From: "Dennis Eldridge" Newsgroups: alt.folklore.computers Subject: Longest running computer Date: Mon, 17 Jul 2000 02:06:25 -0300 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.6700 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6700 Lines: 11 Message-ID: <3972935c$0$8328@wodc7nh6.news.uu.net> NNTP-Posting-Host: 199.172.230.105 X-Trace: reader2.news.uu.net 963810140 8328 199.172.230.105 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!194.25.134.126.MISMATCH!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!newscore.gigabell.net!news-fra.pop.de!ams.uu.net!ffx.uu.net!spool1.news.uu.net!spool0.news.uu.net!reader2.news.uu.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59887 So here's an interesting topic: Old systems still in production use. I'd be very interested to hear stories of systems which are either still in use or recently retired, which have lasted at least 15 years. This was inspired by a system, an MAI Basic 4 8000, donated to the Rhode Island Computer Museum http://www.osfn.org/ricm/ which was bought in 1975 and only put out to pasture in 1997. Anyone? ###### Message-ID: <3972A82B.17177CCC@trailing-edge.com> Date: Mon, 17 Jul 2000 06:31:08 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.0 DEC 3000 Model 300L) MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 19 NNTP-Posting-Host: 63.73.218.130 X-Trace: reader2.news.uu.net 963829869 14722 63.73.218.130 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!nntp.abs.net!uunet!ffx.uu.net!spool1.news.uu.net!spool0.news.uu.net!reader2.news.uu.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59881 Dennis Eldridge wrote: > > So here's an interesting topic: Old systems still in production use. I'd be > very interested to hear stories of systems which are either still in use or > recently retired, which have lasted at least 15 years. > > This was inspired by a system, an MAI Basic 4 8000, donated to the Rhode > Island Computer Museum http://www.osfn.org/ricm/ which was bought in 1975 > and only put out to pasture in 1997. > > Anyone? How about systems that haven't been *rebooted* in 17 years? There's a VAX/VMS system running railroad switching in Ireland that has been running VMS 3.x continuously since it was last rebooted in 1983. Tim. ###### From: "Geoffrey G. Rochat" Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Mon, 17 Jul 2000 10:20:44 -0400 Organization: Kersur Technologies Lines: 40 Message-ID: <8kv4bv$ju4$1@news.kersur.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> NNTP-Posting-Host: 06-091.024.popsite.net X-Trace: news.kersur.net 963843263 20420 216.126.161.91 (17 Jul 2000 14:14:23 GMT) X-Complaints-To: usenet@kersur.net NNTP-Posting-Date: 17 Jul 2000 14:14:23 GMT X-Newsreader: Microsoft Outlook Express 4.72.3110.1 X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!nyc-news-feed1.bbnplanet.com!colt.net!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!portc03.blue.aol.com!uunet!ffx.uu.net!news.kersur.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59870 Dennis Eldridge wrote in message <3972935c$0$8328@wodc7nh6.news.uu.net>... >So here's an interesting topic: Old systems still in production use. I'd be >very interested to hear stories of systems which are either still in use or >recently retired, which have lasted at least 15 years. > >This was inspired by a system, an MAI Basic 4 8000, donated to the Rhode >Island Computer Museum http://www.osfn.org/ricm/ which was bought in 1975 >and only put out to pasture in 1997. > >Anyone? The Rhode Island Computer Museum also has a PDP-9. It was installed in a printing plant in Philadelphia circa 1967, and was taken out of service in 1999 only after some not-very-well-informed soul attempted to do maintenance on the machine by swapping Flip-Chip cards with the power on. I'm in the process of restoring the machine, and hope to apply power some time this summer. I don't know why, but I have an unexplainable affection for old 18-bit DEC machines... The Retro-Computing Society of Rhode Island ('Don't know what it is about Rhode Island, but for a little state they've sure got a lot of people who collect old computers...) has a PB-250 computer from the early 1960s, along with an eclectic assortment of DEC, DG, Wang, Texas Instruments, IBM and everybody-else-too machines, but the oldest thing they have that still runs, IIRC, is a PDP-12. It was powered up and running for about a half hour yesterday. Much of the other old stuff may run too, but it takes a fair amount of courage to power up some of