Newsgroups: alt.sys.pdp10,comp.human-factors
From: pwilson@world.std.com (Pete Wilson)
Subject: command-line completion
Message-ID: <GrDHML.Jso@world.std.com>
Date: Mon, 11 Feb 2002 14:30:21 GMT
Organization: The World Public Access UNIX, Brookline, MA
Lines: 10
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newspeer.radix.net!uunet!ash.uu.net!world!pwilson
Xref: chonsp.franklin.ch alt.sys.pdp10:10277

Dan Murphy, in one of his (so valuable) TOPS-20-archive/memory 
papers, speaks enthusiastically about command completion: type 
ESC after a partial command and the whole thing appears. From 
a programmer's pov, it is a tremendous idea: who wouldn't want 
to demo such a thing? "Now ... just watch THIS!!"

Anybody know if that idea was ever widely used in other OSs?
Ever proven useful in the real (extra-DEC) world?

-- Pete Wilson

######

From: jmfbahciv@aol.com
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: Mon, 11 Feb 02 14:30:57 GMT
Organization: UltraNet Communications, Inc.
Lines: 38
Message-ID: <a48sd6$eva$18@bob.news.rcn.net>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net>
X-Trace: UmFuZG9tSVbun8f6vNACLZeAcqp4PZKEv9d0u6KWVxmQHMOzfBhQGD1reFMZIt5g
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 11 Feb 2002 16:44:22 GMT
X-Newsreader:  News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-255-137
Xref: chonsp.franklin.ch alt.sys.pdp10:10282

In article <3C67EFC7.18E3E3C5@bartek.dontspamme.net>,
   Arthur Krewat <krewat@bartek.dontspamme.net> wrote:
>Pete Wilson wrote:
>> 
>> Dan Murphy, in one of his (so valuable) TOPS-20-archive/memory
>> papers, speaks enthusiastically about command completion: type
>> ESC after a partial command and the whole thing appears. From
>> a programmer's pov, it is a tremendous idea: who wouldn't want
>> to demo such a thing? "Now ... just watch THIS!!"
>> 
>> Anybody know if that idea was ever widely used in other OSs?
>> Ever proven useful in the real (extra-DEC) world?
>
>I don't want to start a flame-war here, but ...
>
>What use was command completion besides showing off to someone?

Our writers loved it when they were producing hardcopy examples.

>
>If I type "DI" I expect a directory - I don't care if I can type
>"DI<ESC>" and it types out DIRECTORY for me... what use is that?

Self-training.  Mapping -20 commands in my head to their
-10 equivalent.  If the command could be completed, then the 
characters typed were the bare mimumum.  


>
>Question mark on TOPS-20 yes, but command completion? A useless
>toy to me...

All I had to do was show a new user the ? and the $ command.  
They figured out how to use the system all on their own.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: Tue, 12 Feb 02 10:39:25 GMT
Organization: UltraNet Communications, Inc.
Lines: 37
Message-ID: <a4b37h$8e6$1@bob.news.rcn.net>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net> <1013451786.72920@dns2.serv.net>
X-Trace: UmFuZG9tSVb3LoIFrX1OxaXAZc+Rc5M9qaiuJwR0WsiToeweDBV+a7OzW2eKt7jl
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 12 Feb 2002 12:53:05 GMT
X-Newsreader:  News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-216-33
Xref: chonsp.franklin.ch alt.sys.pdp10:10294

In article <1013451786.72920@dns2.serv.net>,
   Patrick Scheible <kkt@itchy.serv.net> wrote:
>In alt.sys.pdp10 Arthur Krewat <krewat@bartek.dontspamme.net> wrote:
>
>> I don't want to start a flame-war here, but ...
>
>> What use was command completion besides showing off to someone?
>
>> If I type "DI" I expect a directory - I don't care if I can type
>> "DI<ESC>" and it types out DIRECTORY for me... what use is that?
>
>First, remembering the command you want but not how much of the
>command is a unique abbreviation.  I type di<ESC>, if it's a unique
>command I win because it completes it, if it's not a unique command I
>don't lose much because it just beeps once and waits for me to type
>another character or two (as opposed to giving an error and having to
>retype the whole command string).

There is also the little fact that customers added their own
commands.  So a poor little -20 developer's fingers from Marlboro
would be able to go on-site and not make a huge fool of himself
by not "knowing" the site's command set.  After all, we had to
maintain our mytique ;-).

More seriously, if a site had more than one system but with
different command sets, and only connections through the 
network, acclimating to a system was easier if the command 
sets were different.


/BAH

<snip>

/BAH

Subtract a hundred and four for e-mail.

######

From: Arthur Krewat <krewat@bartek.dontspamme.net>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Organization: Kilonet.net
Lines: 22
Message-ID: <3C67EFC7.18E3E3C5@bartek.dontspamme.net>
References: <GrDHML.Jso@world.std.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.78 [en] (X11; U; SunOS 5.8 i86pc)
X-Accept-Language: en
Date: Mon, 11 Feb 2002 16:27:54 GMT
NNTP-Posting-Host: 24.186.100.134
X-Trace: news02.optonline.net 1013444874 24.186.100.134 (Mon, 11 Feb 2002 11:27:54 EST)
NNTP-Posting-Date: Mon, 11 Feb 2002 11:27:54 EST
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!logbridge.uoregon.edu!gestalt.direcpc.com!cyclone2.usenetserver.com!usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10289

Pete Wilson wrote:
> 
> Dan Murphy, in one of his (so valuable) TOPS-20-archive/memory
> papers, speaks enthusiastically about command completion: type
> ESC after a partial command and the whole thing appears. From
> a programmer's pov, it is a tremendous idea: who wouldn't want
> to demo such a thing? "Now ... just watch THIS!!"
> 
> Anybody know if that idea was ever widely used in other OSs?
> Ever proven useful in the real (extra-DEC) world?

I don't want to start a flame-war here, but ...

What use was command completion besides showing off to someone?

If I type "DI" I expect a directory - I don't care if I can type
"DI<ESC>" and it types out DIRECTORY for me... what use is that?

Question mark on TOPS-20 yes, but command completion? A useless
toy to me...

aak

######

Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net> <a48sd6$eva$18@bob.news.rcn.net>
From: fin@visi.com (Craig A. Finseth)
Lines: 28
Message-ID: <90T98.30791$Wf1.6333289@ruti.visi.com>
Date: Mon, 11 Feb 2002 17:16:53 GMT
NNTP-Posting-Host: 209.98.98.8
X-Complaints-To: abuse@visi.com
X-Trace: ruti.visi.com 1013447813 209.98.98.8 (Mon, 11 Feb 2002 11:16:53 CST)
NNTP-Posting-Date: Mon, 11 Feb 2002 11:16:53 CST
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-out.visi.com!hermes.visi.com!ruti.visi.com!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10286

In article <a48sd6$eva$18@bob.news.rcn.net>,  <jmfbahciv@aol.com> wrote:
>In article <3C67EFC7.18E3E3C5@bartek.dontspamme.net>,
>   Arthur Krewat <krewat@bartek.dontspamme.net> wrote:
>>Pete Wilson wrote:
>>> 
>>> Dan Murphy, in one of his (so valuable) TOPS-20-archive/memory
>>> papers, speaks enthusiastically about command completion: type
>>> ESC after a partial command and the whole thing appears. From
>>> a programmer's pov, it is a tremendous idea: who wouldn't want
>>> to demo such a thing? "Now ... just watch THIS!!"
>>> 
>>> Anybody know if that idea was ever widely used in other OSs?
>>> Ever proven useful in the real (extra-DEC) world?

Completion is heavily used in Unix shells for filecompletion (and some
also do commmand completion).

Completion is heavily used in the Emacs text editor for completion of
all sorts.

Completion is heavily used in many development systems for completing
identifiers while entering code.

It is generally useful when you are selecting from a finite universe
of identifiers.

Craig

######

From: Patrick Scheible <kkt@itchy.serv.net>
Subject: Re: command-line completion
Newsgroups: alt.sys.pdp10,comp.human-factors
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net>
Reply-To: kkt@serv.net
User-Agent: tin/pre-1.4-19990413 ("Endemoniada") (UNIX) (FreeBSD/4.3-RELEASE (i386))
Organization: Ye 'Ol Disorganized NNTPCache groupie
Message-ID: <1013451786.72920@dns2.serv.net>
Cache-Post-Path: dns2.serv.net!unknown@itchy.serv.net
X-Cache: nntpcache 2.4.0b2 (see http://www.nntpcache.org/)
Lines: 29
Date: Mon, 11 Feb 2002 17:22:44 GMT
NNTP-Posting-Host: 199.201.191.2
X-Trace: newsfeed.slurp.net 1013448164 199.201.191.2 (Mon, 11 Feb 2002 11:22:44 CST)
NNTP-Posting-Date: Mon, 11 Feb 2002 11:22:44 CST
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.slurp.net!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10276

In alt.sys.pdp10 Arthur Krewat <krewat@bartek.dontspamme.net> wrote:

> I don't want to start a flame-war here, but ...

> What use was command completion besides showing off to someone?

> If I type "DI" I expect a directory - I don't care if I can type
> "DI<ESC>" and it types out DIRECTORY for me... what use is that?

First, remembering the command you want but not how much of the
command is a unique abbreviation.  I type di<ESC>, if it's a unique
command I win because it completes it, if it's not a unique command I
don't lose much because it just beeps once and waits for me to type
another character or two (as opposed to giving an error and having to
retype the whole command string).

Second, because of the noise words that tell you what parameters are
expected and what order.  If I want to copy a file but I'm not sure
whether the source or the destination is first, I type cop<esc> and it
completes and prompts (FROM FILE).  Then I type enough of the file
name to identify the file and <esc> and it completes the filename and
prompts (TO FILE).

Several Unix shells have command completion, I suspect that TOPS-20
was at least one of the sources of inspiration.  Of course the shells
can't do as good a job because of the design decision that the shells
don't know what kind of parameters programs expect.

-- Patrick Scheible

######

From: "Robert Stankowic" <pcdoktor@netway.at>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: Mon, 11 Feb 2002 18:25:47 +0100
Organization: UTA/netway (Customer)
Lines: 47
Sender: authuser@62.218.77.120
Message-ID: <a48ukl$thd$1@newsreader1.netway.at>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net>
NNTP-Posting-Host: 62.218.77.120
X-Trace: newsreader1.netway.at 1013448149 30253 62.218.77.120 (11 Feb 2002 17:22:29 GMT)
X-Complaints-To: abuse@netway.at
NNTP-Posting-Date: 11 Feb 2002 17:22:29 GMT
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2600.0000
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!fr.clara.net!heighliner.fr.clara.net!news-hub.siol.net!newsfeed.Austria.EU.net!newsfeed.kpnqwest.at!newsmaster-01.atnet.at!atnet.at!news.netway.at!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10288

Hello, this is my first posting here (after having watched this NG for quite
a while), so please forgive me (but point out) if I violate some of the
rules here :-)

"Arthur Krewat" <krewat@bartek.dontspamme.net> schrieb im Newsbeitrag
news:3C67EFC7.18E3E3C5@bartek.dontspamme.net...
> Pete Wilson wrote:
> >
> > Dan Murphy, in one of his (so valuable) TOPS-20-archive/memory
> > papers, speaks enthusiastically about command completion: type
> > ESC after a partial command and the whole thing appears. From
> > a programmer's pov, it is a tremendous idea: who wouldn't want
> > to demo such a thing? "Now ... just watch THIS!!"
> >
> > Anybody know if that idea was ever widely used in other OSs?
> > Ever proven useful in the real (extra-DEC) world?
>
> I don't want to start a flame-war here, but ...
>
> What use was command completion besides showing off to someone?
>
> If I type "DI" I expect a directory - I don't care if I can type
> "DI<ESC>" and it types out DIRECTORY for me... what use is that?
>
> Question mark on TOPS-20 yes, but command completion? A useless
> toy to me...

Completion of typing in my experience can be a really great help - maybe not
especially/not only with command lines. I take MSVC as an example: you type
the name of an aggregate type, for example "struct_something." with the dot
(the dot tells the editor that you want  a member of the type), and a menue
with a list of the members pops up. You type the first letter of the
member's name and the first name beginning with that letter is highlighted.
If it is the member you want, you just enter a space and the whole word is
transferred into your text. If it is not the member, enter the next
character and so on. No typos, no need to remember all the member names..
But I agree, that with command lines maybe just entering an unambiguous
abbreviation may be as good or sometimes even better.
Hmm, except, if you enter a wrong letter, it might be an existing
abbreviation, but not that one you want, and if the command interpreter
completes it for you, there is a good chance that you notice your mistyping
_before_ it causes some very undesired effect.

Kind regards
Robert

######

From: FlyLikeAnEagle@United.Com
Message-ID: <rOdGr40LMPU3-pn2-T0CjWEPnDQsW@tamqfl1-ar1-162-205.biz.dsl.gtei.net>
Newsgroups: comp.human-factors,alt.sys.pdp10
Subject: Re: command-line completion
References: <GrDHML.Jso@world.std.com>
User-Agent: ProNews/2 V1.51.ib110
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 8bit
Lines: 40
X-Trace: +r2abhvZrQlLlpOE/wjaTzP2vXWLchJA0WWLiMJjpmHjylh9nnFCpOOEMGrHCT51tCM7l5TZ+ooc!rDj7AGQmwdh8OhCvC3Q7W8DieNCoIMoifmvWbP8Wl+nJxceb8uZ3HUL6QajtfELASbYuq6CN3YaJ!MBapnGDLlF4Twg+5y4bvVz7Y1CTZjKG02J/OElQ=
X-Complaints-To: abuse@gte.net
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
NNTP-Posting-Date: Mon, 11 Feb 2002 17:29:14 GMT
Distribution: world
Date: Mon, 11 Feb 2002 17:29:14 GMT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!out.nntp.be!propagator-SanJose!in.nntp.be!nntp-relay.ihug.net!ihug.co.nz!logbridge.uoregon.edu!paloalto-snh1.gtei.net!news.gtei.net!paloalto-snr1.gtei.net.POSTED!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10291

  Pete,

  The concept of completion is a major part of the 4DOS, 4OS2, and 4NT
products.  I don't recall the vendor, but a Yahoo search should get it
for you.  They focused on directory name and file name completion.
This should be especially nice for long file and directory names.  I
love it on my system.

  Here is a quick run down.  I'll presume Windows since that is all most
people know.  Type "CD Win<tab>" and you get "CD Windows" on the
command line.  Okay, now if there are multiple matches just keep 
pressing <tab> or <shift+tab> to scroll through them.  You need not enter
a file name.  "CD <tab>" will find the first alphabetic file or directory
name.  It also knows what files need quotes about them.

  I've seen and used command completion programs as far back as the
Apple ][+ DOS.  They aren't as useful unless they include file and
directory completion.  BTW, I don't this 4DOS does command completion,
just file and directory completion.  It does support making your own 
commands, color coding and lots of other neat stuff.  It complements the
command environment very well.

