it one.

It sounds like the approach taken when implementing disk storage
in UNIX was on a record basis rather than a file basis.  From
a TOPS10 point of view, this an upsidedown approach in disk 
organization with respect to access.

Have I gotten it wrong?

/BAH

Subtract a hundred and four for e-mail.

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 8 May 1999 20:55:16 -0700
Organization: Chez Inwap
Message-ID: <7h30v4$o8l$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gn06k$7hn$1@eve.enteract.com> <7gqle7$h6j$1@shell3.ba.best.com> <7gqmbp$r3s$1@eve.enteract.com>
Lines: 47
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 926222122 198 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gqmbp$r3s$1@eve.enteract.com>,
David Scheidt  <dscheidt@enteract.com> wrote:
>In alt.folklore.computers Joe Smith <inwap@best.com> wrote:
>: In article <7gn06k$7hn$1@eve.enteract.com>,
>: David Scheidt  <dscheidt@enteract.com> wrote:
>:>In alt.folklore.computers Joe Smith <inwap@best.com> wrote:
>:>
>:>: The only thing I've seen vaguely like this function is 'ext2ed' on Linux.
>:>: 	-Joe
>:>
>:>That's because the linux "filesystem" doesn't guarantee filesystem consistancy
>
>: You missed my point.  I was talking about ext2ed, not ext2fs.
>
>No.  You clearly missed mine.  I have never needed an ext2ed like utility on 
>a FFS-like filesystem.  

Today's "User Friendly" comic strip has captured the essence of this
(and similar) threads.
	-Joe

http://www.userfriendly.org/cartoons/archives/99may/19990508.html

Scene: Two male computer geeks at a table.  There is an Amiga calendar
       in the background.  A female tech walks in on the second panel.
					-------------------------
A.J.: Rember the days of DOS with CONFIG files?  EDIT was such a
      hopless text editor.

Greg: Wuss.  I used EDLIN.
					-------------------------
A.J.: Oh bite me.  I used "COPY CON".

Greg: Yeah?  Well I wrote my files with ECHO.
					-------------------------
A.J.: Dolt!

Greg: Wimp!

Miranda: Well, I edited the inodes by hand.  With magnets.
					-------------------------
http://www.userfriendly.org/static  (C) by Illiad.

-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: Will Rose <cwr@cts.com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 9 May 1999 09:41:17 GMT
Organization: CTS Network Services
Lines: 15
Message-ID: <7h3l7t$gnh$1@nusku.cts.com>
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gsmuu$635$1@shell3.ba.best.com> <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU> <7h1jtf$jqt$2@ligarius.ultra.net>
NNTP-Posting-Host: crash-i2.cts.com
X-Trace: nusku.cts.com 926242877 17137 205.163.0.7 (9 May 1999 09:41:17 GMT)
X-Complaints-To: usenet@nusku.cts.com
NNTP-Posting-Date: 9 May 1999 09:41:17 GMT
User-Agent: tin/pre-1.4-19990413 ("Endemoniada") (UNIX) (crash/3.2 (i386))
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!news.idt.net!nntp2.cerf.net!mercury.cts.com!nusku.cts.com!not-for-mail

In alt.folklore.computers jmfbahciv@aol.com wrote:
[...]
:>The designers of UNIX apparently never talked to any database guys.  This
:>is typical of OS developers who think that the kernel is the only project
:>for a Real Programmer.

Apparently the version of Unix used for database work within Bell Labs
split off early from (what is now) the standard Unix.  The improvements
they made never got back to the standard version, as no-one wanted to use
the standard version for database work.


Will
cwr@crash.cts.com

######

From: "Alan H. Martin" <AMartin@MA.UltraNet.Com>
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sun, 09 May 1999 11:10:03 -0400
Organization: UltraNet Communications , an RCN Company http://www.ultranet.com/
Lines: 104
Message-ID: <3735A54B.E72F6D38@MA.UltraNet.Com>
References: <371be3e4.0@newsfeed.one.net> <7g8sbh$e88@weyl.math.psu.edu> <7gb4n6$795$1@shell3.ba.best.com> <372CA8A3.C1DE2BBD@MA.UltraNet.Com> <7gj7nd$c1v$1@shell3.ba.best.com>
NNTP-Posting-Host: 209-122-232-202.s456.tnt4.sbo.ma.dialup.rcn.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 9 May 1999 15:11:02 GMT
X-Mailer: Mozilla 4.51 [en] (Win95; U)
X-Accept-Language: en,en-US,en-GB,es
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!not-for-mail

Joe Smith wrote:

> Here's a failure mode that I reported (and got a patch for):
>         .ASSIGN DSK TTY
>         .R BASIC
>         ^C
>         ^C
> 
> BASIC opened logical device TTY for input and output.  It then began to
> immediatly use IN and OUT UUOs without doing a LOOKUP or ENTER, since the
> program always talks to TTY:, right?
> 
> The IN fails because the unprivileged program had done an OPEN and IN without
> an intervening LOOKUP.  It tried to output "I/O error reading from device TTY"
> to device TTY.  But because it is doing an OUT without an intervening ENTER,
> that UUO fails.  The output routine goes into an infinite loop trying to
> output error messages.  In the mean time, the Control-C intercept routine
> was doing its job, preventing Control-C from stopping execution.
> 
> The patch was to exit quietly if TTY was not assigned to a terminal device.

I remember seeing that SPR go by in college.  That was you eh?  That's only
half the story.  (Actually, I just discovered it's only a third of the story).

The original code OPENed TTY:

	MOVEI	T,TTYBUF	;SET UP TTY BUFFS
	MOVEM	T,.JBFF
	INIT	1
	SIXBIT	/TTY/
	XWD	TYO,TYI
	HALT	.-3
	INBUF	1
	OUTBUF	1

You reported BASIC would loop if TTY referred to DSK.  Instead of solving the
problem by requiring that TTY be a TTY, the maintainer "solved" the problem by
requiring that it not be a disk.  The fix would have looked something like:


;	206	16982	IF THE TTY IS ASSIGNED DSK THE TTYIN ROUTINE OPENS
;			TTY  NO CHECK IS MADE FOR VALID DEVICE AND WILL 
;			REMAIN IN A RUN STATE.
...
	SETZ	T,		;[206]SET UP TO CHECK FOR
	DEVCHR	T,		;[206]A TTY DEVICE
	TLNN	T,(DV.TTY)	;[206]MUST BE A TTY DEVICE
	JRST	[OUTSTR	[ASCIZ /?COMMAND DEVICE NOT A TTY - ABORT -
/]
		EXIT]		;[206]DEVICE NOT LEGAL SO ERROR AND EXIT

(Note the trailing space in the middle line of the revision history entry. 
This is a portent of things to come).