these crufty old monsters. I doubt *very* strongly that any of these things are longevity record holders. You might want to look at DoD, NASA and the FAA for that. ###### From: Will Salt Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 17 Jul 2000 20:00:14 +0100 Organization: Pretty disorganised Lines: 47 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> NNTP-Posting-Host: host62-7-98-84.btinternet.com X-Newsreader: Gnus v5.5/XEmacs 20.4 - "Emerald" Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!nyc-news-feed1.bbnplanet.com!news.gtei.net!colt.net!diablo.netcom.net.uk!netcom.net.uk!peer.news.th.u-net.net!u-net!btnet-peer0!btnet!mendelevium.btinternet.com!candle.btinternet.com!nobody Xref: chonsp.franklin.ch alt.folklore.computers:59987 Tim Shoppa writes: > Dennis Eldridge wrote: > > > > So here's an interesting topic: Old systems still in production use. I'd be > > very interested to hear stories of systems which are either still in use or > > recently retired, which have lasted at least 15 years. > > > > This was inspired by a system, an MAI Basic 4 8000, donated to the Rhode > > Island Computer Museum http://www.osfn.org/ricm/ which was bought in 1975 > > and only put out to pasture in 1997. > > > > Anyone? > > How about systems that haven't been *rebooted* in 17 years? > > There's a VAX/VMS system running railroad switching in Ireland > that has been running VMS 3.x continuously since it was last > rebooted in 1983. Really? Do you have any more details? I doubt that it actually does the switching; more likely, it's just used as an event log. On British Rail since the 1960s, new signalling installations (US: interlockings) generally used a couple of minicomputers (1 running, 1 backup) to keep the Train Register; a record of the times at which trains pass each point on the system. The interlocking itself is done with several rooms full of relay logic; the equipment is completely incapable of making decisions. This is probably how most British trains today are signalled. In the 1980s, British Rail developed solid-state equipment for doing this, based (originally) on Motorola CPUs. These systems are called Integrated Electronic Signalling Centres, and can work automatically as long as the trains run to time :-) There are about 10 such centres currently working; the first opened in 1987. The safety-critical core of there centres uses 3 processors running in parallel and comparing results. The less safety-critical parts (such as the Automatic Route Setting and Train Timetable Processor subsystems) are completely seperate from the interlocking processors themselves. Railway signalling engineers never install anything unless they are *sure* it cannot produce wrong-side failure. -- Will Salt ###### Sender: eric@ruckus.brouhaha.com From: Eric Smith Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> X-Disclaimer: Everything I write is false. Organization: Eric Conspiracy Secret Labs X-Eric-Conspiracy: There is no conspiracy. Date: 17 Jul 2000 16:10:01 -0700 Message-ID: Lines: 8 X-Newsreader: Gnus v5.5/Emacs 20.3 NNTP-Posting-Host: ruckus.brouhaha.com X-Trace: 17 Jul 2000 16:09:40 -0700, ruckus.brouhaha.com Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.direct.ca!look.ca!newsfeed.concentric.net!webtv.net!news.spies.com!ruckus.brouhaha.com Xref: chonsp.franklin.ch alt.folklore.computers:59990 Tim Shoppa writes: > There's a VAX/VMS system running railroad switching in Ireland > that has been running VMS 3.x continuously since it was last > rebooted in 1983. What? Not even for provocative maintenance??? :-) Is it a 782 or other multiprocessor configuration? ###### From: "Carl R. Friend" Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Mon, 17 Jul 2000 21:27:13 -0400 Organization: as little as possible! Lines: 54 Message-ID: <3973B271.763C405@prescienttech.com> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8kv4bv$ju4$1@news.kersur.