  David

On Mon, 11 Feb 2002 14:30:21 UTC, pwilson@world.std.com (Pete Wilson) wrote:

> Dan Murphy, in one of his (so valuable) TOPS-20-archive/memory 
> papers, speaks enthusiastically about command completion: type 
> ESC after a partial command and the whole thing appears. From 
> a programmer's pov, it is a tremendous idea: who wouldn't want 
> to demo such a thing? "Now ... just watch THIS!!"
> 
> Anybody know if that idea was ever widely used in other OSs?
> Ever proven useful in the real (extra-DEC) world?
> 
> -- Pete Wilson


-- 

######

From: stamant@haeckel.csc.ncsu.edu (Robert St. Amant)
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 11 Feb 2002 13:01:06 -0500
Organization: NC State University
Lines: 38
Message-ID: <lpny9hzsxst.fsf@haeckel.csc.ncsu.edu>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net> <1013451786.72920@dns2.serv.net>
NNTP-Posting-Host: haeckel.csc.ncsu.edu
X-Trace: uni00nw.unity.ncsu.edu 1013450469 7248 152.14.51.145 (11 Feb 2002 18:01:09 GMT)
X-Complaints-To: news@unity.ncsu.edu
NNTP-Posting-Date: 11 Feb 2002 18:01:09 GMT
X-Newsreader: Gnus v5.7/Emacs 20.7
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!atlnga1-snf1.ops.genuity.net!news.gtei.net!news.futuresouth.com!news-out.spamkiller.net!propagator-la!news-in-la.newsfeeds.com!news-in.superfeed.net!uni00nw.unity.ncsu.edu!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10293

Patrick Scheible <kkt@itchy.serv.net> writes:

> In alt.sys.pdp10 Arthur Krewat <krewat@bartek.dontspamme.net> wrote:
> 
> > I don't want to start a flame-war here, but ...
> 
> > What use was command completion besides showing off to someone?
> 
> > If I type "DI" I expect a directory - I don't care if I can type
> > "DI<ESC>" and it types out DIRECTORY for me... what use is that?
> 
> First, remembering the command you want but not how much of the
> command is a unique abbreviation.  I type di<ESC>, if it's a unique
> command I win because it completes it, if it's not a unique command I
> don't lose much because it just beeps once and waits for me to type
> another character or two (as opposed to giving an error and having to
> retype the whole command string).

Or the system responds by displaying the possible completions of my
prefix, helping me to finish it correctly.

> Second, because of the noise words that tell you what parameters are
> expected and what order.  If I want to copy a file but I'm not sure
> whether the source or the destination is first, I type cop<esc> and it
> completes and prompts (FROM FILE).  Then I type enough of the file
> name to identify the file and <esc> and it completes the filename and
> prompts (TO FILE).
> 
> Several Unix shells have command completion, I suspect that TOPS-20
> was at least one of the sources of inspiration.  Of course the shells
> can't do as good a job because of the design decision that the shells
> don't know what kind of parameters programs expect.
> 
> -- Patrick Scheible

-- 
Rob St. Amant
http://www4.ncsu.edu/~stamant

######

From: Tim Shoppa <shoppa@trailing-edge.com>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 11 Feb 2002 10:01:38 -0800
Organization: Trailing-Edge Technologies
Lines: 35
Message-ID: <a490u201v07@drn.newsguy.com>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net> <a48sd6$eva$18@bob.news.rcn.net> <90T98.30791$Wf1.6333289@ruti.visi.com>
NNTP-Posting-Host: p-460.newsdawg.com
X-Newsreader: Direct Read News 2.91
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!drn
Xref: chonsp.franklin.ch alt.sys.pdp10:10278

In article <90T98.30791$Wf1.6333289@ruti.visi.com>, fin@visi.com says...
>
>In article <a48sd6$eva$18@bob.news.rcn.net>,  <jmfbahciv@aol.com> wrote:
>>In article <3C67EFC7.18E3E3C5@bartek.dontspamme.net>,
>>   Arthur Krewat <krewat@bartek.dontspamme.net> wrote:
>>>Pete Wilson wrote:
>>>> 
>>>> Dan Murphy, in one of his (so valuable) TOPS-20-archive/memory
>>>> papers, speaks enthusiastically about command completion: type
>>>> ESC after a partial command and the whole thing appears. From
>>>> a programmer's pov, it is a tremendous idea: who wouldn't want
>>>> to demo such a thing? "Now ... just watch THIS!!"
>>>> 
>>>> Anybody know if that idea was ever widely used in other OSs?
>>>> Ever proven useful in the real (extra-DEC) world?
>
>Completion is heavily used in Unix shells for filecompletion (and some
>also do commmand completion).

But the way it is done is much less elegant than in TOPS-20.  In TOPS-20,
the command interpreter and the command cooperate as to what the legal elements
of the command were, and thus completion of commands, options, and
filenames are all implemented.  While filename completion under a
Unix shell certainly *is* useful, the shell has no context for completing
command options.

TOPS-20 also has context sensitive help, so if you wanted to know what
your command options were it was easy.

Frank daCruz of Columbia has pointed me towards a context-sensitive
completion/help facility for use *within* a running program (after it
has already begun execution), but AFAIK it doesn't work for shells in general.

Tim.

######

From: Tim Shoppa <shoppa@trailing-edge.com>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 11 Feb 2002 10:05:45 -0800
Organization: Trailing-Edge Technologies
Lines: 38
Message-ID: <a4915p01vjo@drn.newsguy.com>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net> <1013451786.72920@dns2.serv.net>
NNTP-Posting-Host: p-630.newsdawg.com
X-Newsreader: Direct Read News 2.91
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!drn
Xref: chonsp.franklin.ch alt.sys.pdp10:10280

In article <1013451786.72920@dns2.serv.net>, Patrick says...
>Several Unix shells have command completion, I suspect that TOPS-20
>was at least one of the sources of inspiration.

It certainly was.  From "The T in TCSH":

  In 1964, DEC produced the PDP-6. The PDP-10 was a later
  re-implementation. It was re-christened the DECsystem-10 in
  1970 or so when DEC brought out the second model, the KI10. 

  TENEX was created at Bolt, Beranek & Newman (a Cambridge,
  Mass. think tank) in 1972 as an experiment in demand-paged
  virtual memory operating systems. They built a new pager for
  the DEC PDP-10 and created the OS to go with it. It was
  extremely successful in academia. 

  In 1975, DEC brought out a new model of the PDP-10, the
  KL10; they intended to have only a version of TENEX, which
  they had licensed from BBN, for the new box. They called their
  version TOPS-20 (their capitalization is trademarked). A lot of
  TOPS-10 users (`The OPerating System for PDP-10') objected;
  thus DEC found themselves supporting two incompatible
  systems on the same hardware--but then there were 6 on the
  PDP-11! 

  TENEX, and TOPS-20 to version 3, had command completion
  via a user-code-level subroutine library called ULTCMD. With
  version 3, DEC moved all that capability and more into the
  monitor (`kernel' for you Unix types), accessed by the
  COMND% JSYS (`Jump to SYStem' instruction, the supervisor
  call mechanism [are my IBM roots also showing?]). 

  The creator of tcsh was impressed by this feature and several
  others of TENEX and TOPS-20, and created a version of csh
  which mimicked them. 

Tim.

######

From: fdc@columbia.edu (Frank da Cruz)
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 11 Feb 2002 13:48:09 -0500
Organization: Columbia University
Lines: 37
Message-ID: <a493l9$dsu$1@watsol.cc.columbia.edu>
References: <GrDHML.Jso@world.std.com> <a48sd6$eva$18@bob.news.rcn.net> <90T98.30791$Wf1.6333289@ruti.visi.com> <a490u201v07@drn.newsguy.com>
NNTP-Posting-Host: watsol.cc.columbia.edu
X-Trace: newsmaster.cc.columbia.edu 1013453290 19553 128.59.39.139 (11 Feb 2002 18:48:10 GMT)
X-Complaints-To: postmaster@columbia.edu
NNTP-Posting-Date: 11 Feb 2002 18:48:10 GMT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!hub1.nntpserver.com!news-out.spamkiller.net!propagator-maxim!news-in.spamkiller.net!feed.newsfeeds.com!dfw-feed.news.verio.net!phl-feed.news.verio.net!news.verio.net!newsmaster.cc.columbia.edu!news.columbia.edu!news-not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10287

In article <a490u201v07@drn.newsguy.com>,
Tim Shoppa  <shoppa@trailing-edge.com> wrote:
: Frank daCruz of Columbia has pointed me towards a context-sensitive
: completion/help facility for use *within* a running program (after it
: has already begun execution), but AFAIK it doesn't work for shells in
: general. 
:
For the record:

 . Many Kermit programs do this, including MS-DOS Kermit, Kermit 95 (for
   Windows 9x/ME/NT/2000/XP) and C-Kermit for Unix, VMS, AOS/VS, VOS, Amiga,
   etc (not to mention forgotten versions like the ones for the Apple II and
   the Macintosh); thus in the heyday of diversity and Kermit as a universal
   connective tool, people on most platforms were exposed to it, or at least
   had the opportunity to be exposed.

 . The CCMD package is a COMND JSYS implementation in C that has been
   ported to at least Unix, DOS, and (I think) VMS.  It is used in the MM
   program (our C-language clone of TOPS-20 MM), but it can also be used
   in a general way, with any software.  Various people here started
   working on Unix versions of the TOPS-20 EXEC, the TOPS-20 REV program,
   etc, based on CCMD, but these were never finished or released.

Most people today prefer to click on things.  This is unfortunate because
you never have the chance to become an expert and go faster.  Groveling
though menus is fine for beginners but extremely labor intensive compared
to the COMND interface, in which you could get the menus, prompts, and
help if you wanted them, but could skip all that once you knew what you
were doing.  I think the COMND JSYS was one of the best ideas in
computing.

For completeness I should mention that we also had a complete COMND
package written in SAIL, but I'm afraid it's lost to the ages unless Tim
has unearthed it from an old DECUS tape.  After programming in C for the
last 18 years I still miss SAIL...

- Frank

######

Newsgroups: alt.sys.pdp10,comp.human-factors
From: cstacy@theworld.com (Christopher C. Stacy)
Subject: Re: command-line completion
Message-ID: <uy9hz4wpf.fsf@theworld.com>
Sender: cstacy@BONK
Date: Mon, 11 Feb 2002 19:58:36 GMT
References: <GrDHML.Jso@world.std.com>
NNTP-Posting-Host: ppp0c032.std.com
Organization: The World @ Software Tool & Die
X-Newsreader: Gnus v5.7/Emacs 20.7
Lines: 7
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.cwix.com!news.umass.edu!world!news
Xref: chonsp.franklin.ch alt.sys.pdp10:10283

>>>>> On Mon, 11 Feb 2002 14:30:21 GMT, Pete Wilson ("Pete") writes:
 Pete> Anybody know if that idea was ever widely used in other OSs?

Completion was used heavily in Genera, the operating system for the
Symbolics Lisp Machine.  It was based on the ideas and experience
from TOPS-20 completion, but was orders of magnitude fancier.

######

From: Tim Shoppa <shoppa@trailing-edge.com>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 11 Feb 2002 12:03:39 -0800
Organization: Trailing-Edge Technologies
Lines: 23
Message-ID: <a4982r02htm@drn.newsguy.com>
References: <GrDHML.Jso@world.std.com> <a48sd6$eva$18@bob.news.rcn.net> <90T98.30791$Wf1.6333289@ruti.visi.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu>
NNTP-Posting-Host: p-093.newsdawg.com
X-Newsreader: Direct Read News 2.91
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.media.kyoto-u.ac.jp!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!drn
Xref: chonsp.franklin.ch alt.sys.pdp10:10281

In article <a493l9$dsu$1@watsol.cc.columbia.edu>, fdc@columbia.edu says...
>For completeness I should mention that we also had a complete COMND
>package written in SAIL, but I'm afraid it's lost to the ages unless Tim
>has unearthed it from an old DECUS tape.  After programming in C for the
>last 18 years I still miss SAIL...

Seek and ye shall find:
http://pdp-10.trailing-edge.com/decuslib20-01/01/decus/20-0002/comnd.doc.html
http://pdp-10.trailing-edge.com/decuslib20-01/01/decus/20-0002/comnd.mss.html

http://pdp-10.trailing-edge.com/decuslib20-01/01/decus/20-0002/comnd.sai.html

Or look at all of DECUS 20-0002 (SAIL/FAIL):

http://pdp-10.trailing-edge.com/decus/20-2.html

Give the new search engine on the PDP-10 archive home page a try, Frank.  It
works wonders!

http://pdp-10.trailing-edge.com/

Tim.

######

From: Tim Shoppa <shoppa@trailing-edge.com>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 11 Feb 2002 12:06:11 -0800
Organization: Trailing-Edge Technologies
Lines: 23
Message-ID: <a4987j02ifq@drn.newsguy.com>
References: <GrDHML.Jso@world.std.com> <a48sd6$eva$18@bob.news.rcn.net> <90T98.30791$Wf1.6333289@ruti.visi.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu>
NNTP-Posting-Host: p-143.newsdawg.com
X-Newsreader: Direct Read News 2.91
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!howland.erols.net!newsfeed.cwix.com!wn2feed!worldnet.att.net!24.0.0.38!newshub2.rdc1.sfba.home.com!news.home.com!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!drn
Xref: chonsp.franklin.ch alt.sys.pdp10:10279

In article <a493l9$dsu$1@watsol.cc.columbia.edu>, fdc@columbia.edu says...
>For completeness I should mention that we also had a complete COMND
>package written in SAIL, but I'm afraid it's lost to the ages unless Tim
>has unearthed it from an old DECUS tape.  After programming in C for the
>last 18 years I still miss SAIL...

Seek and ye shall find:
http://pdp-10.trailing-edge.com/decuslib20-01/01/decus/20-0002/comnd.doc.html
http://pdp-10.trailing-edge.com/decuslib20-01/01/decus/20-0002/comnd.mss.html

http://pdp-10.trailing-edge.com/decuslib20-01/01/decus/20-0002/comnd.sai.html

Or look at all of DECUS 20-0002 (SAIL/FAIL):

http://pdp-10.trailing-edge.com/decus/20-2.html

Give the new search engine on the PDP-10 archive home page a try, Frank.  It
works wonders!

http://pdp-10.trailing-edge.com/

Tim.

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 11 Feb 2002 20:53:24 GMT
Organization: TSS Inc.
Lines: 18
Message-ID: <a49b04$1j6q$1@citadel.in.taronga.com>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net>
NNTP-Posting-Host: citadel.in.taronga.com
X-Trace: citadel.in.taronga.com 1013460804 52442 10.0.0.43 (11 Feb 2002 20:53:24 GMT)
X-Complaints-To: usenet@taronga.com
NNTP-Posting-Date: 11 Feb 2002 20:53:24 GMT
X-Newsreader: trn 4.0-test72 (19 April 1999)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.stanford.edu!news.kjsl.com!news.usenet2.org!citadel.in.taronga.com!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10304

In article <3C67EFC7.18E3E3C5@bartek.dontspamme.net>,
Arthur Krewat  <krewat@bartek.dontspamme.net> wrote:
>What use was command completion besides showing off to someone?