20 years later, I now see that someone at DEC apparently then complained that
they couldn't type:

.DEASSIGN DSK
.CONTINUE

to recover from the problem.  So they tried again:

;	212	NONE	EDIT 206 DOES NOT RETURN CORRECTLY TO THE MONITOR
;			DOESNT ALLOW FOR A CONTINUE
...
	SETZ	T,		;[206]SET UP TO CHECK FOR
	DEVCHR	T,		;[206]A TTY DEVICE
	TLNN	T,(DV.TTY)	;[206]MUST BE A TTY DEVICE
	JRST	[OUTSTR	[ASCIZ /?COMMAND DEVICE NOT A TTY - ABORT -
/]
		EXIT	1,	;[206]DEVICE NOT LEGAL SO ERROR AND EXIT
		JRST	BASIC]	;[212]RETURN TO COLD START FOR CONTINUE'S


Whether or not the second fix was published, someone in the field (perhaps
Tony Rizzolo at Stevens) saw the first fix and kicked it in the shins:

.ASSIGN NUL TTY
.R BASIC

Since the null device is all things to all people, DEVCHR says it is
simultaneously a card punch, card reader, line printer, display, paper tape
punch, paper tape reader, DECtape, magtape, TTY and disk.  The TLNN skips,
BASIC sends "READY, FOR HELP TYPE HELP." to the bit bucket, waits for input,
gets EOF and restarts.  It opens TTY: (NUL:), sends "READY, FOR HELP TYPE
HELP." to the bit bucket, waits for input, ...  You can't halt it (because of
the ^C intercept).

This is SPRed.  The third try's the charm:

;	235	27382	ASSIGNING "NUL: TTY:" CAUSES BASIC TO LOOP AT STARTUP.
...
	SETZ	T,		;[206]SET UP TO CHECK FOR
	DEVCHR	T,		;[206]A TTY DEVICE
	TLNE	T,(DV.TTY)	;[206][235]MUST BE A TTY DEVICE
	TLNE	T,(DV.DSK!DV.DTA);[235]AND NO OTHERS (IE. NUL)
	JRST	[OUTSTR	[ASCIZ /?COMMAND DEVICE NOT A TTY - ABORT -
/]
		EXIT	1,	;[206]DEVICE NOT LEGAL SO ERROR AND EXIT
		JRST	BASIC]	;[212]RETURN TO COLD START FOR CONTINUE'S
				/AHM
-- 
Alan Howard Martin				AMartin@MA.UltraNet.Com

######

From: cpierce1@ford.com (Clinton Pierce)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Mon, 10 May 1999 14:06:37 GMT
Organization: Hardly
Message-ID: <3737e7cd.1748774@news.ford.com>
References: <373354a6$0$494@news.zetnet.co.uk>
NNTP-Posting-Host: 19.88.81.187
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Newsreader: Forte Agent 1.5/32.452
Lines: 12
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!news-feed.inet.tele.dk!bofh.vszbr.cz!dispose.news.demon.net!demon!europa.netcrusader.net!141.211.144.13!cliffs.rs.itd.umich.edu!jobone!dailyplanet.srl.ford.com!eccws1.dearborn.ford.com!news

On 07 May 1999 21:01:26 GMT, lisard@zetnet.co.uk wrote:
>   :Sheesh... If you want to start old sync vs. async flamewar - at
>   :least take care to RTFDejaNews and learn WTF it is about.
>
>A Unix vs Unix jihad. I'm awestruck.

I'm impressed.  It's what keeps UNIX fresh after 30 years...

-- 
Clinton A. Pierce       "If you rush a Miracle Man, you get rotten
clintp@geeksalad.org        Miracles."  -- Miracle Max, The Princess Bride
http://www.geeksalad.org

######

Message-ID: <fc.1deb005a367800001deb005a926a14bf.4016350@city.magnet.at>
Date: Mon, 10 May 1999 16:06:37 +0200
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
From: cpierce1@ford.com (cpierce1@ford.com)
X-FC-Icon-ID: 23033
X-FC-Options: suppress-ndn
Newsgroups: alt.folklore.computers
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
Lines: 27
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!newsfeed01.univie.ac.at!magnet.at!city.magnet.at!news

On 07 May 1999 21:01:26 GMT, lisard@zetnet.co.uk wrote:
>   :Sheesh... If you want to start old sync vs. async flamewar - at
>   :least take care to RTFDejaNews and learn WTF it is about.
>
>A Unix vs Unix jihad. I'm awestruck.

I'm impressed.  It's what keeps UNIX fresh after 30 years...

-- 
Clinton A. Pierce       "If you rush a Miracle Man, you get rotten
clintp@geeksalad.org        Miracles."  -- Miracle Max, The Princess Bride
http://www.geeksalad.org

Message-ID: <3737e7cd.1748774@news.ford.com>
Path:
magnet.at!newsfeed03.univie.ac.at!newscore.univie.ac.at!howland.erols.net!newsxfer3.itd.umich.edu.MISMATCH!cliffs.rs.itd.umich.edu!jobone!dailyplanet.srl.ford.com!eccws1.dearborn.ford.com!news
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Date: Mon, 10 May 1999 14:06:37 GMT
From: cpierce1@ford.com (Clinton Pierce)
Organization: Hardly
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Lines: 12
Newsgroups: alt.folklore.computers,alt.sys.pdp10
References: <373354a6$0$494@news.zetnet.co.uk>

######

From: inwap@best.com (Joe Smith)
Newsgroups: alt.folklore.computers,alt.sys.pdp10
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 11 May 1999 02:24:02 -0700
Organization: Chez Inwap
Message-ID: <7h8svi$rnb$1@shell3.ba.best.com>
References: <371be3e4.0@newsfeed.one.net> <7gjhrh$lrk@weyl.math.psu.edu> <7gmjn5$lo2$1@shell3.ba.best.com> <7gmko2$aq5$3@antiochus.ultra.net>
Lines: 44
NNTP-Posting-Host: shell3.ba.best.com
X-Trace: nntp1.ba.best.com 926414647 224 inwap@206.184.139.134
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.berkeley.edu!news2.best.com!news3.best.com!nntp1.ba.best.com!not-for-mail

In article <7gmko2$aq5$3@antiochus.ultra.net>,  <jmfbahciv@aol.com> wrote:
>In article <7gmjn5$lo2$1@shell3.ba.best.com>,
>   inwap@best.com (Joe Smith) wrote:
>>In article <7gjhrh$lrk@weyl.math.psu.edu>,
>>Alexander Viro <viro@weyl.math.psu.edu> wrote:
>>>Possible failure mode: directory getting full before CLOSE. How do you 
>>>handle that?
>>
>>Directories don't get full.  There was no limit to the 
>>number of file names that could be stored in a directory.
>
>Well, there were practical limits.  One couldn't have a disk full
>of one RIB...hmm...I wonder what would have happened if we had
>tried to fill the disk with just a RIB and then modified a file?
>To make the file organization extensible there was the notion of
>a RIB...and then there was the spare RIB.

From another message:

> Not strictly true; if the RIB itself filled up, you lose. The monitor
> would not use extended RIBs for directories. I (and others) tried to
> shame TW into fixing that, but he would just snort... Only saw it
> happen once, and that was only because the directory blocks were full
> of zeros from deleted files...

That must have been before they added UFD compression.