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: g4WiiL44VV0XPGHewhJlnwsYgI0NIzJhSP7A+EHATIc= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 18 Jul 2000 01:27:18 GMT X-Accept-Language: en X-Mailer: Mozilla 4.5 [en] (X11; I; Linux 2.0.29 i586) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59967 "Geoffrey G. Rochat" wrote: > > The Retro-Computing Society of Rhode Island ('Don't know what it is > about Rhode Island, but for a little state they've sure got a lot of > people who collect old computers...) has a PB-250 computer from the > early 1960s, along with an eclectic assortment of DEC, DG, Wang, Texas > Instruments, IBM and everybody-else-too machines, but the oldest thing > they have that still runs, IIRC, is a PDP-12. I suspect that there's something in the water in RI (USA) which causes this phenomenon. That particular -12, serial nr. 179 was built, probably, in 1970 and ran up until RCS/RI acquired it in 1996. There's an older one, serial nr. 10, which awaits restoration work. Our Packard Bell 250 was built in April, 1961 and was in active service until well into the 1980s. We are working on getting it running on its 40th "birthday" next spring. The LINC-8 in our collection was built in 1965, and was also in active service until the late '80s. It used to be that computers weren't considered "disposable" the way they are now. As such, they were built to last and to be maintainable. So, it's really no surprise that one can put power to a quarter-century old system and have it work well. I wonder how many PCs purchased last year will be running in 2030.... > [RCS/RI's PDP-12] was powered up and running for about a half hour > yesterday. Much of the other old stuff may run too, but it takes a > fair amount of courage to power up some of these crufty old > monsters. RCS/RI is rather hesitant about simply "putting power" to the old iron to see if it runs. There's a rather lengthy process of making sure that the machine is capable of _taking_ power before we even try. > I doubt *very* strongly that any of these things are longevity record > holders. You might want to look at DoD, NASA and the FAA for that. Agreed. For all the scorn that's heaped upon the military for its adherence to "antique" hardware, one needs to remember that the old stuff has been proven to work. Heck, even the "Space Shuttle" used good old ferrite-core memory until a few years ago. -- +------------------------------------------------+---------------------+ | Carl Richard Friend (UNIX Sysadmin) | West Boylston | | Minicomputer Collector / Enthusiast | Massachusetts, USA | | mailto:crfriend@ma.ultranet.com +---------------------+ | http://www.ultranet.com/~crfriend/museum | ICBM: 42:22N 71:47W | +------------------------------------------------+---------------------+ ###### From: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 18 Jul 2000 01:47:41 GMT Organization: The National Capital FreeNet Lines: 18 Message-ID: <8l0cvt$e50$1@freenet9.carleton.ca> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> Reply-To: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) NNTP-Posting-Host: freenet10 X-Trace: freenet9.carleton.ca 963884861 14496 134.117.136.30 (18 Jul 2000 01:47:41 GMT) X-Complaints-To: complaints@ncf.ca NNTP-Posting-Date: 18 Jul 2000 01:47:41 GMT X-Given-Sender: ab528@freenet10.carleton.ca (Heinz W. Wiggeshoff) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.frontiernet.net!nntp.gctr.net!xcski.com!freenet-news!FreeNet.Carleton.CA!ab528 Xref: chonsp.franklin.ch alt.folklore.computers:59976 Will Salt (wpsalt@btinternet.com) writes: > ... > In the 1980s, British Rail developed solid-state equipment for doing > this, based (originally) on Motorola CPUs. These systems are called > Integrated Electronic Signalling Centres, and can work automatically > as long as the trains run to time :-) There are about 10 such centres > currently working; the first opened in 1987. The safety-critical core > of there centres uses 3 processors running in parallel and comparing > results. The less safety-critical parts (such as the Automatic Route > Setting and Train Timetable Processor subsystems) are completely > seperate from the interlocking processors themselves. Railway > signalling engineers never install anything unless they are *sure* it > cannot produce wrong-side failure. So has the cause of that horrid crash last year been determined yet? ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> From: Ric Werme X-Newsreader: NN version 6.5.0 CURRENT #119 Lines: 25 Message-ID: Date: Tue, 18 Jul 2000 03:50:58 GMT NNTP-Posting-Host: 24.91.12.15 X-Complaints-To: abuse@mediaone.net X-Trace: typhoon.ne.mediaone.net 963892258 24.91.12.15 (Mon, 17 Jul 2000 23:50:58 EDT) NNTP-Posting-Date: Mon, 17 Jul 2000 23:50:58 EDT Organization: Road Runner Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.cwix.com!chnws02.mediaone.net!chnws05.ne.mediaone.net!24.128.8.70!typhoon.ne.mediaone.