Provides "safe" verification that you know what you're doing.

It's really a subset of the "?" help function, though, and as you note
that's much more useful.

The ? behaviour is replicated, with greater or lesser faithfulness, in a
lot of DEC firmware, as well as Cisco and other network vendors, and of
course Kermit.

-- 
Rev. Peter da Silva, ULC.	                                 WWFD?

"Be conservative in what you generate, and liberal in what you accept"
	-- Matthew 10:16 (l.trans)

######

From: fdc@columbia.edu (Frank da Cruz)
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 11 Feb 2002 16:06:15 -0500
Organization: Columbia University
Lines: 31
Message-ID: <a49bo7$k81$1@watsol.cc.columbia.edu>
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu> <a4982r02htm@drn.newsguy.com>
NNTP-Posting-Host: watsol.cc.columbia.edu
X-Trace: newsmaster.cc.columbia.edu 1013461576 25690 128.59.39.139 (11 Feb 2002 21:06:16 GMT)
X-Complaints-To: postmaster@columbia.edu
NNTP-Posting-Date: 11 Feb 2002 21:06:16 GMT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!out.nntp.be!propagator-SanJose!in.nntp.be!news-in-sanjose!dfw-feed.news.verio.net!phl-feed.news.verio.net!news.verio.net!newsmaster.cc.columbia.edu!news.columbia.edu!news-not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10303

In article <a4982r02htm@drn.newsguy.com>,
Tim Shoppa  <shoppa@trailing-edge.com> wrote:
: In article <a493l9$dsu$1@watsol.cc.columbia.edu>, fdc@columbia.edu says...
: >For completeness I should mention that we also had a complete COMND
: >package written in SAIL, but I'm afraid it's lost to the ages unless Tim
: >has unearthed it from an old DECUS tape.  After programming in C for the
: >last 18 years I still miss SAIL...
: 
: Seek and ye shall find:
: http://pdp-10.trailing-edge.com/decuslib20-01/01/decus/20-0002/comnd.doc.html
: http://pdp-10.trailing-edge.com/decuslib20-01/01/decus/20-0002/comnd.mss.html
: 
: http://pdp-10.trailing-edge.com/decuslib20-01/01/decus/20-0002/comnd.sai.html
: 
That's insane.  I haven't seen that stuff in nearly 20 years.
You're a wild man.

: Or look at all of DECUS 20-0002 (SAIL/FAIL):
: 
: http://pdp-10.trailing-edge.com/decus/20-2.html
: 
So has anybody ported Sail to Unix yet? :-)

: Give the new search engine on the PDP-10 archive home page a try, Frank.  It
: works wonders!
: 
: http://pdp-10.trailing-edge.com/
: 
Yikes there's stuff in there I wrote it in 1978...

- Frank

######

From: Rich Alderson <alderson+news@panix.com>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 11 Feb 2002 19:55:25 -0500
Organization: Systems Administration, XKL LLC, Redmond WA 98052
Lines: 18
Sender: alderson+news@panix1.panix.com
Message-ID: <mddit93v7r6.fsf@panix1.panix.com>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net> <1013451786.72920@dns2.serv.net> <a4915p01vjo@drn.newsguy.com>
NNTP-Posting-Host: panix1.panix.com
X-Trace: news.panix.com 1013475329 15166 166.84.1.1 (12 Feb 2002 00:55:29 GMT)
X-Complaints-To: abuse@panix.com
NNTP-Posting-Date: 12 Feb 2002 00:55:29 GMT
X-Newsreader: Gnus v5.7/Emacs 20.7
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!out.nntp.be!propagator-SanJose!in.nntp.be!micro-heart-of-gold.mit.edu!bloom-beacon.mit.edu!panix!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10302

Tim Shoppa <shoppa@trailing-edge.com> writes:

> In article <1013451786.72920@dns2.serv.net>, Patrick says...
> >Several Unix shells have command completion, I suspect that TOPS-20
> >was at least one of the sources of inspiration.
> 
> It certainly was.  From "The T in TCSH":

[ snip ]

Does my name still appear in the man page credits?  That text was taken from a
posting by yours truly in alt.folklore.computers written more than 10 (and less
than 16) years ago; I was truly flattered when I noticed it in the SunOS 4.1
man pages!

-- 
Rich Alderson                                          alderson+news@panix.com
  "You get what anybody gets.  You get a lifetime."  --Death, of the Endless

######

From: Daniel B Dobkin <dbd@gnlawyers.com>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
References: <GrDHML.Jso@world.std.com> <a48sd6$eva$18@bob.news.rcn.net> <90T98.30791$Wf1.6333289@ruti.visi.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu>
User-Agent: MT-NewsWatcher/3.1 (PPC)
Message-ID: <dbd-800F0C.20203211022002@news2.srv.hcvlny.cv.net>
Lines: 31
Date: Tue, 12 Feb 2002 01:20:31 GMT
NNTP-Posting-Host: 24.46.115.64
X-Trace: news02.optonline.net 1013476831 24.46.115.64 (Mon, 11 Feb 2002 20:20:31 EST)
NNTP-Posting-Date: Mon, 11 Feb 2002 20:20:31 EST
Organization: Optimum Online
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sunqbc.risq.qc.ca!falcon.america.net!cyclone2.usenetserver.com!usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10313

In article <a493l9$dsu$1@watsol.cc.columbia.edu>,
 fdc@columbia.edu (Frank da Cruz) wrote:
>  . The CCMD package is a COMND JSYS implementation in C that has been
>    ported to at least Unix, DOS, and (I think) VMS.  It is used in the MM
>    program (our C-language clone of TOPS-20 MM), but it can also be used
>    in a general way, with any software.  Various people here started
>    working on Unix versions of the TOPS-20 EXEC, the TOPS-20 REV program,
>    etc, based on CCMD, but these were never finished or released.

I wrote an Exec for SunOS, using the CCMD routines, back around 1989-90;
I even added a set of TBADD/TBDEL/TBLUK calls: very useful for getting
back the same Emacs fork, for example, and generally for CONTINUEing
suspended forks.

It had a very basic command set --- about what was built into the MS-DOS
shell.  The idea was to let people accustomed to DOS migrate to SunOS
without having to learn anything about UNIX.

> Most people today prefer to click on things.  This is unfortunate because
> you never have the chance to become an expert and go faster.  Groveling
> though menus is fine for beginners but extremely labor intensive compared
> to the COMND interface, in which you could get the menus, prompts, and
> help if you wanted them, but could skip all that once you knew what you
> were doing.  I think the COMND JSYS was one of the best ideas in
> computing.

Dave Kashtan wrote an article for UNIX World about fifteen years ago that
said, in effect, "Command lines are where it's at."


\dbd

######

From: Carl Shapiro <cshapiro+spam@panix.com>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 11 Feb 2002 22:21:33 -0500
Organization: PANIX Public Access Internet and UNIX, NYC
Lines: 12
Sender: cshapiro@panix3.panix.com
Message-ID: <ouybsev5qrm.fsf@panix3.panix.com>
References: <GrDHML.Jso@world.std.com> <uy9hz4wpf.fsf@theworld.com>
NNTP-Posting-Host: panix3.panix.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Trace: reader2.panix.com 1013484091 14749 166.84.1.3 (12 Feb 2002 03:21:31 GMT)
X-Complaints-To: abuse@panix.com
NNTP-Posting-Date: Tue, 12 Feb 2002 03:21:31 +0000 (UTC)
User-Agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-out.spamkiller.net!propagator-maxim!news-in.spamkiller.net!hub1.nntpserver.com!bloom-beacon.mit.edu!panix!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10307

cstacy@theworld.com (Christopher C. Stacy) writes:

> >>>>> On Mon, 11 Feb 2002 14:30:21 GMT, Pete Wilson ("Pete") writes:
>  Pete> Anybody know if that idea was ever widely used in other OSs?
> 
> Completion was used heavily in Genera, the operating system for the
> Symbolics Lisp Machine.  It was based on the ideas and experience
> from TOPS-20 completion, but was orders of magnitude fancier.

Hey, don't forget CLIM, the Common Lisp Interface Manager, which also
has a completion facility!  Unsurprisingly, is extremely similar to
the Command Processor facility of Dynamic Windows.

######

From: klh@panix.com (Ken Harrenstien)
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 12 Feb 2002 01:03:38 -0500
Organization: PANIX Public Access Internet and UNIX, NYC
Lines: 20
Message-ID: <a4ab7q$rjm$1@panix2.panix.com>
References: <GrDHML.Jso@world.std.com> <90T98.30791$Wf1.6333289@ruti.visi.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu>
NNTP-Posting-Host: panix2.panix.com
X-Trace: news.panix.com 1013493819 19909 166.84.1.2 (12 Feb 2002 06:03:39 GMT)
X-Complaints-To: abuse@panix.com
NNTP-Posting-Date: 12 Feb 2002 06:03:39 GMT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-out.spamkiller.net!propagator-maxim!news-in.spamkiller.net!newsfeed.fast.net!howland.erols.net!panix!panix2.panix.com!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10299

In article <a493l9$dsu$1@watsol.cc.columbia.edu>,
Frank da Cruz <fdc@columbia.edu> wrote:
> . The CCMD package is a COMND JSYS implementation in C that has been
>   ported to at least Unix, DOS, and (I think) VMS.  It is used in the MM
>   program (our C-language clone of TOPS-20 MM), but it can also be used
>   in a general way, with any software.  Various people here started
>   working on Unix versions of the TOPS-20 EXEC, the TOPS-20 REV program,
>   etc, based on CCMD, but these were never finished or released.

If you don't already know this you might find it interesting...  while
at SRI-NIC, we actually back-ported the CCMD API to TOPS-20 (using
COMND% of course) so that a number of our NIC programs could be built
on both SunOS (using cc) and TOPS-20 (using KCC).  Ian Macky did most
of the work on that.

I still maintain/port my own copy to the various Unices I need to run
MM on.  Solaris, Tru64, FreeBSD, NetBSD, Linux, etc.  I didn't realize
anyone else still cared about it, if in fact that's true.

--Ken

######

Newsgroups: alt.sys.pdp10,comp.human-factors
From: Patricia Tressel <tressel@cs.washington.edu>
Subject: Re: command-line completion
Content-Type: text/plain; charset=us-ascii
X-Accept-Language: en
Sender: news@beaver.cs.washington.edu (News Subsystem)
Content-Transfer-Encoding: 7bit
Organization: Computer Science & Engineering, U of Washington, Seattle
Message-ID: <3C68FD44.79949089@cs.washington.edu>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net>
X-Mailer: Mozilla 4.79 [en] (X11; U; Linux 2.4.7-UP.26 i686)
Mime-Version: 1.0
X-Nntp-Posting-Host: beryllium.cs.washington.edu
Date: Tue, 12 Feb 2002 03:32:20 -0800
Lines: 62
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!out.nntp.be!propagator-SanJose!in.nntp.be!micro-heart-of-gold.mit.edu!uw-beaver!news
Xref: chonsp.franklin.ch alt.sys.pdp10:10301

Arthur Krewat wrote:
 
> What use was command completion...

A more general form of completion is word prediction, which is used (for
instance) in assistive text editors.

The main difference is that there may be a large number of words that
match the
prefix of characters already typed, and the editor uses such features as
word
frequencies and grammar to select just the most probable words to show
the user.
For command completion, there were few enough options that all could be
shown.

Since this is going to comp.human-factors as well as alt.sys.pdp10, I'll
add:

Word-predicting editors are most commonly used in one of two cases:

  The user has physical limitations that preclude typing at a reasonable
speed,
  e.g. the user may only be able to operate a single switch.  (Stephen
Hawking
  uses a predictive editor.)

  The user has a language difficulty.  (Predicting editors, and
especially ones
  that also suggest correct grammar, are used to help children who're
having
  trouble acquiring writing skills.)

I can imagine another use (though I haven't seen it suggested):

  Patients with RSI might use it to reduce the number of keystrokes they
need to
  type.

We're just starting up a project at the University of Washington to
provide a
user interface for completely paralyzed patients, using EEG as the input
device.
(There are several such projects already in progress at other
universities.)
Needless to say, word prediction (*and* command completion, and goal /
plan
recognition) will be of use...

In case you're interested in seeing how these editors present their word
(and
phrase) completions, there's a list at:
  www.cs.washington.edu/homes/tressel/bci/editors.html
All are GUI-based, so they tend to present the completion somewhere
other than
inline with the text being entered.

(If you do happen to look at the list, and notice any editors I've
omitted, let
me know -- this is intended to be an exhaustive list.  Thanks!)

-- Pat Tressel

######

Newsgroups: alt.sys.pdp10,comp.human-factors
From: Patricia Tressel <tressel@cs.washington.edu>
Subject: Re: command-line completion
Content-Type: text/plain; charset=us-ascii
X-Accept-Language: en
Sender: news@beaver.cs.washington.edu (News Subsystem)
Content-Transfer-Encoding: 7bit
Organization: Computer Science and Engineering, University of Washington
Message-ID: <3C690667.44CB3177@cs.washington.edu>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net> <a48sd6$eva$18@bob.news.rcn.net>
X-Mailer: Mozilla 4.79 [en] (X11; U; Linux 2.4.7-UP.26 i686)
Mime-Version: 1.0
X-Nntp-Posting-Host: beryllium.cs.washington.edu
Date: Tue, 12 Feb 2002 04:11:19 -0800
Lines: 59
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-out.spamkiller.net!propagator-maxim!news-in.spamkiller.net!tethys.csu.net!news-hog.berkeley.edu!ucberkeley!nntp.cs.ubc.ca!uw-beaver!news
Xref: chonsp.franklin.ch alt.sys.pdp10:10300

jmfbahciv@aol.com wrote:

> All I had to do was show a new user the ? and the $ command.
> They figured out how to use the system all on their own.

This is no exaggeration.  I was a systems programmer at at TOPS-10 shop
for many
years.  Once, we received a tape in DUMPER format.  We didn't have
anything that
would read it under TOPS-10, but we did have a working copy of
TOPS-20...on our
KLAD pack.  So we booted that, and my co-worker remembered the field
service
default password.  I'd never used TOPS-20, but I'd heard about $ and ?.