The directory would get shorter if there were 64 consecutive deleted
file slots.  The first implementation had a bug on "DELETE *.*" on a
very big directory:  The first 64 files would be deleted, then the
UFD was compressed by removing the first block, shifting all the other
file names forward one block.  The program reading the UFD would continue
where it thought it had left off, and go after the third batch of 64 names
now residing in the second block of the UFD.  The end result was that only
half of the names got deleted on the first pass.  A directory N blocks
long needed log2(N) delete operations.

The fix was to defer UFD compression if the UFD was open for reading by
a program, and to delay an OPEN of the UFD if compression was in progress.
        -Joe
-- 
INWAP.COM is Joe Smith, Sally Smith and our cat Murdock.
  (The O'Hallorans and their cats moved to http://www.tyedye.org/ Nov-98.)
See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets"

######

From: michael.wojcik@merant.com (Michael Wojcik)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 11 May 1999 18:49:00 GMT
Organization: MERANT Inc.
Lines: 63
Message-ID: <7h9u2s$129@news1.newsguy.com>
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gsmuu$635$1@shell3.ba.best.com> <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU> <7h1jtf$jqt$2@ligarius.ultra.net>
NNTP-Posting-Host: p-627.newsdawg.com
X-Newsreader: xrn 9.00
Originator: mww@raederle.microfocus.com
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.eurocyber.net!howland.erols.net!newsxfer3.itd.umich.edu.MISMATCH!cliffs.rs.itd.umich.edu!logbridge.uoregon.edu!pln-w!spln!extra.newsguy.com!newsp.newsguy.com!michael.wojcik


[Barb's message is a bit old, but I haven't seen a response to it yet,
so...]

In article <7h1jtf$jqt$2@ligarius.ultra.net>, jmfbahciv@aol.com writes:

> It sounds like the approach taken when implementing disk storage
> in UNIX was on a record basis rather than a file basis.  From
> a TOPS10 point of view, this an upsidedown approach in disk 
> organization with respect to access.
> 
> Have I gotten it wrong?

It's worse than that, actually.  I don't share Mark Crispin's
vehemence on this issue, but I have to agree that file locking is
not an area where Unix (qua least-common-denominator-of-popular-
current-implementations-thereof) shines.

Unix implements disk storage on a file basis, loosely speaking.
Unfortunately, it implements file locking on a file-plus-offset-
for-length basis.  When you get a (traditional Unix) file lock, it's
for some length (zero or greater, in bytes; Unix has no built-in
concept of records) from either a specified position or the current
position in the file associated with that file descriptor, depending
on the API(s) your Unix implementation supports.

In most things, Unix deals with files of bytes.  For file types that
support the concept (eg. regular files, as opposed to pipes and
streaming device files and the like), there's a byte offset into the
file associated with each descriptor - what might be called a cursor
in some contexts.

File locks, on the other hand, apply to regions in files, one region
per lock.  The system guarantees (assuming it's working, which is
not always a safe assumption - viz NFS in many cases) that attempts
to lock overlapping regions will fail.  But the region concept doesn't
fit well with other system file primitives; it's clearly a half-assed
attempt at record-based locking, but it leaves the precise semantics
up to the application.  If applications need to share files that
agreement may be hard to come by.

There's a little relief in "zero-length" locks, which are defined to
"start at the current location and extend forward to the maximum
possible file size" - a lock on the whole file, in other words.  But
that may be a larger grain than your application wants, and there's
no guarantee that other applications were written to use zero-length
locks.

Oh, and non-zero regions can extend past the end of the file.  In
fact, they can begin past the end of the file.  That could potentially
trip up developers, I suppose.  Certainly it's not obvious if you
don't understand how locks are implemented.  (Yes, yes, in a perfect
world, all Unix programmers would know how the OS works.  But then
we'd have a better standard locking model, too.)


Michael Wojcik          	        michael.wojcik@merant.com
AAI Development, MERANT                 (block capitals are a company mandate)
Department of English, Miami University

He described a situation where a man is there to feed a dog and the dog
is there to keep the man from touching the equipment.
-- Anthony F. Giombetti

######

From: cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 11 May 1999 19:36:04 GMT
Organization: Honest Chris' Sysadmin Emporium
Message-ID: <7ha0r4$5qr$8@teabag.demon.co.uk>
References: <371be3e4.0@newsfeed.one.net>
    <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC>
    <7gqrr1$b0o$1@shell3.ba.best.com>
    <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU>
    <7gsmuu$635$1@shell3.ba.best.com>
    <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU>
    <7h1jtf$jqt$2@ligarius.ultra.net> <7h9u2s$129@news1.newsguy.com>
NNTP-Posting-Host: localhost
X-NNTP-Posting-Host: teabag.demon.co.uk:193.237.4.110
X-Trace: news.demon.co.uk 926452142 nnrp-04:16419 NO-IDENT teabag.demon.co.uk:193.237.4.110
X-Complaints-To: abuse@demon.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Newsreader: knews 1.0b.0
Lines: 14
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!cam-news-hub1.bbnplanet.com!news.gtei.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!teabag.demon.co.uk!not-for-mail

In article <7h9u2s$129@news1.newsguy.com>,
	michael.wojcik@merant.com (Michael Wojcik) writes:
> Unix implements disk storage on a file basis, loosely speaking.
> Unfortunately, it implements file locking on a file-plus-offset-
> for-length basis.  When you get a (traditional Unix) file lock, it's
> for some length (zero or greater, in bytes; Unix has no built-in
> concept of records) from either a specified position or the current
> position in the file associated with that file descriptor, depending
> on the API(s) your Unix implementation supports.

I suppose it's still an improvement on VM/CMS allegedly implementing
a lock strategy which locks out the entire minidisc for the duration...

Chris.

######

From: "Joel C. Ewing" <jcewing@acm.org>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 12 May 1999 07:24:07 -0500
Content-Transfer-Encoding: 7bit
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gsmuu$635$1@shell3.ba.best.com> <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU> <7h1jtf$jqt$2@ligarius.ultra.net> <7h9u2s$129@news1.newsguy.com> <7ha0r4$5qr$8@teabag.demon.co.uk>
X-Posted-Path-Was: not-for-mail
X-Accept-Language: en
Content-Type: text/plain; charset=us-ascii
X-ELN-Date: 12 May 1999 12:24:37 GMT
X-ELN-Insert-Date: Wed May 12 05:25:12 1999
Organization: EarthLink Network, Inc.
Lines: 49
Mime-Version: 1.0
NNTP-Posting-Host: sdn-ar-014mokcitp277.dialsprint.net
Message-ID: <373972E6.10863FCD@acm.org>
X-Mailer: Mozilla 4.51 [en] (Win98; U)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feeder.qis.net!newsfeed1.earthlink.net!nntp.earthlink.net!posted-from-earthlink!not-for-mail



Chris Hedley wrote:
> 
> In article <7h9u2s$129@news1.newsguy.com>,
>         michael.wojcik@merant.com (Michael Wojcik) writes:
> > Unix implements disk storage on a file basis, loosely speaking.
> > Unfortunately, it implements file locking on a file-plus-offset-
> > for-length basis.  When you get a (traditional Unix) file lock, it's
> > for some length (zero or greater, in bytes; Unix has no built-in
> > concept of records) from either a specified position or the current
> > position in the file associated with that file descriptor, depending
> > on the API(s) your Unix implementation supports.
> 
> I suppose it's still an improvement on VM/CMS allegedly implementing
> a lock strategy which locks out the entire minidisc for the duration...
> Chris.