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59996 "Dennis Eldridge" writes: >So here's an interesting topic: Old systems still in production use. I'd be >very interested to hear stories of systems which are either still in use or >recently retired, which have lasted at least 15 years. Didn't we discuss this recently? I mentioned the Bailey Meter 756 (world's first commercially successful multiprocessor system). It was a process control computer used all over the world. One in Australia was installed in 1965 for a power plant that went online in 1965. The system was shut down and sent to the Australian Computer Museum Society some 30 years later. My father designed a lot of it and taught me how to program the data logging processor. Decimal machine, executed off a drum, I need to write a WWW page for it. A Bailey 721 analog computer actually ran the plant. See http://www.terrigal.net.au/~acms/a4.htm to order a 48 page booklet on the 756. -- Ric Werme | werme@nospam.mediaone.net http://people.ne.mediaone.net/werme | ^^^^^^^ delete ###### From: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 18 Jul 2000 04:21:05 GMT Organization: The National Capital FreeNet Lines: 13 Message-ID: <8l0lvh$quk$1@freenet9.carleton.ca> References: <3972935c$0$8328@wodc7nh6.news.uu.net> Reply-To: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) NNTP-Posting-Host: freenet10 X-Trace: freenet9.carleton.ca 963894065 27604 134.117.136.30 (18 Jul 2000 04:21:05 GMT) X-Complaints-To: complaints@ncf.ca NNTP-Posting-Date: 18 Jul 2000 04:21:05 GMT X-Given-Sender: ab528@freenet10.carleton.ca (Heinz W. Wiggeshoff) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!howland.erols.net!nntp.primenet.com.MISMATCH!nntp.frontiernet.net!nntp.gctr.net!xcski.com!freenet-news!FreeNet.Carleton.CA!ab528 Xref: chonsp.franklin.ch alt.folklore.computers:59979 Ric Werme (werme@nospam.mediaone.net) writes: > > I mentioned the Bailey Meter 756 (world's first commercially successful > multiprocessor system). It was a process control computer used all > over the world. One in Australia was installed in 1965 for a power plant > that went online in 1965. The system was shut down and sent to the > Australian Computer Museum Society some 30 years later. Although I've never heard of it, (process control being a secretive black art, if one believes the priests^Wpractioners), why was it shut down? In other words, which MicroSoft Excell or Word product replaced it? ==> B-) <== ###### From: Charles Eicher Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 18 Jul 2000 01:23:52 -0700 Organization: http://extra.newsguy.com Lines: 26 Message-ID: <8l146o$14ca@edrn.newsguy.com> References: <3972935c$0$8328@wodc7nh6.news.uu.net> NNTP-Posting-Host: p-109.newsdawg.com Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!209.50.235.254!europa.netcrusader.net!209.113.65.250!korova.insync.net!solomon.io.com!nntp.giganews.com!nntp2.giganews.com!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!edrn Xref: chonsp.franklin.ch alt.folklore.computers:59954 In article , Ric says... > >I mentioned the Bailey Meter 756 (world's first commercially successful >multiprocessor system). It was a process control computer used all >over the world. One in Australia was installed in 1965 for a power plant >that went online in 1965. The system was shut down and sent to the >Australian Computer Museum Society some 30 years later. From the early computer era, 30 years doesn't seem to be so rare. One of my CompSci professors worked on design of the Univac 1, back in the late 1970s he said he was actively maintaining one of the first Univac 1 models, which was still in daily use running AP and Payroll. At that time, the CPU would easily have been at least 25 years old. Of course, this professor used to proudly describe this Univac 1 to each incoming class of students, because there would always be some smart aleck who declare it would be cheaper to set up a cheap PC and a spreadsheet. And the professor would reply, "Well, what's the fun in THAT?" But the main point he was trying to make, companies with working systems are reluctant to change, as long as the current system works adequately (even if it's just BARELY adequate). Well, that was over 20 years ago, somehow I doubt that this Univac 1 is still operational, especially since the professor retired, and I heard he died several years ago. I would like to believe that the machine outlived the professor, doing accurate computations until it eventually blew a tube and then he was no longer around to repair it. The last of the Univac repairmen, keeping alive the last of the Univacs.... ###### From: Alexandre Pechtchanski Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Organization: Rockefeller University Hospital (GCRC), New York Message-ID: <7go8nssrpbbpcgt8f55v37be89524ee2ia@4ax.com> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8kv4bv$ju4$1@news.kersur.net> <3973B271.763C405@prescienttech.com> X-Newsreader: Forte Agent 1.7/32.534 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 23 Date: Tue, 18 Jul 2000 10:39:07 -0400 NNTP-Posting-Host: 129.85.24.56 X-Trace: rockyd.rockefeller.edu 963931347 129.85.24.56 (Tue, 18 Jul 2000 10:42:27 EDT) NNTP-Posting-Date: Tue, 18 Jul 2000 10:42:27 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.nyu.edu!rockyd.rockefeller.