With just the inline help and prompts, I was able to: create an account
and
filesystem, mount the tape, read it onto disk with DUMPER, find an
appropriate
tape output program (I forget what it was...), mount and label a new
tape, and
write the tape contents back out.  (And, yes, it was readable under
TOPS-10. ;-)

I was *very* pleased with the $ and ?.  I can't imagine doing the same
with man
pages or Windows help -- those require that the user *know something
already* (or
at least have good luck at guessing) to get help.  With VMS help, at
least one
got a top-down list of ~all the options, from which one could, at worst,
do a
complete traversal.  In the above situation, I did start off with an
equivalent
choice -- I typed ? at the bare command prompt, the moral equivalent of
typing
"help" to VMS DCL.  But because the VMS help data was separate from the
command
interpreter, it was possible for the two to get out of sync.  For COMND,
even
if the author of the COMND driver files didn't include any help info,
one still
got authoritative lists of what one could type next.

Tim Shoppa wrote:

> Frank daCruz of Columbia has pointed me towards a context-sensitive
> completion/help facility for use *within* a running program (after it
> has already begun execution), but AFAIK it doesn't work for shells in general.

Just for completeness sake:  Unless I'm spacing out totally, there was a
user-mode
library for COMND-style input that came with TOPS-10.  I certainly
remember
modifying the COMND driver files for a configuration tool that was part
of the
distribution...

-- Pat Tressel

######

From: Alan Greig <a.greig@virgin.net>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: Tue, 12 Feb 2002 14:03:46 +0000
Organization: http://extra.newsguy.com
Lines: 35
Message-ID: <d18i6ucqe7ir0jv19t00n1632498ao79ej@4ax.com>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net> <a48sd6$eva$18@bob.news.rcn.net> <3C690667.44CB3177@cs.washington.edu>
NNTP-Posting-Host: p-874.newsdawg.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Newsreader: Forte Agent 1.8/32.548
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-out.nuthinbutnews.com!propagator!feed2.newsfeeds.com!newsfeeds.com!cyclone-sf.pbi.net!165.113.238.17!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!enews4
Xref: chonsp.franklin.ch alt.sys.pdp10:10311

On Tue, 12 Feb 2002 04:11:19 -0800, Patricia Tressel
<tressel@cs.washington.edu> wrote:

>jmfbahciv@aol.com wrote:
>
>With just the inline help and prompts, I was able to: create an account
>and
>filesystem, mount the tape, read it onto disk with DUMPER, find an
>appropriate
>tape output program (I forget what it was...), mount and label a new
>tape, and

DUMPER in "interchange" mode presumably. BACKUP could read DUMPER
interchange format tapes and vice versa.

>write the tape contents back out.  (And, yes, it was readable under
>TOPS-10. ;-)
>
>I was *very* pleased with the $ and ?.  I can't imagine doing the same
>
>Just for completeness sake:  Unless I'm spacing out totally, there was a
>user-mode
>library for COMND-style input that came with TOPS-10.  I certainly
>remember

Part of GLXLIB wasn't it? OPR on TOPS-10 used it.

>modifying the COMND driver files for a configuration tool that was part
>of the
>distribution...
>
>-- Pat Tressel

--
Alan

######

From: fdc@columbia.edu (Frank da Cruz)
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 12 Feb 2002 11:16:18 -0500
Organization: Columbia University
Lines: 64
Message-ID: <a4bf4i$99k$1@watsol.cc.columbia.edu>
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu> <a4ab7q$rjm$1@panix2.panix.com>
NNTP-Posting-Host: watsol.cc.columbia.edu
X-Trace: newsmaster.cc.columbia.edu 1013530579 12802 128.59.39.139 (12 Feb 2002 16:16:19 GMT)
X-Complaints-To: postmaster@columbia.edu
NNTP-Posting-Date: 12 Feb 2002 16:16:19 GMT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!howland.erols.net!panix!newsmaster.cc.columbia.edu!news.columbia.edu!news-not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10305

In article <a4ab7q$rjm$1@panix2.panix.com>,
Ken Harrenstien <klh@panix.com> wrote:
: In article <a493l9$dsu$1@watsol.cc.columbia.edu>,
: Frank da Cruz <fdc@columbia.edu> wrote:
: > . The CCMD package is a COMND JSYS implementation in C that has been
: >   ported to at least Unix, DOS, and (I think) VMS.  It is used in the MM
: >   program (our C-language clone of TOPS-20 MM), but it can also be used
: >   in a general way, with any software.  Various people here started
: >   working on Unix versions of the TOPS-20 EXEC, the TOPS-20 REV program,
: >   etc, based on CCMD, but these were never finished or released.
: 
: If you don't already know this you might find it interesting...  while
: at SRI-NIC, we actually back-ported the CCMD API to TOPS-20 (using
: COMND% of course) so that a number of our NIC programs could be built
: on both SunOS (using cc) and TOPS-20 (using KCC).  Ian Macky did most
: of the work on that.
: 
If I ever knew that, I forgot.

: I still maintain/port my own copy to the various Unices I need to run
: MM on.  Solaris, Tru64, FreeBSD, NetBSD, Linux, etc.  I didn't realize
: anyone else still cared about it, if in fact that's true.
: 
Gee, maybe it's time to gather up all the pieces and put it back together.
At Columbia, CCMD got the Y2K and y2K+1 patches (announced here) and that
was about the end of it.

Meanwhile I still use MM, as do other survivors from the DEC-20 era.
Recently Columbia migrated its user file system to humongous NFS servers,
and now we are all faced with the "NFS bug": Edit a message in EMACS under
MM, save the message in EMACS, exit EMACS, return to MM, find the result
truncated, sometimes partially, sometimes totally.  EMACS has told Unix
(Solaris in this case, but it also happens elsewhere) to save the file and
then exits.  Does EMACS do an fsync()?  (I don't know.)  In any case, since
this is an NFS disk, the file is written when Unix feels like getting
around to it AND when NFS feels like getting around to it, if ever, and MM
often reads it back before it is fully written.

The problem is alleviated somewhat if you add a pause to MM at the point
where it gets control back from EMACS but before it reads in the temp file.
But since there is no way of knowing how long the pause must be, it can
(and does) still fail, no matter what you choose.

On "bad NFS days", you can lose one message after another this way, so
I started adding code to MM to try to catch the situation and fix it, but
so far have not succeeded (it's only a spare time project and I don't have
any).  I know Mark says NFS and Pine just don't mix and maybe the same is
true of MM and nothing can be done about it.

Once this is fixed, my list for MM includes:

 . Make it allow 8-bit data (Latin-1, UTF-8, etc) in messages and headers.
 . Give it a command to announce the charset when sending a message.
 . Make it do something reasonable with incoming charset announcers.

The latter might be to send an ISO-2022 character-set designation escape
sequence if the charset is a standard one (i.e. registered in the ISO
register), thus allowing the terminal, console, or emulator to switch
automatically to the appropriate character set for each message.

Actually I have much longer list but will probably never get to it so no
point including it.

- Frank

######

Path: chonsp.franklin.ch!not-for-mail
From: Neil Franklin <neil@franklin.ch.remove>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 12 Feb 2002 23:45:22 +0100
Organization: My own Private Self
Lines: 25
Message-ID: <6uwuxiwc8t.fsf@chonsp.franklin.ch>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net> <1013451786.72920@dns2.serv.net> <a4915p01vjo@drn.newsguy.com> <mddit93v7r6.fsf@panix1.panix.com>
NNTP-Posting-Host: chonsp.franklin.ch
X-Trace: chonsp.franklin.ch 1013553925 2040 10.0.3.2 (12 Feb 2002 22:45:25 GMT)
X-Complaints-To: news@chonsp.franklin.ch
NNTP-Posting-Date: 12 Feb 2002 22:45:25 GMT
X-Newsreader: Gnus v5.7/Emacs 20.4
Xref: chonsp.franklin.ch alt.sys.pdp10:10314

Rich Alderson <alderson+news@panix.com> writes:

> Tim Shoppa <shoppa@trailing-edge.com> writes:
>
> > It certainly was.  From "The T in TCSH":
>
> Does my name still appear in the man page credits?

neil@chonsp 23:42:08 ~> uname -a
Linux chonsp 2.2.13 #127 Thu Oct 21 13:13:20 CDT 1999 i586 unknown
neil@chonsp 23:42:11 ~> date
Tue Feb 12 23:42:13 MET 2002
neil@chonsp 23:42:13 ~> man tcsh | grep Alderson
       Richard  M. Alderson III, for writing the `T in tcsh' sec-
neil@chonsp 23:42:23 ~>

Still in the tcsh manual on this Linux system. Together with many others.

And the T IN TCSH text is in there too, full as you posted it.


--
Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/
Hacker, Unix Guru, El Eng HTL/BSc, Sysadmin, Archer, Roleplayer
- Intellectual Property is Intellectual Robbery

######

From: jmfbahciv@aol.com
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: Wed, 13 Feb 02 09:18:12 GMT
Organization: UltraNet Communications, Inc.
Lines: 17
Message-ID: <a4dirg$sac$1@bob.news.rcn.net>
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu> <a4ab7q$rjm$1@panix2.panix.com> <a4bf4i$99k$1@watsol.cc.columbia.edu> <3C69C053.889CB9A0@bartek.dontspamme.net>
X-Trace: UmFuZG9tSVaeIB3BcIAJWCou31nTDYoe1SAnszrSoHwTvRMrVXS69hjevS/znVH9
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 13 Feb 2002 11:32:00 GMT
X-Newsreader:  News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-255-33
Xref: chonsp.franklin.ch alt.sys.pdp10:10319

In article <3C69C053.889CB9A0@bartek.dontspamme.net>,
   Arthur Krewat <krewat@bartek.dontspamme.net> wrote:
>
>My original key point was ? was usefull, but I fail to see
>the usefulness of command-completion. 
>
>I guess I spent too much time with limited TOPS-10 access,
>so I had to make sure I had worked out what I wanted to do
>beforehand :)

Sure.  The -10 was for the experienced who wanted to get work
done.  The -20 was for newbies.  I'd better include a :-) for
Mark.  ;-)

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: Wed, 13 Feb 02 13:11:41 GMT
Organization: UltraNet Communications, Inc.
Lines: 22
Message-ID: <a4e0h7$8is$2@bob.news.rcn.net>
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com>
	 <a493l9$dsu$1@watsol.cc.columbia.edu> <a4ab7q$rjm$1@panix2.panix.com>
	 <a4bf4i$99k$1@watsol.cc.columbia.edu> <3C69C053.889CB9A0@bartek.dontspamme.net> <Pine.NXT.4.50.0202121751400.5148-100000@Tomobiki-Cho.CAC.Washington.EDU> <3C6A7193.C54E957B@bartek.dontspamme.net>
X-Trace: UmFuZG9tSVa4E/LBGVdfc6i0usBCUt0eZuAN+Pll8WJrRJmXS2Mh4U1tXTUyk4uO
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 13 Feb 2002 15:25:27 GMT
X-Newsreader:  News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!209-122-233-240
Xref: chonsp.franklin.ch alt.sys.pdp10:10316

In article <3C6A7193.C54E957B@bartek.dontspamme.net>,
   Arthur Krewat <krewat@bartek.dontspamme.net> wrote:
>Mark Crispin wrote:
<snip>

>> People who have never lived with command completion can not appreciate 
who
>> wonderful it is to have it.  This is to be expected.  There are many
>> examples in this world of seemingly "useless" luxuries which are
>> unappreciated by those who have never enjoyed them, and are inadequately
>> appreciated by those who have enjoyed them until the day they are lost.
>
>You are quite right - and I didn't start this to piss people off, it was
>truly an intellectual curiosity.
I think the question produced an excellent discussion.  I hadn't
thought about my training technique for a -20 user in years.
Back then, I had assumed that everybody would build their command
structure so that self-training was possible.  Boy!  Was I wrong!!!

/BAH

Subtract a hundred and four for e-mail.

######

From: Arthur Krewat <krewat@bartek.dontspamme.net>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Organization: Kilonet.net
Lines: 9
Message-ID: <3C69C053.889CB9A0@bartek.dontspamme.net>
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu> <a4ab7q$rjm$1@panix2.panix.com> <a4bf4i$99k$1@watsol.cc.columbia.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.78 [en] (X11; U; SunOS 5.8 i86pc)
X-Accept-Language: en
Date: Wed, 13 Feb 2002 01:28:00 GMT
NNTP-Posting-Host: 24.186.100.134
X-Trace: news02.optonline.net 1013563680 24.186.100.134 (Tue, 12 Feb 2002 20:28:00 EST)
NNTP-Posting-Date: Tue, 12 Feb 2002 20:28:00 EST
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cyclone2.usenetserver.com!usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10324


My original key point was ? was usefull, but I fail to see
the usefulness of command-completion. 

I guess I spent too much time with limited TOPS-10 access,
so I had to make sure I had worked out what I wanted to do
beforehand :)

aak

######

From: "JD" <dyson@jdyson.com>
Newsgroups: alt.sys.pdp10,comp.human-factors
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu> <a4ab7q$rjm$1@panix2.panix.com> <a4bf4i$99k$1@watsol.cc.columbia.edu> <3C69C053.889CB9A0@bartek.dontspamme.net> <Pine.NXT.4.50.0202121751400.5148-100000@Tomobiki-Cho.CAC.Washington.EDU>
Subject: Re: command-line completion
Lines: 18
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2600.0000
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000
Message-ID: <M_ja8.568$K%4.139464@news1.iquest.net>
Date: Tue, 12 Feb 2002 21:24:07 -0500
NNTP-Posting-Host: 198.70.149.90
X-Complaints-To: news-abuse@iquest.net
X-Trace: news1.iquest.net 1013566508 198.70.149.90 (Tue, 12 Feb 2002 21:15:08 EST)
NNTP-Posting-Date: Tue, 12 Feb 2002 21:15:08 EST
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!rcn!dca6-feed2.news.digex.net!intermedia!newsfeed1.cidera.com!Cidera!iquest!news1.iquest.net!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10321


"Mark Crispin" <mrc@CAC.Washington.EDU> wrote in message
news:Pine.NXT.4.50.0202121751400.5148-100000@Tomobiki-Cho.CAC.Washington.EDU...
> On Wed, 13 Feb 2002, Arthur Krewat wrote:
> > My original key point was ? was usefull, but I fail to see
> > the usefulness of command-completion.
>
> If your fingers are trained for command completion, then when a surprise
> comes up, you can recover instead of having to retype the whole damn
> command.
>
That is what I use command line editing under ksh/bash/(sh(on BSD)) for.  In
vi mode (you can also use emacs), one can do ESC-k with k's until the right
command to be edited, then use subsequent vi, emacs commands to modify
the command line.