...One of the reasons that VM is called a control program rather than an
Operating System.  The hardware VM is emulating with a minidisk doesn't
have the physical capability of locking on a finer level than the full
disk (reserve/release or online/offline), so neither does the emulated
disk.  You can, however, share a minidisk among multiple virtual
machines, and then it is the responsibility of the operating system
running on that virtual machine to perform the required logical locking
on finer levels.  CMS, not being a full-function operating system, does
have its limitations.

I see in another part of this thread discussion on mandatory versus
advisory locking for protection for file updates.  I am amazed that
anyone would even consider implementing significant production databases
that could be shared across applications or across multiple platforms in
an environment where the database integrity issues were not strongly
enforced and guaranteed by the underlying operating system, and not at
the mercy of application code or clients independently deciding to use a
consistent and safe protocol for updating indexed or relational
databases.  Considering the time and effort that was expended by IBM on
the MVS-OS/390 platform to address integrity and recoverability issues
in DB2, and to a lesser degree the update integrity even of VSAM files,
these complex issues are not the kind of thing that you want every
application to try to solve in their own, most-likely inadequate, way. 
Leaving these issues to the individual applications for
cross-application data is a high-risk decision.  This may be OK for
"toy" data, but anyone implementing mission-critical stuff in this way
ought to have their head examined.

-- 
Joel C. Ewing, Fort Smith, AR        jcewing@acm.org
Sr Technical Administrator, Mainframe Systems, DTC

######

From: Greg <nospam@erols.com>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 12 May 1999 08:44:44 -0400
Lines: 37
Sender: gregm@europa
Message-ID: <m3zp3a4fur.fsf@erols.com>
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gsmuu$635$1@shell3.ba.best.com> <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU> <7h1jtf$jqt$2@ligarius.ultra.net> <7h9u2s$129@news1.newsguy.com> <7ha0r4$5qr$8@teabag.demon.co.uk> <373972E6.10863FCD@acm.org>
X-Trace: 04obIxZknK/5bnssXL5qPTIrYvUU3RUXVknG2ifNBqQ=
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 12 May 1999 12:40:39 GMT
X-Newsreader:  Gnus v5.5/XEmacs 20.4 - "Emerald"
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!bignews.mediaways.net!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!not-for-mail

"Joel C. Ewing" <jcewing@acm.org> writes:

> 
> I see in another part of this thread discussion on mandatory versus
> advisory locking for protection for file updates.  I am amazed that
> anyone would even consider implementing significant production databases
> that could be shared across applications or across multiple platforms in
> an environment where the database integrity issues were not strongly
> enforced and guaranteed by the underlying operating system, and not at
> the mercy of application code or clients independently deciding to use a
> consistent and safe protocol for updating indexed or relational
> databases.  Considering the time and effort that was expended by IBM on
> the MVS-OS/390 platform to address integrity and recoverability issues
> in DB2, and to a lesser degree the update integrity even of VSAM files,
> these complex issues are not the kind of thing that you want every
> application to try to solve in their own, most-likely inadequate, way. 
> Leaving these issues to the individual applications for
> cross-application data is a high-risk decision.  This may be OK for
> "toy" data, but anyone implementing mission-critical stuff in this way
> ought to have their head examined.
> 

Absolutely right.  In my case it wasn't my head at issue, my clients
wanted it on the PC- there were a variety of other factors as well,
but any big, critical system really should not be on a PC.  

They're paying for it now- the usually lousy support staffing, buggy
OS, buggy database, buggy 3rd party libs, non-existent tech support,
typical stuff.  Nothing like what is required for a real system with
long term requirements.  

I wouldn't be suprised if this system will need to be rewritten in a
few years once Microsoft has abandoned compatibility with their older
software/databases.

Gregm

######

From: Mike Swaim <swaim@gemini.c-com.net>
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Newsgroups: alt.folklore.computers
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gsmuu$635$1@shell3.ba.best.com> <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU> <7h1jtf$jqt$2@ligarius.ultra.net> <7h9u2s$129@news1.newsguy.com> <7ha0r4$5qr$8@teabag.demon.co.uk> <373972E6.10863FCD@acm.org>
Organization: PointeCom
User-Agent: tin/pre-1.4-980818 ("Laura") (UNIX) (FreeBSD/2.2.8-RELEASE (i386))
Lines: 23
Message-ID: <pgg_2.3816$qQ4.78630@news2.giganews.com>
NNTP-Posting-Date: Wed, 12 May 1999 09:30:45 CDT
X-Trace: sv1-ELQp/5yNa1uUmiSnhJPX71O3aJlfek0HwNllMj6UB3XR0az7QWRo4oB7M0AYIQqJcl180hpgLxzmCGx!qYXDbLiNHxg=
X-Complaints-To: abuse@GigaNews.Com
X-Abuse-Info: Please be sure to forward a copy of ALL headers
X-Abuse-Info: Otherwise we will be unable to process your complaint properly
Date: Wed, 12 May 1999 14:30:45 GMT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsgate.cistron.nl!het.net!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!nntp.giganews.com!nntp2.giganews.com!news2.giganews.com.POSTED!gemini.c-com.net!swaim

Joel C. Ewing <jcewing@acm.org> wrote:

: I see in another part of this thread discussion on mandatory versus
: advisory locking for protection for file updates.  I am amazed that
: anyone would even consider implementing significant production databases
: that could be shared across applications or across multiple platforms in
: an environment where the database integrity issues were not strongly
: enforced and guaranteed by the underlying operating system, and not at
: the mercy of application code or clients independently deciding to use a
: consistent and safe protocol for updating indexed or relational
: databases.

  Who says that the clients have direct access to the files? In all of the
C/S database systems I've used, the clients have talked to a server
application that did all of the file I/O, usually ignoring advanced O/S
file locking features. (Heck, some UNIX SQL Servers write to their own
disk partitions.)


-- 
Mike Swaim, Avatar of Chaos: Disclaimer:I sometimes lie.
Home: swaim@c-com.net
Alum: swaim@alumni.rice.edu  Quote: "Boingie"^4 Y,W&D

######

From: Dave Daniels <a__fake__address@127.0.0.1>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 14 May 1999 19:01:17 BST
Organization: ArgoNet, but does not reflect its views
Lines: 46
Distribution: world
Message-ID: <na.f430d64902.a70220a__fake__address@127.0.0.1>
References: <371be3e4.0@newsfeed.one.net> <7ha0r4$5qr$8@teabag.demon.co.uk>
Reply-To: Dave Daniels <a__fake__address@127.0.0.1>
NNTP-Posting-Host: userl746.uk.uudial.com
X-Trace: lure.pipex.net 926753381 2591 193.149.76.51 (15 May 1999 07:29:41 GMT)
X-Complaints-To: abuse@uk.uu.net
NNTP-Posting-Date: 15 May 1999 07:29:41 GMT
X-Newsreader: NewsAgent 0.85 for RISC OS
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!tank.news.pipex.net!pipex!bore.news.pipex.net!pipex!argonet.co.uk!argbq79

In article <7ha0r4$5qr$8@teabag.demon.co.uk>,
cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) wrote:
> I suppose it's still an improvement on VM/CMS allegedly implementing
> a lock strategy which locks out the entire minidisc for the duration...