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:59995 On Mon, 17 Jul 2000 21:27:13 -0400, "Carl R. Friend" wrote: [ snip ] > RCS/RI is rather hesitant about simply "putting power" to the old >iron to see if it runs. There's a rather lengthy process of making >sure that the machine is capable of _taking_ power before we even >try. Which reminds me of a scary anecdote from my Soviet days. We just received a CM-4 (Soviet clone of PDP-11), and were going to bring it up. As our previous experience was with the genuine PDP-11, which survived handling by russian load handlers, and in these long past times I was a starry-eyed optimist, I was all for switching it on immediately (after assembly), especially as it carried all the usual complement of quality control certificates. Fortunately, our electronic engineer was less trusting, and decided to check power cables [1]. He was right - they were wired to short 220V to ground. [1] In these times power cables were _hand-assembled_, not extruded. -- [ When replying, remove *'s from address ] Alexandre Pechtchanski, Systems Manager, RUH, NY ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Wed, 19 Jul 00 09:00:37 GMT Organization: UltraNet Communications, Inc. Lines: 27 Message-ID: <8l453j$kdc$2@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8kv4bv$ju4$1@news.kersur.net> <3973B271.763C405@prescienttech.com> <7go8nssrpbbpcgt8f55v37be89524ee2ia@4ax.com> X-Trace: LLOOpQRPg0Y5rUdNGFVkPYK83DitM/03obyV3tSWVRQ= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 19 Jul 2000 11:57:39 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed.berkeley.edu!ucberkeley!newspeer.monmouth.com!europa.netcrusader.net!207.172.3.37!feed1.news.rcn.net!rcn!207-172-245-32 Xref: chonsp.franklin.ch alt.folklore.computers:60029 In article <7go8nssrpbbpcgt8f55v37be89524ee2ia@4ax.com>, Alexandre Pechtchanski wrote: >On Mon, 17 Jul 2000 21:27:13 -0400, "Carl R. Friend" > wrote: >[ snip ] > >> RCS/RI is rather hesitant about simply "putting power" to the old >>iron to see if it runs. There's a rather lengthy process of making >>sure that the machine is capable of _taking_ power before we even >>try. > >Which reminds me of a scary anecdote from my Soviet days. >We just received a CM-4 (Soviet clone of PDP-11), and >were going to bring it up. >As our previous experience was with the genuine PDP-11, >which survived handling by russian load handlers, Heh. Heh. Those PDP-11s seemed to survive anything. It never occured to me before now that lading procedures could be used as a stress test. :-) I guess we'ld (USA) have to use airline baggage handlers here. /BAH Subtract a hundred and four for e-mail. ###### From: "Eric Schweitzer (archy)" Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Tue, 18 Jul 2000 13:45:27 -0400 Lines: 24 Distribution: world Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l146o$14ca@edrn.newsguy.com> NNTP-Posting-Host: schmooze.hunter.cuny.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <8l146o$14ca@edrn.newsguy.com> Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!news.msfc.nasa.gov!info.usuhs.mil!uky.edu!news.cuny.edu!schmooze.hunter.cuny.edu!ershc Xref: chonsp.franklin.ch alt.folklore.computers:60052 On 18 Jul 2000, Charles Eicher wrote: > Date: 18 Jul 2000 01:23:52 -0700 > From: Charles Eicher Well, that was over 20 years ago, somehow I doubt that this Univac 1 is still > operational, especially since the professor retired, and I heard he died several > years ago. I would like to believe that the machine outlived the professor, > doing accurate computations until it eventually blew a tube and then he was no > longer around to repair it. The last of the Univac repairmen, keeping alive the > last of the Univacs.... I don't know about today, but 10 years ago or so I was teaching a class, and one of of my students ran a Univac 1 for Manufacturers Hanover Bank here in NYC. All it did was count proxies (whatever that means to a bank) but it was operational. Of course, he was back in school to get a degree to get another job as it didn't look like the machine would run forever. Certainly ManiHani didn't. ###### From: Charles Eicher Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 18 Jul 2000 10:55:03 -0700 Organization: http://extra.newsguy.com Lines: 26 Message-ID: <8l25ln$8re@edrn.newsguy.com> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8kv4bv$ju4$1@news.kersur.net> <3973B271.763C405@prescienttech.com> <3974935b$0$4933@reader3> NNTP-Posting-Host: p-408.newsdawg.com Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.stanford.edu!