John

######

From: "JD" <dyson@jdyson.com>
Newsgroups: alt.sys.pdp10,comp.human-factors
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu> <a4ab7q$rjm$1@panix2.panix.com> <a4bf4i$99k$1@watsol.cc.columbia.edu> <3C69C053.889CB9A0@bartek.dontspamme.net> <Pine.NXT.4.50.0202121751400.5148-100000@Tomobiki-Cho.CAC.Washington.EDU> <M_ja8.568$K%4.139464@news1.iquest.net> <Pine.NXT.4.50.0202121830050.5235-100000@Tomobiki-Cho.CAC.Washington.EDU>
Subject: Re: command-line completion
Lines: 22
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2600.0000
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000
Message-ID: <SDka8.573$K%4.139868@news1.iquest.net>
Date: Tue, 12 Feb 2002 22:07:57 -0500
NNTP-Posting-Host: 198.70.149.90
X-Complaints-To: news-abuse@iquest.net
X-Trace: news1.iquest.net 1013569138 198.70.149.90 (Tue, 12 Feb 2002 21:58:58 EST)
NNTP-Posting-Date: Tue, 12 Feb 2002 21:58:58 EST
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!newsfeed1.cidera.com!Cidera!iquest!news1.iquest.net!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10320


"Mark Crispin" <mrc@CAC.Washington.EDU> wrote in message
news:Pine.NXT.4.50.0202121830050.5235-100000@Tomobiki-Cho.CAC.Washington.EDU...
> On Tue, 12 Feb 2002, JD wrote:
> > > > My original key point was ? was usefull, but I fail to see
> > > > the usefulness of command-completion.
> > > If your fingers are trained for command completion, then when a surprise
> > > comes up, you can recover instead of having to retype the whole damn
> > > command.
> > That is what I use command line editing under ksh/bash/(sh(on BSD)) for.  In
> > vi mode (you can also use emacs), one can do ESC-k with k's until the right
> > command to be edited, then use subsequent vi, emacs commands to modify
> > the command line.
>
> Consider how many iterations you have to go through when it turns out that
> you're completely mistaken about the form of that command.
>
Oh, sort of an AI type thing?   :-).   Instead of editing a nice, long pipe command,
the shell automatically reads my mind :-).

John

######

From: Tim Shoppa <shoppa@trailing-edge.com>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 12 Feb 2002 19:31:42 -0800
Organization: Trailing-Edge Technologies
Lines: 15
Message-ID: <a4cmmu018ad@drn.newsguy.com>
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu> <a4ab7q$rjm$1@panix2.panix.com> <a4bf4i$99k$1@watsol.cc.columbia.edu> <3C69C053.889CB9A0@bartek.dontspamme.net> <Pine.NXT.4.50.0202121751400.5148-100000@Tomobiki-Cho.CAC.Washington.EDU> <M_ja8.568$K%4.139464@news1.iquest.net> <Pine.NXT.4.50.0202121830050.5235-100000@Tomobiki-Cho.CAC.Washington.EDU> <SDka8.573$K%4.139868@news1.iquest.net>
NNTP-Posting-Host: p-106.newsdawg.com
X-Newsreader: Direct Read News 2.91
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!cyclone.bc.net!logbridge.uoregon.edu!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!drn
Xref: chonsp.franklin.ch alt.sys.pdp10:10317

In article <SDka8.573$K%4.139868@news1.iquest.net>, "JD" says...
>Oh, sort of an AI type thing?   :-).   Instead of editing a nice, long pipe
>command,
>the shell automatically reads my mind :-).

No, it's more like a good old mechanical interlocking.  You are
given a very positive "feel" for each step of the command composition
and are given very definite cues indicating what is expected to
come next.

It isn't you dictating to the machine; it's not the machine reading your
mind; it's you and the machine working together to get the job done.

Tim.

######

From: Enrico Badella <enrico.badella@softstar.it>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: Wed, 13 Feb 2002 12:54:25 +0100
Organization: SoftStar
Lines: 34
Message-ID: <3C6A53F1.7894CDA2@softstar.it>
References: <GrDHML.Jso@world.std.com> <a48sd6$eva$18@bob.news.rcn.net> <90T98.30791$Wf1.6333289@ruti.visi.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu>
NNTP-Posting-Host: 193.42.2.4
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.75 [en] (WinNT; U)
X-Accept-Language: en,it
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!newsfeed.gamma.ru!Gamma.RU!Radio-MSU.net!mtu.ru!teleglobe.net!teleglobe.net!server-b.cs.interbusiness.it!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10315



Frank da Cruz wrote:
> 
> In article <a490u201v07@drn.newsguy.com>,
> Tim Shoppa  <shoppa@trailing-edge.com> wrote:
> : Frank daCruz of Columbia has pointed me towards a context-sensitive
> : completion/help facility for use *within* a running program (after it

 [snip]

> Most people today prefer to click on things.  This is unfortunate because
> you never have the chance to become an expert and go faster.  Groveling
> though menus is fine for beginners but extremely labor intensive compared

Let me add that they prefer been sloppy instead of using their brain; in a 
project I'm involved the decision to use an m$ box instead of unix one was 
made because the developers of the web pages wouldn't have to care about 
case sensitive file names. This made me wonder why I'm still working in 
IT and not becoming a drain cleaner.

e.

 
========================================================================
Enrico Badella                       email:   enrico.badella@softstar.it
Soft*Star srl                                 eb@vax.cnuce.cnr.it
InterNetworking Specialists          tel:     +39-011-746092            
Via Camburzano 9                     fax:     +39-011-746487
10143 Torino, Italy

  Wanted, for hobbyist use, any type of PDP and microVAX hardware,software,
  manuals,schematics,etc. and DEC-10 docs or manuals
==========================================================================

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 13 Feb 2002 12:46:55 GMT
Organization: TSS Inc.
Lines: 15
Message-ID: <a4dn7v$t71$1@citadel.in.taronga.com>
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu> <3C6A53F1.7894CDA2@softstar.it>
NNTP-Posting-Host: citadel.in.taronga.com
X-Trace: citadel.in.taronga.com 1013604415 29921 10.0.0.43 (13 Feb 2002 12:46:55 GMT)
X-Complaints-To: usenet@taronga.com
NNTP-Posting-Date: 13 Feb 2002 12:46:55 GMT
X-Newsreader: trn 4.0-test72 (19 April 1999)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.stanford.edu!news.kjsl.com!news.usenet2.org!citadel.in.taronga.com!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10323

In article <3C6A53F1.7894CDA2@softstar.it>,
Enrico Badella  <enrico.badella@softstar.it> wrote:
>Let me add that they prefer been sloppy instead of using their brain; in a 
>project I'm involved the decision to use an m$ box instead of unix one was 
>made because the developers of the web pages wouldn't have to care about 
>case sensitive file names.

If they can't find a UNIX based webserver that can't be convinced to deal
with file names in a case-insensitive fashion, I'll write one for them.

-- 
Rev. Peter da Silva, ULC.	                                 WWFD?

"Be conservative in what you generate, and liberal in what you accept"
	-- Matthew 10:16 (l.trans)

######

From: Arthur Krewat <krewat@bartek.dontspamme.net>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Organization: Kilonet.net
Lines: 19
Message-ID: <3C6A6E6F.32D6EC08@bartek.dontspamme.net>
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu> <a4ab7q$rjm$1@panix2.panix.com> <a4bf4i$99k$1@watsol.cc.columbia.edu> <3C69C053.889CB9A0@bartek.dontspamme.net> <a4dirg$sac$1@bob.news.rcn.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.78 [en] (X11; U; SunOS 5.8 i86pc)
X-Accept-Language: en
Date: Wed, 13 Feb 2002 13:52:56 GMT
NNTP-Posting-Host: 24.186.100.134
X-Trace: news02.optonline.net 1013608376 24.186.100.134 (Wed, 13 Feb 2002 08:52:56 EST)
NNTP-Posting-Date: Wed, 13 Feb 2002 08:52:56 EST
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cyclone2.usenetserver.com!usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10326

jmfbahciv@aol.com wrote:
> 
> In article <3C69C053.889CB9A0@bartek.dontspamme.net>,
>    Arthur Krewat <krewat@bartek.dontspamme.net> wrote:
> >
> >My original key point was ? was usefull, but I fail to see
> >the usefulness of command-completion.
> >
> >I guess I spent too much time with limited TOPS-10 access,
> >so I had to make sure I had worked out what I wanted to do
> >beforehand :)
> 
> Sure.  The -10 was for the experienced who wanted to get work
> done.  The -20 was for newbies.  I'd better include a :-) for
> Mark.  ;-)

ROTFL!

aak

######

From: Arthur Krewat <krewat@bartek.dontspamme.net>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Organization: Kilonet.net
Lines: 41
Message-ID: <3C6A7193.C54E957B@bartek.dontspamme.net>
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com>
	 <a493l9$dsu$1@watsol.cc.columbia.edu> <a4ab7q$rjm$1@panix2.panix.com>
	 <a4bf4i$99k$1@watsol.cc.columbia.edu> <3C69C053.889CB9A0@bartek.dontspamme.net> <Pine.NXT.4.50.0202121751400.5148-100000@Tomobiki-Cho.CAC.Washington.EDU>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.78 [en] (X11; U; SunOS 5.8 i86pc)
X-Accept-Language: en
Date: Wed, 13 Feb 2002 14:02:57 GMT
NNTP-Posting-Host: 24.186.100.134
X-Trace: news02.optonline.net 1013608977 24.186.100.134 (Wed, 13 Feb 2002 09:02:57 EST)
NNTP-Posting-Date: Wed, 13 Feb 2002 09:02:57 EST
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cyclone2.usenetserver.com!usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10325

Mark Crispin wrote:
> 
> On Wed, 13 Feb 2002, Arthur Krewat wrote:
> > My original key point was ? was usefull, but I fail to see
> > the usefulness of command-completion.
> 
> If your fingers are trained for command completion, then when a surprise
> comes up, you can recover instead of having to retype the whole damn
> command.
> 
> The best example of such a surprise is an abbreviation which ceases to
> work because a new command was added which makes that abbreviation
> ambiguous.  Or a change in syntax to the command.

Interesting - I just realized that I was never a "user" of a machine
that I didn't administer, therefor no suprise command changes. And the
only TOPS-20 experience I had was hacking the Internet when it first
went IP. The ? was very usefull in this case :)

> People who have never lived with command completion can not appreciate who
> wonderful it is to have it.  This is to be expected.  There are many
> examples in this world of seemingly "useless" luxuries which are
> unappreciated by those who have never enjoyed them, and are inadequately
> appreciated by those who have enjoyed them until the day they are lost.

You are quite right - and I didn't start this to piss people off, it was
truly an intellectual curiosity.
 
> I also understand that there is a tribe of "human factors engineers" whose
> profession it is to tell people that they don't really want what they say
> they want.  I don't bother arguing with that tribe any more; I just don't
> buy their products.

I used to try to do that in user communities, especially CAD people.
I was able to deflect the course of some people, but there were sticks
in the mud you couldn't budge. I eventually gave up and just gave people
what they wanted - after they cried that's not what they really wanted,
I did it my way and they conformed. An extra step, but much less fighting
in user group meetings :)

aak

######

Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
References: <GrDHML.Jso@world.std.com> <M_ja8.568$K%4.139464@news1.iquest.net> <Pine.NXT.4.50.0202121830050.5235-100000@Tomobiki-Cho.CAC.Washington.EDU> <SDka8.573$K%4.139868@news1.iquest.net>
Sender: a.nonymous@somewhere.invalid
From: a.nonymous@somewhere.invalid
Originator: a.nonymous@somewhere.invalid
Organization: Not Much
X-Newsreader: trn 4.0-test69 (20 September 1998)
Originator: root@c-ns. (Operator)
Lines: 27
Message-ID: <GFza8.959$A%3.16025@ord-read.news.verio.net>
Date: Wed, 13 Feb 2002 20:04:54 GMT
NNTP-Posting-Host: 207.241.52.60
X-Complaints-To: abuse@verio.net
X-Trace: ord-read.news.verio.net 1013630694 207.241.52.60 (Wed, 13 Feb 2002 20:04:54 GMT)
NNTP-Posting-Date: Wed, 13 Feb 2002 20:04:54 GMT
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-feeds.jump.net!dfw-peer!news.verio.net!ord-read.news.verio.net.POSTED!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10333

In article <SDka8.573$K%4.139868@news1.iquest.net>,
JD <dyson@jdyson.com> wrote:
>
>"Mark Crispin" <mrc@CAC.Washington.EDU> wrote in message
>news:Pine.NXT.4.50.0202121830050.5235-100000@Tomobiki-Cho.CAC.Washington.EDU...
>> On Tue, 12 Feb 2002, JD wrote:
>> > > > My original key point was ? was usefull, but I fail to see
>> > > > the usefulness of command-completion.
>> > > If your fingers are trained for command completion, then when a surprise
>> > > comes up, you can recover instead of having to retype the whole damn
>> > > command.
>> > That is what I use command line editing under ksh/bash/(sh(on BSD)) for.  In
>> > vi mode (you can also use emacs), one can do ESC-k with k's until the right
>> > command to be edited, then use subsequent vi, emacs commands to modify
>> > the command line.
>>
>> Consider how many iterations you have to go through when it turns out that
>> you're completely mistaken about the form of that command.
>>
>Oh, sort of an AI type thing?   :-).   Instead of editing a nice, long
>pipe command,
>the shell automatically reads my mind :-).

Sorry, no.    ROM does _not_ stand for <R>ead <O>perator's <M>ind.


{grins, ducks. and *runs* away}

######

From: jmfbahciv@aol.com
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: Fri, 15 Feb 02 08:32:25 GMT
Organization: UltraNet Communications, Inc.
Lines: 67
Message-ID: <a4iou9$ss5$1@bob.news.rcn.net>
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com>
 <a493l9$dsu$1@watsol.cc.columbia.edu> <a4ab7q$rjm$1@panix2.panix.com>
 <a4bf4i$99k$1@watsol.cc.columbia.edu> <3C69C053.889CB9A0@bartek.dontspamme.net>
 <a4dirg$sac$1@bob.news.rcn.net> <Pine.NXT.4.50.0202131724210.6067-100000@Tomobiki-Cho.CAC.Washington.EDU>
X-Trace: UmFuZG9tSVYKpnqZcBoJZKZt1fX0V2pLFUnG1rc+prrpD09etwO9TG+QUEChGnSQ
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 15 Feb 2002 10:46:33 GMT
X-Newsreader:  News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-245-212
Xref: chonsp.franklin.ch alt.sys.pdp10:10345

In article 
<Pine.NXT.4.50.0202131724210.6067-100000@Tomobiki-Cho.CAC.Washington.EDU>,
   Mark Crispin <mrc@CAC.Washington.EDU> wrote:
>On Wed, 13 Feb 2002 jmfbahciv@aol.com wrote:
>> Sure.  The -10 was for the experienced who wanted to get work
>> done.  The -20 was for newbies.  I'd better include a :-) for
>> Mark.  ;-)
>
>Yah, you had better include a :-) on that Barb.

<GRIN>  

>
>I could say, "The -10 was for people who only did single monolithic tasks.

Only for those who didn't know about batch.

>The -20 was for people who did multi-tasking."

Yup.  They would type one COMPILE command and then catch up
on their Notebook Updates while waiting.  ;-)

>
>On the -10, the shell was built into the kernel.  On the -20, the shell
>was an ordinary user program, as it is on most modern operating systems.