My knowledge of VM is out of date, but certainly in the older
releases (like the one we still run at work) there was no *file*
locking mechanism whatsoever provided by the operating system.
What control there was was limited to how a virtual machine could
gain access to a minidisk. In general, one virtual machine can had
read/write access to a minidisk and any number of others can have
read only access. It is also possible for everyone to have
read/write access but this is a very bad idea. If the virtual
machines are running CMS then you can guarentee that the disk will
be corrupted. The reason for this is that each virtual machine (or
to be more precise, each copy of CMS) holds its own independent
copy of the disk directory in memory and works from this. There
is/was no way of keeping them in sync automatically, and 'multi
write' access is just a disaster waiting to happen.

If you want CMS file locking then, yes, there are ways to do it.
Probably the easiest is to arrange for your application to link to
the minidisk only when it needs to read or write a file and to
release the disk when that operation is complete. You gaining
read/write access to the minidisk will restrict anybody else to
read only access until you detach the disk. I guess that what you
have seen is a form of this.
      
As I said, my knowledge is way out of date and I have no idea how
the later Shared Filing System deals with this. Perhaps somebody
else could fill in those details.

Dave


-- 
ANTISPAM: Please note that the email address above is false. My
correct address is:

    dave_daniels<at>argonet<dot>co<dot>uk

Please replace the <at> and <dot>s with @ and . respectively when
replying - Thanks!

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Tue, 18 May 99 10:14:21 GMT
Organization: UltraNet Communications, Inc.
Lines: 103
Message-ID: <7hrkss$e9g$1@ligarius.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gsmuu$635$1@shell3.ba.best.com> <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU> <7h1jtf$jqt$2@ligarius.ultra.net> <7h9u2s$129@news1.newsguy.com>
NNTP-Posting-Host: d11.dial-16.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 18 May 1999 12:02:36 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d11

In article <7h9u2s$129@news1.newsguy.com>,
   michael.wojcik@merant.com (Michael Wojcik) wrote:
>
>[Barb's message is a bit old, but I haven't seen a response to it yet,
>so...]

Thanks.  I was beginning to think I had asked a nonsense question. 
>
>In article <7h1jtf$jqt$2@ligarius.ultra.net>, jmfbahciv@aol.com writes:
>
>> It sounds like the approach taken when implementing disk storage
>> in UNIX was on a record basis rather than a file basis.  From
>> a TOPS10 point of view, this an upsidedown approach in disk 
>> organization with respect to access.
>> 
>> Have I gotten it wrong?
>
>It's worse than that, actually.  I don't share Mark Crispin's
>vehemence on this issue, but I have to agree that file locking is
>not an area where Unix (qua least-common-denominator-of-popular-
>current-implementations-thereof) shines.
>
>Unix implements disk storage on a file basis, loosely speaking.

Files are a way to organize data based on a top-down approach
no matter what the data is.  That way an operating system can
manage data storage integrity without interfering with its usage.
I'm not sure how much influence the disk geometry had in this
top-down design (but I'm sure it had something to do with it).



>Unfortunately, it implements file locking on a file-plus-offset-
>for-length basis.  When you get a (traditional Unix) file lock, it's
>for some length (zero or greater, in bytes; Unix has no built-in
>concept of records) from either a specified position or the current
>position in the file associated with that file descriptor, depending
>on the API(s) your Unix implementation supports.

I think this is awful (but that could be because I was brought
up on the -10 where multiple writers of a file were a no-no).
>
>In most things, Unix deals with files of bytes. 

I disagree with this approach because it involves the monitor
in the data.  I don't think it's the monitor's business to
meddle with the data; that's the app's business.  Is there
an exception?

> For file types that
>support the concept (eg. regular files, as opposed to pipes and
>streaming device files and the like), there's a byte offset into the
>file associated with each descriptor - what might be called a cursor
>in some contexts.
>
>File locks, on the other hand, apply to regions in files, one region
>per lock.  The system guarantees (assuming it's working, which is
>not always a safe assumption - viz NFS in many cases) that attempts
>to lock overlapping regions will fail.  But the region concept doesn't
>fit well with other system file primitives; it's clearly a half-assed
>attempt at record-based locking, but it leaves the precise semantics
>up to the application.  If applications need to share files that
>agreement may be hard to come by.

There's sharing files; then there's sharing data.  I often confuse
the two myself.  It's the monitor's responsibility to ensure
file integrity (e.g., a writer said write, so the monitor writes);
but it should _never_ have to know the type of data.  Maybe that's
why I don't like byte-addressable machines.  
>
>There's a little relief in "zero-length" locks, which are defined to
>"start at the current location and extend forward to the maximum
>possible file size" - a lock on the whole file, in other words.  But
>that may be a larger grain than your application wants, and there's
>no guarantee that other applications were written to use zero-length
>locks.
>
>Oh, and non-zero regions can extend past the end of the file.

But the monitor ensures that the extension past the EOF is
zero doesn't it?

>  In
>fact, they can begin past the end of the file.  That could potentially
>trip up developers, I suppose.  Certainly it's not obvious if you
>don't understand how locks are implemented.  (Yes, yes, in a perfect
>world, all Unix programmers would know how the OS works.  But then
>we'd have a better standard locking model, too.)

If the lock begins past the EOF, does that make a holey file?  Or
is the implied data logically contiguous, in other words the hole 
automatically becomes a part of the file.   No, that can't 
work...There'ld be no stopping a runaway updater.  Actually, 
does the data have to be physically contiguous, too?

It's clear that people confuse the physical location of data, with
the logical organization of data, with the physical distribution
of data (I do it all the time myself until I get a cold bucket
of water).  It sounds like UNIX has sort of munged some this together.

/BAH

Subtract a hundred and four for e-mail.

######

From: John Hughes <john@AtlanTech.COM>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 18 May 1999 18:28:29 +0200
Organization: Atlantic Technologies INC.
Lines: 30
Message-ID: <ufpv3y2vgy.fsf@microlite.CalvaCom.FR>
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gsmuu$635$1@shell3.ba.best.com> <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU> <7h1jtf$jqt$2@ligarius.ultra.net> <7h9u2s$129@news1.newsguy.com> <7hrkss$e9g$1@ligarius.ultra.net>
NNTP-Posting-Host: microlite.calvacom.fr
X-Trace: guinness.serv.psi.calva.net 927044909 14891 195.134.198.33 (18 May 1999 16:28:29 GMT)
X-Complaints-To: usenet@calvacom.fr
NNTP-Posting-Date: 18 May 1999 16:28:29 GMT
X-Newsreader: Gnus v5.5/XEmacs 20.3 - "Vatican City"
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.planete.net!psinet-france!not-for-mail

jmfbahciv@aol.com writes:

> In article <7h9u2s$129@news1.newsguy.com>,
>    michael.wojcik@merant.com (Michael Wojcik) wrote:
> >Oh, and non-zero regions can extend past the end of the file.
> >In fact, they can begin past the end of the file.

> If the lock begins past the EOF, does that make a holey file?  Or
> is the implied data logically contiguous, in other words the hole 
> automatically becomes a part of the file.   No, that can't 
> work...There'ld be no stopping a runaway updater.  Actually, 
> does the data have to be physically contiguous, too?

A lock starts at some location for some length.  The location is just
a number, the length is just a number.  When an application locks part 
of a file the OS just checks whether the specified region overlaps any 
existing lock region.  It doesn't care whether the region is in the file 
or not.

(C-ISAM and it's clone D-ISAM use this - their protocol for locking
record #n is to lock "SOME HUGE ADDRESS" + n for one byte.  The fact
that this address doesn't exist in the file doesn't matter).

It's pretty wierd.

-- 
John Hughes <john@Calva.COM>,
	Atlantic Technologies Inc.		Tel: +33-1-4313-3131
	66 rue du Moulin de la Pointe,		Fax: +33-1-4313-3139
	75013 PARIS.

######

Message-ID: <3742B53F.D61FA5E7@gazonk.del>
Date: Wed, 19 May 1999 08:57:35 -0400
From: "Foobar T. Clown" <fubar@gazonk.del>
Organization: Blurp
X-Mailer: Mozilla 4.5 [en] (WinNT; I)
X-Accept-Language: en
MIME-Version: 1.0
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gsmuu$635$1@shell3.ba.best.com> <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU> <7h1jtf$jqt$2@ligarius.ultra.net> <7h9u2s$129@news1.newsguy.com> <7hrkss$e9g$1@ligarius.ultra.net> <ufpv3y2vgy.fsf@microlite.CalvaCom.FR> <7hu6ef$j8n$2@antiochus.ultra.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
NNTP-Posting-Host: 198.133.110.49
X-Trace: 19 May 1999 09:00:20 -0500, 198.133.110.49
Lines: 15
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!nntp.abs.net!newsfeed.enteract.com!ix.netcom.com!zeus.nomos.com!198.133.110.49

jmfbahciv@aol.com:
> If [a] lock begins past the EOF, does that make a holey file?

John Hughes <john@AtlanTech.COM>:
> A lock starts at some location for some length.  The location is
> just a number, the length is just a number.  [...]

jmfbahciv@aol.com:
> I understand that.  Your comment didn't answer my question.  Perhaps
> I didn't word it clearly?

I doubt that many programmers who've been working for less than fifteen
years have ever seen the message, "File has holes," and I bet a lot of
them would think your were putting them on if you tried to explain what
it meant.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 19 May 99 09:26:01 GMT
Organization: UltraNet Communications, Inc.
Lines: 27
Message-ID: <7hu6ef$j8n$2@antiochus.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gsmuu$635$1@shell3.ba.best.com> <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU> <7h1jtf$jqt$2@ligarius.ultra.net> <7h9u2s$129@news1.newsguy.com> <7hrkss$e9g$1@ligarius.ultra.net> <ufpv3y2vgy.fsf@microlite.CalvaCom.FR>
NNTP-Posting-Host: d3.dial-17.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 19 May 1999 11:14:23 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!news.ultranet.com!d3

In article <ufpv3y2vgy.fsf@microlite.CalvaCom.FR>,
   John Hughes <john@AtlanTech.COM> wrote:
>jmfbahciv@aol.com writes:
>
>> In article <7h9u2s$129@news1.newsguy.com>,
>>    michael.wojcik@merant.com (Michael Wojcik) wrote:
>> >Oh, and non-zero regions can extend past the end of the file.
>> >In fact, they can begin past the end of the file.
>
>> If the lock begins past the EOF, does that make a holey file?  Or
>> is the implied data logically contiguous, in other words the hole 
>> automatically becomes a part of the file.   No, that can't 
>> work...There'ld be no stopping a runaway updater.  Actually, 
>> does the data have to be physically contiguous, too?
>
>A lock starts at some location for some length.  The location is just
>a number, the length is just a number.  When an application locks part 
>of a file the OS just checks whether the specified region overlaps any 
>existing lock region.  It doesn't care whether the region is in the file 
>or not.

I understand that.  Your comment didn't answer my question.  Perhaps
I didn't word it clearly?

/BAH

Subtract a hundred and four for e-mail.

######

From: jt5555@epix.net (Julian Thomas)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Wed, 19 May 1999 11:17:05 -0400
Organization: epix Internet Services
Lines: 20
Message-ID: <3742d615$1$wg$mr2ice@news.epix.net>
References: <371be3e4.0@newsfeed.one.net> 
      <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> 
      <7gqrr1$b0o$1@shell3.ba.best.com> 
      <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> 
      <7gsmuu$635$1@shell3.ba.best.com> 
      <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU> 
      <7h1jtf$jqt$2@ligarius.ultra.net> 
      <7h9u2s$129@news1.newsguy.com> 
      <7hrkss$e9g$1@ligarius.ultra.net> 
      <ufpv3y2vgy.fsf@microlite.CalvaCom.FR> 
      <7hu6ef$j8n$2@antiochus.ultra.net> 
      <3742B53F.D61FA5E7@gazonk.del>
NNTP-Posting-Host: itha-125ppp96.epix.net
X-Newsreader: MR/2 Internet Cruiser Edition for OS/2 v1.60 b60 
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!news-xfer.newsread.com!netaxs.com!newsread.com!news-xfer.epix.net!news1.epix.net!epix-news

In <3742B53F.D61FA5E7@gazonk.del>, on 05/19/99 
   at 08:57 AM, "Foobar T. Clown" <fubar@gazonk.del> said:

>I doubt that many programmers who've been working for less than fifteen
>years have ever seen the message, "File has holes," and I bet a lot of
>them would think your were putting them on if you tried to explain what
>it meant.

Along the same vein, does anyone else remember U-shaped files on tape?
 
-- 
 Julian Thomas: jt 5555 at epix dot net  http://home.epix.net/~jt
 remove numerics for email
 Boardmember of POSSI.org - Phoenix OS/2 Society, Inc  http://www.possi.org
 In the beautiful Finger Lakes Wine Country of New York State!
 -- --
 A flying saucer results when a nudist spills his coffee.

######

From: fishbowl@fotd.hosting.netcom.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 19 May 1999 11:52:12 -0500
Organization: Disorganization of Nonmembers
Lines: 14
Message-ID: <7huq7s$1mm$1@fotd.hosting.netcom.com>
References: <371be3e4.0@newsfeed.one.net> <ufpv3y2vgy.fsf@microlite.CalvaCom.FR> <7hu6ef$j8n$2@antiochus.ultra.net> <3742B53F.D61FA5E7@gazonk.del>
Reply-To: "J.L.M." <fishbowl+mail@hosting.netcom.com>
NNTP-Posting-Host: fit.hosting.netcom.com
X-NETCOM-Date: Wed May 19  9:53:15 AM PDT 1999
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsfeed.ecrc.net!newsfeed.enteract.com!ix.netcom.com!not-for-mail

In article <3742B53F.D61FA5E7@gazonk.del>,
Foobar T. Clown <fubar@gazonk.del> wrote:
>I doubt that many programmers who've been working for less than fifteen
>years have ever seen the message, "File has holes," and I bet a lot of
>them would think your were putting them on if you tried to explain what
>it meant.