I never did figure out a good methodology to use this feature. 
It didn't match my style of work, I think.

>As on UNIX, people could (and *did*) use alternative shells.
>
>The -20 had I/O redirection, the -10 did not.

What do you mean by this?

>
>In terms of comparison to UNIX:
>
>The -10's shell was much more primitive than the UNIX shell.  The -20's
>shell was quite a bit better, but didn't measure up to the UNIX shell or
>for that matter to the full capability of the -20's operating system.
>
>The -10's kernel was more primitive than the UNIX kernel.  With one
>exception, the -20's kernel was quite a bit more advanced than the UNIX
>kernel.
>
>That exception had to do with processes; the -20 had forks which shared a
>great deal of context (including the connected directory) with other forks
>in the process tree and (unlike UNIX) a fork could not be separated from
>its tree to be an independent tree.  However, -20 forks were much more
>independent of each other than UNIX threads.  This particular issue is
>irrelevant on the -10, since on the -10 you only had a single job which
>represented a single virtual machine and got recycled whenever you ran a
>new program.

The -10 was slowly getting acclimated when we finally got a 
programmer who could change SCNSER without breaking the world.
He did implement PUSH/POP but I never found a reason to use it.

I will say, Mark ;-), that if I had to write a user mode program
in MACRO, the UUO^WJSYS interface was the best on the -20.
We never did figure out how to document using the SCAN interface
and no monies were allotted to work on that one.

/BAH

Subtract a hundred and four for e-mail.

######

From: "Zane H. Healy" <healyzh@shell1.aracnet.com>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 14 Feb 2002 02:10:49 GMT
Organization: Aracnet
Lines: 16
Message-ID: <a4f6b90efd@enews1.newsguy.com>
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu> <3C6A53F1.7894CDA2@softstar.it> <a4dn7v$t71$1@citadel.in.taronga.com>
NNTP-Posting-Host: p-543.newsdawg.com
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686))
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!sunqbc.risq.qc.ca!headwall.stanford.edu!newsfeed.stanford.edu!logbridge.uoregon.edu!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10352

In alt.sys.pdp10 Peter da Silva <peter@taronga.com> wrote:
> In article <3C6A53F1.7894CDA2@softstar.it>,
> Enrico Badella  <enrico.badella@softstar.it> wrote:
>>Let me add that they prefer been sloppy instead of using their brain; in a 
>>project I'm involved the decision to use an m$ box instead of unix one was 
>>made because the developers of the web pages wouldn't have to care about 
>>case sensitive file names.

> If they can't find a UNIX based webserver that can't be convinced to deal
> with file names in a case-insensitive fashion, I'll write one for them.

Unless I'm mistaken webservers running on OpenVMS don't care about case
sensitivity, and OpenVMS is one H*** of a lot more secure than Microslop
OS's!

			Zane

######

From: jmfbahciv@aol.com
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: Sat, 16 Feb 02 09:59:11 GMT
Organization: UltraNet Communications, Inc.
Lines: 86
Message-ID: <a4lid8$6bt$3@bob.news.rcn.net>
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com>
 <a493l9$dsu$1@watsol.cc.columbia.edu> <a4ab7q$rjm$1@panix2.panix.com>
 <a4bf4i$99k$1@watsol.cc.columbia.edu> <3C69C053.889CB9A0@bartek.dontspamme.net>
 <a4dirg$sac$1@bob.news.rcn.net> <Pine.NXT.4.50.0202131724210.6067-100000@Tomobiki-Cho.CAC.Washington.EDU>
 <a4iou9$ss5$1@bob.news.rcn.net> <Pine.NXT.4.50.0202150936580.7590-100000@Tomobiki-Cho.CAC.Washington.EDU>
X-Trace: UmFuZG9tSVZuL7ExFDh1lrleI5i0rt32pY9oxLzDOGiJW+wgM8h1qrFIT1xE8MiL
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 16 Feb 2002 12:13:28 GMT
X-Newsreader:  News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-245-42
Xref: chonsp.franklin.ch alt.sys.pdp10:10380

In article 
<Pine.NXT.4.50.0202150936580.7590-100000@Tomobiki-Cho.CAC.Washington.EDU>,
   Mark Crispin <mrc@CAC.Washington.EDU> wrote:
>On Fri, 15 Feb 2002 jmfbahciv@aol.com wrote:
>> >The -20 was for people who did multi-tasking."
>> Yup.  They would type one COMPILE command and then catch up
>> on their Notebook Updates while waiting.  ;-)
>
>That was on the pathethic little 128K 2040 systems that DEC insisted on
>selling as TOPS-20 systems because TOPS-20 had virtual memory and TOPS-10
>would no longer support anything that small.
>
>I can assure you that compiles went just fine on a system with adequate
>memory.  And not, system 2102 was not such a system.

None of ours were.

>
>> >On the -10, the shell was built into the kernel.  On the -20, the shell
>> >was an ordinary user program, as it is on most modern operating 
systems.
>> I never did figure out a good methodology to use this feature.
>> It didn't match my style of work, I think.
>
>Well, perhaps it's time to install Linux on your machine and learn what
>modern designs are all about. 

Someday.  I'm not looking forward to yet another ramp-up effort.

> .. One problem with Windows is that its shell
>encourages you to think in the TOPS-10 way,

With the added feature of making sure my thumbs are duct-taped
to my ass.

> .. even though Windows is
>perfectly capable of doing things the modern way.
>
>> >The -20 had I/O redirection, the -10 did not.
>> What do you mean by this?
>
>In TOPS-10 terms:
>
>I am running program FOO that does INCHAR and and
>OUTSTR.  I don't want to type into the program.  So I do
>	.R FOO < INPUT.TXT
>and now every time FOO does an INCHAR it reads the next character from the
>file.
>
>I am running program FOO that does INCHAR and and
>OUTSTR.  I don't want its output typed on my terminal, I want it to go
>into a file.  So I do
>	.R FOO > OUTPUT.TXT
>and now every time FOO does an OUTSTR it appends to that file.
>
>Of course, I can do:
>	.R FOO < INPUT.TXT > OUTPUT.TXT
>
>This is UNIX syntax, but TOPS-20 had the same thing.  The closest that
>TOPS-10 had was MIC.

Ah, I see.  We always knew that terminal I/O on the -10 left
something to be desired.  

TTYs were not "real" I/O devices in TOPS-10 philosophy.  I don't
know any developer that was satisfied with SCNSER.  But then,
it was a first stab at a complicated problem.

>
>> The -10 was slowly getting acclimated when we finally got a
>> programmer who could change SCNSER without breaking the world.
>> He did implement PUSH/POP but I never found a reason to use it.
>
>PUSH/POP are poor substitutes for real process control,

Yes, I know.  It was a first step to getting the terminal I/O
out of the stone age.

> .. which just about
>every system except for TOPS-10 and DEC's original shell for TOPS-20
>(which was about the first thing that we in the non-DEC TOPS-20 world
>fixed!) had.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: Sat, 16 Feb 02 11:20:06 GMT
Organization: UltraNet Communications, Inc.
Lines: 47
Message-ID: <a4ln4t$6bt$15@bob.news.rcn.net>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net> <a48sd6$eva$18@bob.news.rcn.net> <3C690667.44CB3177@cs.washington.edu>
X-Trace: UmFuZG9tSVa/hEsWFV7YJfZ1d32WqmNnMtN6FmVmNFIEOAdRfvvMmq6G2WFG6Osi
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 16 Feb 2002 13:34:21 GMT
X-Newsreader:  News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-245-42
Xref: chonsp.franklin.ch alt.sys.pdp10:10386

In article <3C690667.44CB3177@cs.washington.edu>,
   Patricia Tressel <tressel@cs.washington.edu> wrote:
>jmfbahciv@aol.com wrote:
>
>> All I had to do was show a new user the ? and the $ command.
>> They figured out how to use the system all on their own.
>
>This is no exaggeration.  

Yea.  You wouldn't believe how much of a work load those two
characters took off my back.

> ..I was a systems programmer at at TOPS-10 shop
>for many years.  Once, we received a tape in DUMPER format.  
>We didn't have anything that would read it under TOPS-10, ...

Oh, grump.  Do you know how much time I spent trying to make
sure that people knew about /INTERCHANGE switches to DUMPER
and BACKUP?  I never did figure out a way to "advertise" 
that the -10 and -20 was not mutually exclusive.


> ... but we did have a working copy of
>TOPS-20...on our KLAD pack.  

=:-}  What?  How did that happen?  I didn't ship any -20 
software on the KLAD...they wouldn't let me.

>So we booted that, and my co-worker remembered the field
>service
>default password.  I'd never used TOPS-20, but I'd heard about $ and ?.
>
>With just the inline help and prompts, I was able to: create an account
>and
>filesystem, mount the tape, read it onto disk with DUMPER, find an
>appropriate
>tape output program (I forget what it was...), mount and label a new
>tape, and
>write the tape contents back out.  (And, yes, it was readable under
>TOPS-10. ;-)
<snip>

How to learn all about TOPS-20 in one swell foop :-).

/BAH

Subtract a hundred and four for e-mail.

######

Newsgroups: alt.sys.pdp10
From: pwilson@world.std.com (Pete Wilson)
Subject: Re: command-line completion
Message-ID: <GrnHHo.MGE@world.std.com>
Date: Sun, 17 Feb 2002 00:03:23 GMT
Organization: The World Public Access UNIX, Brookline, MA
Lines: 11
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newspeer.radix.net!uunet!ash.uu.net!world!pwilson
Xref: chonsp.franklin.ch alt.sys.pdp10:10404

Could I ask you please to trim comp.human-factors when
responding to this thread?

I was the one who started the thread in both groups, but
now the discussion has wandered far, far away from the
original topic and probably has little interest any more
for the human-factors folks.

Thanks so much!

-- Pete Wilson

######

From: "Mike McMahon" <mmcm@attbi.com>
Newsgroups: alt.sys.pdp10,comp.human-factors
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu> <a4ab7q$rjm$1@panix2.panix.com> <a4bf4i$99k$1@watsol.cc.columbia.edu> <3C69C053.889CB9A0@bartek.dontspamme.net> <a4dirg$sac$1@bob.news.rcn.net> <Pine.NXT.4.50.0202131724210.6067-100000@Tomobiki-Cho.CAC.Washington.EDU> <a4iou9$ss5$1@bob.news.rcn.net> <Pine.NXT.4.50.0202150936580.7590-100000@Tomobiki-Cho.CAC.Washington.EDU>
Subject: Re: command-line completion
Lines: 30
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2600.0000
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000
Message-ID: <C4Gb8.5670$TU4.3260727@typhoon.ne.ipsvc.net>
Date: Sun, 17 Feb 2002 04:12:18 GMT
NNTP-Posting-Host: 65.96.243.27
X-Complaints-To: abuse@mediaone.net
X-Trace: typhoon.ne.ipsvc.net 1013919138 65.96.243.27 (Sat, 16 Feb 2002 23:12:18 EST)
NNTP-Posting-Date: Sat, 16 Feb 2002 23:12:18 EST
Organization: ATT Broadband
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.cwix.com!chnws02.mediaone.net!cyclone.ne.ipsvc.net!65.96.0.182!typhoon.ne.ipsvc.net.POSTED!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10409

Sorry for coming in late.  This is relevant to the original thread topic.

As I understand it, Tenex style command-line completion descends from
something on the SDS 940.  This was mentioned a few years ago in this
newsgroup, but not in this latest discussion.

ULTCMD, the original command processor for MM, had a different style of
predictive command completion.  You got the completion as you typed, but
could type characters that had already been completed .  URL completion in
recent versions of IE works a little like this.  Unfortunately, the thing
just beeped if you made a typo.  This interface was based on the command
processor in LLL RISOS.  It was called ULTCMD because when I asked Jed
Donnelley what he called it there, he responded, "the Ultimate Command
Processor".  For a while, MM had a switch where you could choose between
ULTCMD and COMND[%], while Mark and I debated the relative merits of each.

LISP Machines had completion based on what we called "chunks".  Earlier
words of a multi-word entry could be ambiguous, provided later additions
resolved it before you hit the Complete key.  The same thing worked for
completing from the thousands of function names in the system for M-.
("meta-dot"), but with dashes instead of spaces, like in GNU Emacs m-X.  See
DLW's description in
http://groups.google.com/groups?hl=en&selm=DLW.93Jun4225302%40butterball.odi
.com for a little more info.  Symbolics keyboards really was a key labeled
Complete on the keyboard, since the keyboard was designed for the system and
not vice versa.  Similar reasoning called for unshifted parenthesis and
colon keys.  See
http://www.abstractscience.freeserve.co.uk/symbolics/photos/IO/index.html.

######

From: jmfbahciv@aol.com
Newsgroups: alt.sys.pdp10
Subject: Re: command-line completion
Date: Mon, 18 Feb 02 09:24:06 GMT
Organization: UltraNet Communications, Inc.
Lines: 41
Message-ID: <a4qp43$rks$1@bob.news.rcn.net>
References: <a4ab7q$rjm$1@panix2.panix.com>
 <a4bf4i$99k$1@watsol.cc.columbia.edu> <3C69C053.889CB9A0@bartek.dontspamme.net>
 <a4dirg$sac$1@bob.news.rcn.net> <Pine.NXT.4.50.0202131724210.6067-100000@Tomobiki-Cho.CAC.Washington.EDU>
 <a4iou9$ss5$1@bob.news.rcn.net> <Pine.NXT.4.50.0202150936580.7590-100000@Tomobiki-Cho.CAC.Washington.EDU>
 <a4lid8$6bt$3@bob.news.rcn.net> <Pine.NXT.4.50.0202161020540.8469-100000@Tomobiki-Cho.CAC.Washington.EDU>
X-Trace: UmFuZG9tSVYwvliUpq2qTJNt8aG8iTydj2gHqZ6knCJAyO6sH2EG40u/Q8CzHQDi
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 18 Feb 2002 11:38:43 GMT
X-Newsreader:  News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!feedme.news.mediaways.net!news0.de.colt.net!peernews2.colt.net!fr.colt.net!isdnet!nntp.abs.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-216-38
Xref: chonsp.franklin.ch alt.sys.pdp10:10419

[pare a newsgroup]

In article 
<Pine.NXT.4.50.0202161020540.8469-100000@Tomobiki-Cho.CAC.Washington.EDU>,
   Mark Crispin <mrc@CAC.Washington.EDU> wrote:
>On Sat, 16 Feb 2002 jmfbahciv@aol.com wrote:
>> >I can assure you that compiles went just fine on a system with adequate
>> >memory.  And not, system 2102 was not such a system.
>> None of ours were.
>
>And that was one of DEC's problems.  It's a miracle that the TOPS-20 group
>got any work done at all on system 2102.