On the other hand, if you say "sparse files" you might find more
folks who understand.  Modern unix filesystems do this (and linux/ext2).

-- 
James
http://ssdd.conservatory.com

######

From: cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 19 May 1999 13:17:02 GMT
Organization: Honest Chris' Sysadmin Emporium
Message-ID: <7hudke$4gt$1@teabag.demon.co.uk>
References: <371be3e4.0@newsfeed.one.net>
    <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC>
    <7gqrr1$b0o$1@shell3.ba.best.com>
    <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU>
    <7gsmuu$635$1@shell3.ba.best.com>
    <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU>
    <7h1jtf$jqt$2@ligarius.ultra.net> <7h9u2s$129@news1.newsguy.com>
    <7hrkss$e9g$1@ligarius.ultra.net>
    <ufpv3y2vgy.fsf@microlite.CalvaCom.FR>
    <7hu6ef$j8n$2@antiochus.ultra.net> <3742B53F.D61FA5E7@gazonk.del>
NNTP-Posting-Host: localhost
X-NNTP-Posting-Host: teabag.demon.co.uk:193.237.4.110
X-Trace: news.demon.co.uk 927122482 nnrp-12:11099 NO-IDENT teabag.demon.co.uk:193.237.4.110
X-Complaints-To: abuse@demon.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
NNTP-Posting-Date: 19 May 1999 13:17:02 GMT
X-Newsreader: knews 1.0b.0
Lines: 15
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!newsgate.cistron.nl!het.net!bullseye.news.demon.net!demon!news.demon.co.uk!demon!teabag.demon.co.uk!localhost!not-for-mail

In article <3742B53F.D61FA5E7@gazonk.del>,
	"Foobar T. Clown" <fubar@gazonk.del> writes:
> I doubt that many programmers who've been working for less than fifteen
> years have ever seen the message, "File has holes," and I bet a lot of
> them would think your were putting them on if you tried to explain what
> it meant.

I'd bet most Unix admins worth their salt are familiar with holey files,
and the problems therein; most notably restoring from a backup with, for
example, dbm files on it and finding it won't fit on the same disc it was
taken off (tar and cpio tend to fill in the holes with allocated, zeroed-
out blocks, which is an arse with a big INN history file and not too much
space to play around with)

Chris.

######

From: benh@lsl.co.uk (Ben Hutchings)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 19 May 1999 13:27:04 GMT
Organization: Laser-Scan Ltd.
Lines: 54
Message-ID: <7hue78$cq5@relay.lsl.co.uk>
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gsmuu$635$1@shell3.ba.best.com> <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU> <7h1jtf$jqt$2@ligarius.ultra.net> <7h9u2s$129@news1.newsguy.com> <7hrkss$e9g$1@ligarius.ultra.net>
NNTP-Posting-Host: 192.9.200.110
X-Newsreader: TIN [version 1.2 PL2]
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.icl.net!newspeer.clara.net!news.clara.net!news-lond.gip.net!news.gsl.net!gip.net!tank.news.pipex.net!pipex!warm.news.pipex.net!pipex!news.lsl.co.uk!benh

jmfbahciv@aol.com wrote:
: In article <7h9u2s$129@news1.newsguy.com>,
:    michael.wojcik@merant.com (Michael Wojcik) wrote:
<snip>
: >In most things, Unix deals with files of bytes. 

: I disagree with this approach because it involves the monitor
: in the data.  I don't think it's the monitor's business to
: meddle with the data; that's the app's business.  Is there
: an exception?

Bytes are the basic indivisible unit of storage on machines running
Unix.  So I think what Michael means is that Unix does *not* concern
itself with the structure of files.

<snip>
: >There's a little relief in "zero-length" locks, which are defined to
: >"start at the current location and extend forward to the maximum
: >possible file size" - a lock on the whole file, in other words.  But
: >that may be a larger grain than your application wants, and there's
: >no guarantee that other applications were written to use zero-length
: >locks.
: >
: >Oh, and non-zero regions can extend past the end of the file.

: But the monitor ensures that the extension past the EOF is
: zero doesn't it?

It's a little more complex than that.  It's possible to write data
to a file beyond the current end-of-file, leaving gaps.  This will
change the file size without storing any data on disk for the gap.
Reading from a gap in a file gives you zeroes.  I guess that every-
thing beyond end-of-file counts as a gap and will result in zeroes
too.

: >  In
: >fact, they can begin past the end of the file.  That could potentially
: >trip up developers, I suppose.  Certainly it's not obvious if you
: >don't understand how locks are implemented.  (Yes, yes, in a perfect
: >world, all Unix programmers would know how the OS works.  But then
: >we'd have a better standard locking model, too.)

: If the lock begins past the EOF, does that make a holey file?  Or
: is the implied data logically contiguous, in other words the hole 
: automatically becomes a part of the file.   No, that can't 
: work...There'ld be no stopping a runaway updater.  Actually, 
: does the data have to be physically contiguous, too?
<snip>

See the explanation above.  Files don't have to be logically or
physically contiguous.

--
Any opinions expressed are my own and not necessarily those of Laser-Scan.

######

From: jmfbahciv@aol.com
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Thu, 20 May 99 11:39:55 GMT
Organization: UltraNet Communications, Inc.
Lines: 33
Message-ID: <7i12lo$ed3$5@ligarius.ultra.net>
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gsmuu$635$1@shell3.ba.best.com> <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU> <7h1jtf$jqt$2@ligarius.ultra.net> <7h9u2s$129@news1.newsguy.com> <7hrkss$e9g$1@ligarius.ultra.net> <ufpv3y2vgy.fsf@microlite.CalvaCom.FR> <7hu6ef$j8n$2@antiochus.ultra.net> <ufiu9n3mwx.fsf@microlite.CalvaCom.FR>
NNTP-Posting-Host: d7.dial-14.mbo.ma.ultra.net
X-Complaints-To: abuse@ultra.net
X-Ultra-Time: 20 May 1999 13:28:24 GMT
X-Newsreader: News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!feed2.news.luth.se!luth.se!newspump.monmouth.com!newspeer.monmouth.com!news.ultranet.com!d7

In article <ufiu9n3mwx.fsf@microlite.CalvaCom.FR>,
   John Hughes <john@AtlanTech.COM> wrote:
>jmfbahciv@aol.com writes:
>
>> In article <ufpv3y2vgy.fsf@microlite.CalvaCom.FR>,
>>    John Hughes <john@AtlanTech.COM> wrote:
>> >jmfbahciv@aol.com writes:
>> >
>> >> In article <7h9u2s$129@news1.newsguy.com>,
>> >>    michael.wojcik@merant.com (Michael Wojcik) wrote:
>> >> >Oh, and non-zero regions can extend past the end of the file.
>> >> >In fact, they can begin past the end of the file.
>> >
>> >> If the lock begins past the EOF, does that make a holey file? 
>> >
>> >A lock starts at some location for some length.  The location is just
>> >a number, the length is just a number.
>>
>> I understand that.  Your comment didn't answer my question.  Perhaps
>> I didn't word it clearly?
>
>There is (if we ignore the "lock to eof" mode and the "autolock on
>read" mode) *NO* connection between the bytes in the file, the
>addresses of those bytes and the regions specified by the locks.
>
>Taking a lock on a region doesn't change the data in the file.
>
But, based on comments to my question, holiness appears
to change the data of the file (this is different from
changing data _in_ the file) without changing the data
of the disk.  [shudder--that's fraught with problems]