They just took the stuff over to the TOPS-10 system to do their
builds.  :-)))

>
>There was an article, several years ago, about the divorce between
>Microsoft and IBM in which IBM took custody of the child (OS/2).  Bill
>Gates visited IBM, and was shocked to see the museum pieces that IBM
>made its developers use.  Apparently, some of the OS/2 developers were
>still using 286s.

At least they had the hardware.  There were times when TW insisted that
a piece of new hardware be installed on the system before he wrote
code.  <gasp!>  That meant that the customer couldn't get the first
one off the production line.

JMF had an awful time getting an Alpha when he was supposed to do
the CPU work. 

>
>It explains a lot about the Evil Empire; how they are able to crank out so
>much software, and also why they are somewhat insensitive to bloat issues.

Yup.  It's like air.  You don't think about it until you run out
of it.

/BAH

Subtract a hundred and four for e-mail.

######

From: jmfbahciv@aol.com
Newsgroups: alt.sys.pdp10
Subject: Re: command-line completion
Date: Mon, 18 Feb 02 09:25:41 GMT
Organization: UltraNet Communications, Inc.
Lines: 35
Message-ID: <a4qp72$rks$2@bob.news.rcn.net>
References: <a4bf4i$99k$1@watsol.cc.columbia.edu> <3C69C053.889CB9A0@bartek.dontspamme.net> <a4dirg$sac$1@bob.news.rcn.net> <Pine.NXT.4.50.0202131724210.6067-100000@Tomobiki-Cho.CAC.Washington.EDU> <a4iou9$ss5$1@bob.news.rcn.net> <Pine.NXT.4.50.0202150936580.7590-100000@Tomobiki-Cho.CAC.Washington.EDU> <a4lid8$6bt$3@bob.news.rcn.net> <Pine.NXT.4.50.0202161020540.8469-100000@Tomobiki-Cho.CAC.Washington.EDU> <slrna7153u.7jv.shannon@daydream.shannon.net>
X-Trace: UmFuZG9tSVb6ztjpr3imu/xjTc3Nc83PZkIlC+h+95NodE5GaEgYHr9vZ8ImBEg2
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 18 Feb 2002 11:40:18 GMT
X-Newsreader:  News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-216-38
Xref: chonsp.franklin.ch alt.sys.pdp10:10420

In article <slrna7153u.7jv.shannon@daydream.shannon.net>,
   Charles Shannon Hendrix <shannon@widomaker.com> wrote:
>In article 
<Pine.NXT.4.50.0202161020540.8469-100000@Tomobiki-Cho.CAC.Washington.EDU>, 
Mark Crispin wrote:
>
>> There was an article, several years ago, about the divorce between
>> Microsoft and IBM in which IBM took custody of the child (OS/2).  Bill
>> Gates visited IBM, and was shocked to see the museum pieces that IBM
>> made its developers use.  Apparently, some of the OS/2 developers were
>> still using 286s.
>> 
>> It explains a lot about the Evil Empire; how they are able to crank out 
so
>> much software, and also why they are somewhat insensitive to bloat 
issues.
>
>Just to clarify... do you mean they don't care about bloat or were
>insensitive to those who created bloat?

If you have so much memory that dribbling isn't detected, you don't
fix the leaks because you cannot know that they exist.



>
>I actually think it can be good to make people write code on small
>machines, though that might be better during training and school.
>

No.  Not write code.  Run the code.

/BAH

Subtract a hundred and four for e-mail.

######

From: Charles Shannon Hendrix <shannon@widomaker.com>
Newsgroups: alt.sys.pdp10
Subject: Re: command-line completion
Date: Mon, 18 Feb 2002 00:45:53 -0500
Organization: Too Much
Message-ID: <slrna7153u.7jv.shannon@daydream.shannon.net>
References: <GrDHML.Jso@world.std.com> <a490u201v07@drn.newsguy.com> <a493l9$dsu$1@watsol.cc.columbia.edu> <a4ab7q$rjm$1@panix2.panix.com> <a4bf4i$99k$1@watsol.cc.columbia.edu> <3C69C053.889CB9A0@bartek.dontspamme.net> <a4dirg$sac$1@bob.news.rcn.net> <Pine.NXT.4.50.0202131724210.6067-100000@Tomobiki-Cho.CAC.Washington.EDU> <a4iou9$ss5$1@bob.news.rcn.net> <Pine.NXT.4.50.0202150936580.7590-100000@Tomobiki-Cho.CAC.Washington.EDU> <a4lid8$6bt$3@bob.news.rcn.net> <Pine.NXT.4.50.0202161020540.8469-100000@Tomobiki-Cho.CAC.Washington.EDU>
Reply-To: shannon@nospam.widomaker.com
User-Agent: slrn/0.9.7.3 (Linux)
X-Complaints-To: newsabuse@supernews.com
Lines: 17
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!194.25.134.126.MISMATCH!newsfeed01.sul.t-online.de!t-online.de!newsfeed.media.kyoto-u.ac.jp!out.nntp.be!propagator-SanJose!in.nntp.be!xmission!isdnet!sn-xit-02!sn-post-01!supernews.com!corp.supernews.com!escape.shannon.net!nobody
Xref: chonsp.franklin.ch alt.sys.pdp10:10412

In article <Pine.NXT.4.50.0202161020540.8469-100000@Tomobiki-Cho.CAC.Washington.EDU>, Mark Crispin wrote:

> There was an article, several years ago, about the divorce between
> Microsoft and IBM in which IBM took custody of the child (OS/2).  Bill
> Gates visited IBM, and was shocked to see the museum pieces that IBM
> made its developers use.  Apparently, some of the OS/2 developers were
> still using 286s.
> 
> It explains a lot about the Evil Empire; how they are able to crank out so
> much software, and also why they are somewhat insensitive to bloat issues.

Just to clarify... do you mean they don't care about bloat or were
insensitive to those who created bloat?

I actually think it can be good to make people write code on small
machines, though that might be better during training and school.

######

From: Charles Shannon Hendrix <shannon@widomaker.com>
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: Mon, 18 Feb 2002 01:10:12 -0500
Organization: Too Much
Message-ID: <slrna716hh.7jv.shannon@daydream.shannon.net>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net> <a48sd6$eva$18@bob.news.rcn.net> <3C690667.44CB3177@cs.washington.edu>
Reply-To: shannon@nospam.widomaker.com
User-Agent: slrn/0.9.7.3 (Linux)
X-Complaints-To: newsabuse@supernews.com
Lines: 49
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sn-xit-03!sn-post-01!supernews.com!corp.supernews.com!escape.shannon.net!nobody
Xref: chonsp.franklin.ch alt.sys.pdp10:10413

In article <3C690667.44CB3177@cs.washington.edu>, Patricia Tressel wrote:
> jmfbahciv@aol.com wrote:

> This is no exaggeration.  I was a systems programmer at at TOPS-10
> shop for many years.  Once, we received a tape in DUMPER format.  We
> didn't have anything that would read it under TOPS-10, but we did have
> a working copy of TOPS-20...on our KLAD pack.  So we booted that, and
> my co-worker remembered the field service default password.  I'd never
> used TOPS-20, but I'd heard about $ and ?.

I first hit TOPS-20 from my university account when I found a machine
with an open account.  I had just been told to go there to get some
files I needed and use it until a local system could be made available.

I learned pretty quickly.  Within a couple of days I was building programs
and within a week I had gone further than I had in the previous 3 months
with PrimeOS or the bizarre world of Honeywell DPV.  Sadly, that was
the first and also the last TOPS-20 system I ever used, outside of ftp
service on simtel.mil.  About two weeks later I no longer needed it,
and moved to the local system.

> I was *very* pleased with the $ and ?.  I can't imagine doing the same
> with man pages or Windows help -- those require that the user *know
> something already* 

I learned UNIX by just reading the man pages at first, and supplementing
that with various programming texts.  I never did understand why people
had trouble with it, but then I used to sit in front of the 12-foot wide
UNIX book in the lab at school (all of the BSD and SunOS man pages bound
in a single volume) and read for hours.

I also managed to learn how to use a Prime 9955 with just the online help,
since the idiots that ran it locked all the documentation away.

Having said that, the thing I believe is missing from man pages is
task-oriented help and a catalyst or "getting started" page.  There isn't
enough reference from one program to another.

I managed to learn by reading man pages, but in addition to "man mt;
man dump; etc" I should have been able to do "man backups".

Also, just typing man should get you into the same sort of thing
that ? gave you on TOPS.  I still cannot understand this particular
ommission, or the lack of task oriented documentation.

Every system should have these two elements in any online help,
otherwise I might was well just have a book.

######

From: peter@taronga.com (Peter da Silva)
Newsgroups: alt.sys.pdp10,comp.human-factors
Subject: Re: command-line completion
Date: 18 Feb 2002 11:56:34 GMT
Organization: TSS Inc.
Lines: 24
Message-ID: <a4qq5i$2k6a$1@citadel.in.taronga.com>
References: <GrDHML.Jso@world.std.com> <a48sd6$eva$18@bob.news.rcn.net> <3C690667.44CB3177@cs.washington.edu> <slrna716hh.7jv.shannon@daydream.shannon.net>
NNTP-Posting-Host: citadel.in.taronga.com
X-Trace: citadel.in.taronga.com 1014033394 86218 10.0.0.43 (18 Feb 2002 11:56:34 GMT)
X-Complaints-To: usenet@taronga.com
NNTP-Posting-Date: 18 Feb 2002 11:56:34 GMT
X-Newsreader: trn 4.0-test72 (19 April 1999)
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.stanford.edu!news-proxy.baileynm.com!citadel.in.taronga.com!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10428

In article <slrna716hh.7jv.shannon@daydream.shannon.net>,
Charles Shannon Hendrix  <shannon@nospam.widomaker.com> wrote:
>I managed to learn by reading man pages, but in addition to "man mt;
>man dump; etc" I should have been able to do "man backups".

That's what the KWIC (Key Words In Context) index was for. Unfortunately
it wasn't always online, though at Berkeley you could search the keywords
with the "apropos" command, and all that was described on the handout
that new users were given.

Apropos was later rolled into the man command as the "-k" option.

>Also, just typing man should get you into the same sort of thing
>that ? gave you on TOPS.  I still cannot understand this particular
>ommission, or the lack of task oriented documentation.

There's actually a lot of that in the system, but it's hidden in /usr/doc
or later /usr/share/doc, and meant to be printed out.

-- 
Rev. Peter da Silva, ULC.	                                 WWFD?

"Be conservative in what you generate, and liberal in what you accept"
	-- Matthew 10:16 (l.trans)

######

Newsgroups: alt.sys.pdp10,comp.human-factors
From: Patricia Tressel <tressel@cs.washington.edu>
Subject: Re: command-line completion
Content-Type: text/plain; charset=us-ascii
X-Accept-Language: en
Sender: news@beaver.cs.washington.edu (News Subsystem)
Content-Transfer-Encoding: 7bit
Organization: Computer Science and Engineering, University of Washington
Message-ID: <3C77419C.ABB7B0CF@cs.washington.edu>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net> <a48sd6$eva$18@bob.news.rcn.net> <3C690667.44CB3177@cs.washington.edu> <a4ln4t$6bt$15@bob.news.rcn.net>
X-Mailer: Mozilla 4.79 [en] (X11; U; Linux 2.4.7-UP.26 i686)
Mime-Version: 1.0
X-Nntp-Posting-Host: beryllium.cs.washington.edu
Date: Fri, 22 Feb 2002 23:15:40 -0800
Lines: 46
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.mailgate.org!cyclone.bc.net!nntp.cs.ubc.ca!uw-beaver!news
Xref: chonsp.franklin.ch alt.sys.pdp10:10462

jmfbahciv@aol.com wrote:

> > ..I was a systems programmer at at TOPS-10 shop
> >for many years.  Once, we received a tape in DUMPER format.
> >We didn't have anything that would read it under TOPS-10, ...
> 
> Oh, grump.  Do you know how much time I spent trying to make
> sure that people knew about /INTERCHANGE switches to DUMPER
> and BACKUP?  I never did figure out a way to "advertise"
> that the -10 and -20 was not mutually exclusive.

D'oh!  /INTERCHANGE we knew about, but I guess we thought it was
just to get rid of the protection and account info, i.e. for
exchange with another site running the *same* OS -- we didn't
realize that the format was otherwise mutually readable.

> > ... but we did have a working copy of
> >TOPS-20...on our KLAD pack.
> 
> =:-}  What?  How did that happen?  I didn't ship any -20
> software on the KLAD...they wouldn't let me.

Hmm...  It seemed very Official.  Perhaps Field Service added it?
It was on there because Field Service needed a known-stable OS for
testing.  TOPS-10 sites had a habit of hacking their OSs heavily,
leading to some cases where either 1) something that looked like a
hardware problem was actually a bug introduced by the customer, or
2) commands needed by FS would not behave as expected.

Technically, customers weren't supposed to touch the KLAD pack, but
we had nice FS folks, who weren't terribly interested in coming out
every time we needed a disk pack formatted.  And they appreciated
it that we'd start up the diagnostics before they arrived.  They
also left their fiche set out and unlocked.  I spent much quality
time hunkered down beside skinned RPO7s, with the Pathfinder fiche
and a scope...  Mark C. -- you know Ron Aley, right?  He generally
didn't want to wait for FS to arrive, so they'd show up and find
us swapping processor boards to see if we could make that (e.g.)
SRAM parity error move so we could isolate the bad board.

Ah, nostalgia.

(If we continue with this, should we start a separate thread, and
not hassle the comp.human-factors folks?)

-- Pat

######

From: jmfbahciv@aol.com
Newsgroups: alt.sys.pdp10
Subject: Re: command-line completion
Date: Tue, 26 Feb 02 09:02:02 GMT
Organization: UltraNet Communications, Inc.
Lines: 30
Message-ID: <a5fqt6$l7a$1@bob.news.rcn.net>
References: <a4bf4i$99k$1@watsol.cc.columbia.edu> <Pine.NXT.4.50.0202161020540.8469-100000@Tomobiki-Cho.CAC.Washington.EDU> <slrna7153u.7jv.shannon@daydream.shannon.net> <a4qp72$rks$2@bob.news.rcn.net> <smme5a.deg.ln@escape.shannon.net> <3C7B006F.BCC4E469@jetnet.ab.ca>
X-Trace: UmFuZG9tSVY1gPguLRhMcEwDvMeRjPImD5RtRkMYVnmRzbM4WbE8Vwo6BnvJPctB
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 26 Feb 2002 11:17:58 GMT
X-Newsreader:  News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-102-66
Xref: chonsp.franklin.ch alt.sys.pdp10:10503

In article <3C7B006F.BCC4E469@jetnet.ab.ca>,
   Ben Franchuk <bfranchuk@jetnet.ab.ca> wrote:
>Charles Shannon Hendrix wrote:
>> 
>> Make them write code where they cannot just use a machine with 1GB of
>> RAM and a 1GHz CPU and they will be better programmers, especially if
>> you also require the code to be readable.
>
>Gasp! What ever happened to 256 KB and 2Mhz 6809's running OS/9 version
>2
> and 120 Meg HD's? That had both well written and fast code. 
>
>BTW did the PDP-10 have well written code or did it suffer from BLOAT
>near the end of its life? 