/BAH

######

From: John Hughes <john@AtlanTech.COM>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: 20 May 1999 15:12:30 +0200
Organization: Atlantic Technologies INC.
Lines: 30
Message-ID: <ufiu9n3mwx.fsf@microlite.CalvaCom.FR>
References: <371be3e4.0@newsfeed.one.net> <Pine.WNT.4.10.9905042324390.-3960447-100000@MRC> <7gqrr1$b0o$1@shell3.ba.best.com> <Pine.NXT.4.20.9905052105420.26847-100000@Tomobiki-Cho.CAC.Washington.EDU> <7gsmuu$635$1@shell3.ba.best.com> <Pine.NXT.4.20.9905061408560.27405-100000@Tomobiki-Cho.CAC.Washington.EDU> <7h1jtf$jqt$2@ligarius.ultra.net> <7h9u2s$129@news1.newsguy.com> <7hrkss$e9g$1@ligarius.ultra.net> <ufpv3y2vgy.fsf@microlite.CalvaCom.FR> <7hu6ef$j8n$2@antiochus.ultra.net>
NNTP-Posting-Host: microlite.calvacom.fr
X-Trace: guinness.serv.psi.calva.net 927205951 19777 195.134.198.33 (20 May 1999 13:12:31 GMT)
X-Complaints-To: usenet@calvacom.fr
NNTP-Posting-Date: 20 May 1999 13:12:31 GMT
X-Newsreader: Gnus v5.5/XEmacs 20.3 - "Vatican City"
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.belnet.be!isdnet!newsfeed.planete.net!psinet-france!not-for-mail

jmfbahciv@aol.com writes:

> In article <ufpv3y2vgy.fsf@microlite.CalvaCom.FR>,
>    John Hughes <john@AtlanTech.COM> wrote:
> >jmfbahciv@aol.com writes:
> >
> >> In article <7h9u2s$129@news1.newsguy.com>,
> >>    michael.wojcik@merant.com (Michael Wojcik) wrote:
> >> >Oh, and non-zero regions can extend past the end of the file.
> >> >In fact, they can begin past the end of the file.
> >
> >> If the lock begins past the EOF, does that make a holey file? 
> >
> >A lock starts at some location for some length.  The location is just
> >a number, the length is just a number.
>
> I understand that.  Your comment didn't answer my question.  Perhaps
> I didn't word it clearly?

There is (if we ignore the "lock to eof" mode and the "autolock on
read" mode) *NO* connection between the bytes in the file, the
addresses of those bytes and the regions specified by the locks.

Taking a lock on a region doesn't change the data in the file.

-- 
John Hughes <john@Calva.COM>,
	Atlantic Technologies Inc.		Tel: +33-1-4313-3131
	66 rue du Moulin de la Pointe,		Fax: +33-1-4313-3139
	75013 PARIS.

######

From: atrn@zeta.org.au (Andy Newman)
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Fri, 21 May 1999 01:01:53 +1000
Organization: is for the organized.
Lines: 10
Message-ID: <1581i7.g8f.ln@bebop.bsn>
References: <371be3e4.0@newsfeed.one.net>
    <ufpv3y2vgy.fsf@microlite.CalvaCom.FR>
    <7hu6ef$j8n$2@antiochus.ultra.net> <3742B53F.D61FA5E7@gazonk.del>
    <7huq7s$1mm$1@fotd.hosting.netcom.com>
NNTP-Posting-Host: d194.syd2.zeta.org.au
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Trace: phaedrus.zeta.org.au 927238282 8522 203.26.9.66 (20 May 1999 22:11:22 GMT)
X-Complaints-To: abuse@zeta.org.au
NNTP-Posting-Date: 20 May 1999 22:11:22 GMT
X-Newsreader: knews 1.0b.0
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!news-feed.inet.tele.dk!bofh.vszbr.cz!newsfeed.stanford.edu!newsfeed.berkeley.edu!intgwpad.nntp.telstra.net!news1.optus.net.au!optus!newsfeed.zip.com.au!zeta.org.au!news.zeta.org.au!bebop.bsn!nobody

In article <7huq7s$1mm$1@fotd.hosting.netcom.com>,
	fishbowl@fotd.hosting.netcom.com writes:
> On the other hand, if you say "sparse files" you might find more
> folks who understand.  Modern unix filesystems do this (and linux/ext2).

"Modern" Unix file systems have been doing sparse files for a long time
which sort of makes them not so modern. Interestingly NT's NTFS is catching
up. Sparse files are to be a feature of Windows 2000 (along with auto-freeing
of zeroed blocks) but you have to convert your file to a sparse file or some
other typical Win32 API ugliness. New Technology indeed.

######

From: Dave Daniels <a__fake__address@127.0.0.1>
Newsgroups: alt.folklore.computers
Subject: Re: How in Hell did the Great Unix to NT Migration begin??
Date: Sat, 22 May 1999 15:56:55 +0100
Organization: None
Lines: 21
Message-ID: <49063ea599a__fake__address@127.0.0.1>
References: <371be3e4.0@newsfeed.one.net>
    <ufpv3y2vgy.fsf@microlite.CalvaCom.FR>
    <7hu6ef$j8n$2@antiochus.ultra.net> <3742B53F.D61FA5E7@gazonk.del>
    <7huq7s$1mm$1@fotd.hosting.netcom.com> <1581i7.g8f.ln@bebop.bsn>
NNTP-Posting-Host: userbk86.uk.uudial.com
X-Trace: lure.pipex.net 927453633 28005 62.188.144.94 (23 May 1999 10:00:33 GMT)
X-Complaints-To: abuse@uk.uu.net
NNTP-Posting-Date: 23 May 1999 10:00:33 GMT
User-Agent: Pluto/1.11k  (RISC-OS/3.7)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newspeer.monmouth.com!tank.news.pipex.net!pipex!bore.news.pipex.net!pipex!argonet.co.uk!argbq79

In article <1581i7.g8f.ln@bebop.bsn>,
   Andy Newman <atrn@zeta.org.au> wrote:
> "Modern" Unix file systems have been doing sparse files for a long time
> which sort of makes them not so modern.

I have never tried any tests to find out, but ISTR reading
somewhere that CMS has sparse files. Can anybody confirm or
deny this?

Dave

-- 
ANTISPAM: Please note that the email address above is false. My
correct address is:

    dave_daniels<at>argonet<dot>co<dot>uk

Please replace the <at> and <dot>s with @ and . respectively when
replying - Thanks!