The BLISS edict did not help.  Also the hardware designs promoted
bloat.  TW invented new swears when he wrote the device drivers
for the RA8n.  This device required protocols, not unlike DECnet
(which was an awful bloat producer).  

Having more address space also produced larger programs because
we no longer had to split things up into GETSEGs.  

One of the ways we "saved" core was the notion of multiple
sharable high segments where many jobs could use just one
copy of a piece of code.

/BAH

Subtract a hundred and four for e-mail.

######

From: shannon@news.widomaker.com (Charles Shannon Hendrix)
Newsgroups: alt.sys.pdp10
Subject: Re: command-line completion
Date: Mon, 25 Feb 2002 20:00:12 -0500
Organization: Tidewater Virginia
Message-ID: <smme5a.deg.ln@escape.shannon.net>
References: <a4bf4i$99k$1@watsol.cc.columbia.edu> <Pine.NXT.4.50.0202161020540.8469-100000@Tomobiki-Cho.CAC.Washington.EDU> <slrna7153u.7jv.shannon@daydream.shannon.net> <a4qp72$rks$2@bob.news.rcn.net>
Reply-To: shannon@widomaker.com
X-Newsreader: trn 4.0-test76 (Apr 2, 2001)
Originator: shannon@news.widomaker.com (Charles Shannon Hendrix)
X-Complaints-To: newsabuse@supernews.com
Lines: 26
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sn-xit-03!sn-post-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10508

In article <a4qp72$rks$2@bob.news.rcn.net>,  <jmfbahciv@aol.com> wrote:

> >Just to clarify... do you mean they don't care about bloat or were
> >insensitive to those who created bloat?
> 
> If you have so much memory that dribbling isn't detected, you don't
> fix the leaks because you cannot know that they exist.

Gotcha... though I have 512MB of RAM on my workstation, and I can detect
memory leaks.  There is a huge amount of just not caring in the equation.

> >I actually think it can be good to make people write code on small
> >machines, though that might be better during training and school.
> >
> 
> No.  Not write code.  Run the code.

Well, I think both is necessary.

Make them write code where they cannot just use a machine with 1GB of
RAM and a 1GHz CPU and they will be better programmers, especially if
you also require the code to be readable.

-- 

UNIX/Perl/C/Pizza__________________________________shannon@widomaker.com

######

From: Ben Franchuk <bfranchuk@jetnet.ab.ca>
Newsgroups: alt.sys.pdp10
Subject: Re: command-line completion
Date: Mon, 25 Feb 2002 20:26:39 -0700
Organization: Posted via Supernews, http://www.supernews.com
Message-ID: <3C7B006F.BCC4E469@jetnet.ab.ca>
X-Mailer: Mozilla 4.79 [en] (Win95; U)
X-Accept-Language: en
MIME-Version: 1.0
References: <a4bf4i$99k$1@watsol.cc.columbia.edu> <Pine.NXT.4.50.0202161020540.8469-100000@Tomobiki-Cho.CAC.Washington.EDU> <slrna7153u.7jv.shannon@daydream.shannon.net> <a4qp72$rks$2@bob.news.rcn.net> <smme5a.deg.ln@escape.shannon.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Complaints-To: newsabuse@supernews.com
Lines: 15
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!fr.clara.net!heighliner.fr.clara.net!newsfeed.online.be!sn-xit-01!sn-post-01!supernews.com!corp.supernews.com!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10499

Charles Shannon Hendrix wrote:
> 
> Make them write code where they cannot just use a machine with 1GB of
> RAM and a 1GHz CPU and they will be better programmers, especially if
> you also require the code to be readable.

Gasp! What ever happened to 256 KB and 2Mhz 6809's running OS/9 version
2
 and 120 Meg HD's? That had both well written and fast code. 

BTW did the PDP-10 have well written code or did it suffer from BLOAT
near the end of its life? 
-- 
Ben Franchuk - Dawn * 12/24 bit cpu *
www.jetnet.ab.ca/users/bfranchuk/index.html

######

From: Arthur Krewat <krewat@bartek.dontspamme.net>
Newsgroups: alt.sys.pdp10
Subject: Re: command-line completion
Organization: Kilonet.net
Lines: 10
Message-ID: <3C7D0BA8.C1428B79@bartek.dontspamme.net>
References: <a4bf4i$99k$1@watsol.cc.columbia.edu> <Pine.NXT.4.50.0202161020540.8469-100000@Tomobiki-Cho.CAC.Washington.EDU> <slrna7153u.7jv.shannon@daydream.shannon.net> <a4qp72$rks$2@bob.news.rcn.net> <smme5a.deg.ln@escape.shannon.net> <3C7B006F.BCC4E469@jetnet.ab.ca>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.78 [en] (X11; U; SunOS 5.8 i86pc)
X-Accept-Language: en
Date: Wed, 27 Feb 2002 16:43:12 GMT
NNTP-Posting-Host: 24.186.100.134
X-Trace: news02.optonline.net 1014828192 24.186.100.134 (Wed, 27 Feb 2002 11:43:12 EST)
NNTP-Posting-Date: Wed, 27 Feb 2002 11:43:12 EST
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed.nextra.ch!news.nextra.ch!nextra.at!newsfeed4.cidera.com!newsfeed1.cidera.com!Cidera!cyclone2.usenetserver.com!usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10509

Ben Franchuk wrote:
> 
> BTW did the PDP-10 have well written code or did it suffer from BLOAT
> near the end of its life?

Yes, it was called TOPS-20 :)

No flames please, I'm trying to be funny...

aak

######

From: jmfbahciv@aol.com
Newsgroups: alt.sys.pdp10
Subject: Re: command-line completion
Date: Thu, 28 Feb 02 08:28:56 GMT
Organization: UltraNet Communications, Inc.
Lines: 43
Message-ID: <a5l1nl$hl0$1@bob.news.rcn.net>
References: <a4bf4i$99k$1@watsol.cc.columbia.edu> <Pine.NXT.4.50.0202161020540.8469-100000@Tomobiki-Cho.CAC.Washington.EDU> <slrna7153u.7jv.shannon@daydream.shannon.net> <a4qp72$rks$2@bob.news.rcn.net> <smme5a.deg.ln@escape.shannon.net> <3C7B006F.BCC4E469@jetnet.ab.ca> <76kq7u0glk1vfah2r7m5rbgo9bokaskj39@4ax.com>
X-Trace: UmFuZG9tSVYPGbBXRGjqj41TbX3V/tDRYig+CM/nZHrPoPmTuKwXu9DnwFwwKTHZ
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 28 Feb 2002 10:45:09 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!feed2.news.rcn.net!feed1.news.rcn.net!rcn!208-59-181-106
Xref: chonsp.franklin.ch alt.sys.pdp10:10517

In article <76kq7u0glk1vfah2r7m5rbgo9bokaskj39@4ax.com>,
   Brine <please.spam@somebody.else> wrote:
>Well, tops-10 itself was protected for a very long time by the Tuesday
>morning review meetings,
>
>If you wanted to get a change in, you wrote a MCO, marked up the
>listings (so everyone could look at your code), and went to the
>meeting to defend (or explain) your edits.
>
>If you wrote crap code, you would hear about it.

There weren't many who did write crap code.  I guess the monitor
meeting helped since it could intimidating.  Most of the "bad code"
was handled before the meeting.  By the time the meeting happened
Magee was ready to do the edit.
>
>If you didn't show up, chances were that unless you were CDO, JMF or
>TW, the edits didn't get in.

That's not quite true.  If there was a problem, showing up was
imperative.  Most of the time, the code was already "cleared"
before the meeting.  Prudent people who couldn't stay up (or
get up) for the meeting, usually arranged to have somebody
represent him.  

From what I hear about what other OSes did, the way we did
our biz was more flexible and didn't have dependencies.
>
>IIRC, the meetings started at something like 7:30 or so. 

<grin>  That 7:30 time was one of the biggest controversies.

> .. Only those
>who were interested showed up.  I had stand-alone from midnight to
>3:00 AM, so I was still up (but it was late for me.)
<snip>

Hmmm....somebody who had the 00:00-3:00 slot.  Now that's a big
clue to identity :-).

/BAH

Subtract a hundred and four for e-mail.

######

From: Brine <please.spam@somebody.else>
Newsgroups: alt.sys.pdp10
Subject: Re: command-line completion
Date: Wed, 27 Feb 2002 16:40:10 -0500
Organization: Poor
Message-ID: <76kq7u0glk1vfah2r7m5rbgo9bokaskj39@4ax.com>
Reply-To: 127.0.0.1
References: <a4bf4i$99k$1@watsol.cc.columbia.edu> <Pine.NXT.4.50.0202161020540.8469-100000@Tomobiki-Cho.CAC.Washington.EDU> <slrna7153u.7jv.shannon@daydream.shannon.net> <a4qp72$rks$2@bob.news.rcn.net> <smme5a.deg.ln@escape.shannon.net> <3C7B006F.BCC4E469@jetnet.ab.ca>
X-Newsreader: Forte Agent 1.9/32.560
X-No-Archive: yes
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Complaints-To: newsabuse@supernews.com
Lines: 31
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newscore.univie.ac.at!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sn-xit-03!sn-post-01!supernews.com!corp.supernews.com!Barn.
Xref: chonsp.franklin.ch alt.sys.pdp10:10524

Well, tops-10 itself was protected for a very long time by the Tuesday
morning review meetings,

If you wanted to get a change in, you wrote a MCO, marked up the
listings (so everyone could look at your code), and went to the
meeting to defend (or explain) your edits.

If you wrote crap code, you would hear about it.

If you didn't show up, chances were that unless you were CDO, JMF or
TW, the edits didn't get in.

IIRC, the meetings started at something like 7:30 or so.  Only those
who were interested showed up.  I had stand-alone from midnight to
3:00 AM, so I was still up (but it was late for me.)

The insane mind of Ben Franchuk seems to have said:

>Charles Shannon Hendrix wrote:
>> 
>> Make them write code where they cannot just use a machine with 1GB of
>> RAM and a 1GHz CPU and they will be better programmers, especially if
>> you also require the code to be readable.
>
>Gasp! What ever happened to 256 KB and 2Mhz 6809's running OS/9 version
>2
> and 120 Meg HD's? That had both well written and fast code. 
>
>BTW did the PDP-10 have well written code or did it suffer from BLOAT
>near the end of its life? 

######

From: Arthur Krewat <krewat@bartek.dontspamme.net>
Newsgroups: alt.sys.pdp10
Subject: Re: command-line completion
Organization: Bartek, Inc.
Lines: 20
Message-ID: <3C7C2AD5.B829D32A@bartek.dontspamme.net>
References: <a4bf4i$99k$1@watsol.cc.columbia.edu>
	 <Pine.NXT.4.50.0202161020540.8469-100000@Tomobiki-Cho.CAC.Washington.EDU>
	 <slrna7153u.7jv.shannon@daydream.shannon.net> <a4qp72$rks$2@bob.news.rcn.net>
	 <smme5a.deg.ln@escape.shannon.net> <3C7B006F.BCC4E469@jetnet.ab.ca>
	 <3C7D0BA8.C1428B79@bartek.dontspamme.net> <Pine.NXT.4.50.0202270947180.18518-100000@Tomobiki-Cho.CAC.Washington.EDU>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.78 [en] (Win95; U)
X-Accept-Language: en
Date: Wed, 27 Feb 2002 21:48:13 GMT
NNTP-Posting-Host: 24.186.100.134
X-Trace: news02.optonline.net 1014846493 24.186.100.134 (Wed, 27 Feb 2002 16:48:13 EST)
NNTP-Posting-Date: Wed, 27 Feb 2002 16:48:13 EST
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news.stealth.net!newsfeed.nyc.globix.net!news-out.cwix.com!newsfeed.cwix.com!newsfeed1.cidera.com!Cidera!cyclone2.usenetserver.com!usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail
Xref: chonsp.franklin.ch alt.sys.pdp10:10523



Mark Crispin wrote:

> On Wed, 27 Feb 2002, Arthur Krewat wrote:
> > Ben Franchuk wrote:
> > > BTW did the PDP-10 have well written code or did it suffer from BLOAT
> > > near the end of its life?
> > Yes, it was called TOPS-20 :)
> >
> > No flames please, I'm trying to be funny...
>
> You failed miserably.
>

How did I know you would say that ? :)

aak

######

From: jmfbahciv@aol.com
Newsgroups: alt.sys.pdp10
Subject: Re: command-line completion
Date: Sat, 02 Mar 02 09:24:48 GMT
Organization: UltraNet Communications, Inc.
Lines: 34
Message-ID: <a5qdp3$17a$1@bob.news.rcn.net>
References: <GrDHML.Jso@world.std.com> <3C67EFC7.18E3E3C5@bartek.dontspamme.net> <a48sd6$eva$18@bob.news.rcn.net> <3C690667.44CB3177@cs.washington.edu> <a4ln4t$6bt$15@bob.news.rcn.net> <3C77419C.ABB7B0CF@cs.washington.edu>
X-Trace: UmFuZG9tSVblXqsHWYh4m6oR4uYKxCXa25JEr2hZMGlJaPKfWBQ2xOEbDnkulLMj
X-Complaints-To: abuse@rcn.com
NNTP-Posting-Date: 2 Mar 2002 11:41:23 GMT
X-Newsreader:  News Xpress Version 1.0 Beta #4
Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!208-59-181-8
Xref: chonsp.franklin.ch alt.sys.pdp10:10530

In article <3C77419C.ABB7B0CF@cs.washington.edu>,
   Patricia Tressel <tressel@cs.washington.edu> wrote:
>jmfbahciv@aol.com wrote:
>
>> > ..I was a systems programmer at at TOPS-10 shop
>> >for many years.  Once, we received a tape in DUMPER format.
>> >We didn't have anything that would read it under TOPS-10, ...
>> 
>> Oh, grump.  Do you know how much time I spent trying to make
>> sure that people knew about /INTERCHANGE switches to DUMPER
>> and BACKUP?  I never did figure out a way to "advertise"
>> that the -10 and -20 was not mutually exclusive.
>
>D'oh!  /INTERCHANGE we knew about, but I guess we thought it was
>just to get rid of the protection and account info, i.e. for
>exchange with another site running the *same* OS -- we didn't
>realize that the format was otherwise mutually readable.

That's how software began to get transferred from the -10 to the
-20 and back again.  BACKUP can also read FAILSA format (although
I seem to remember changing the feature test default to be off
after many, many years of no mention of FAILSA).

<snip>

>(If we continue with this, should we start a separate thread, and
>not hassle the comp.human-factors folks?)

I pared the newsgroup down.  It wouldn't hurt them to know about
one program knowing about old formats and why it's necessary.

/BAH

Subtract a hundred and four for e-mail.