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!edrn Xref: chonsp.franklin.ch alt.folklore.computers:60032 In article <3974935b$0$4933@reader3>, > >Carl R. Friend heeft geschreven in bericht ><3973B271.763C405@prescienttech.com>... >> Agreed. For all the scorn that's heaped upon the military for >>its adherence to "antique" hardware, one needs to remember that the >>old stuff has been proven to work. Heck, even the "Space Shuttle" >>used good old ferrite-core memory until a few years ago. > > >I've always wondered, what kind of system they used in the Shuttle. >Vague rumors about triple redundancy come to mind, but does anyone >know exactly what computer(s) were used on the first mission? >And, what are they using now? Hmm.. I recall reading an article when they first started flying, but that was a hell of a long time ago. I recall reading about mag core memories and being surprised it was still in use. The onboard computers are custom IBM designs, I vaguely recall the model being an adaptation of the IBM 370 but I don't recall specifics. There probably wouldn't be a model number corresponding to their main product line, as these were custom units. I do recall reading about their "voting" system, where the three units "voted" on the correct answer to every calculation, and if the vote wasn't unanimous, it was assumed to be in error on all machines, and recalculated. ###### From: glass2@glass2.lexington.ibm.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 18 Jul 2000 18:15:21 GMT Organization: IBM Austin Lines: 40 Message-ID: <8l26rp$1fpk$1@ausnews.austin.ibm.com> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8kv4bv$ju4$1@news.kersur.net> <3973B271.763C405@prescienttech.com> <3974935b$0$4933@reader3> Reply-To: wa4qal@vnet.ibm.com NNTP-Posting-Host: glass2.cv.lexington.ibm.com X-Newsreader: IBM NewsReader/2 2.0 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cs.utexas.edu!geraldo.cc.utexas.edu!tivoli.com!ausnews.austin.ibm.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60064 In <3974935b$0$4933@reader3>, writes: > >Carl R. Friend heeft geschreven in bericht ><3973B271.763C405@prescienttech.com>... >> Agreed. For all the scorn that's heaped upon the military for >>its adherence to "antique" hardware, one needs to remember that the >>old stuff has been proven to work. Heck, even the "Space Shuttle" >>used good old ferrite-core memory until a few years ago. > > >I've always wondered, what kind of system they used in the Shuttle. >Vague rumors about triple redundancy come to mind, but does anyone >know exactly what computer(s) were used on the first mission? >And, what are they using now? > >Frits According to the limited information I have available (and this information is several years old, so it may have changed by now), the shuttle computers are IBM AP-101 machines, formerly manufactured by IBM FSD (Federal Service Division) in Owego, before FSD was sold (to Loral, which was bought by Martin-Marietta, which merged with Lockheed, which...?). The memory, which used to be core memory, was changed to battery backed, error detecting and correcting, CMOS memory. One source indicated that there are six computers on board the shuttle for flight control purposes. I seem to remember that three of them are running the flight control software, with the outputs going through a voter circuit to eliminate any errors. Another is running a different version of the flight control program, and can replace the others in the event of a software error. A fifth one is powered up, and has the flight control program loaded, but not running (e.g., hot standby). The sixth one is unpowered, but can be installed to replace a failing system once the shuttle is on orbit. Of course, all of this is from fuzzy memory, and may not be correct, or the details may have changed. Dave P.S. Standard Disclaimer: I work for them, but I don't speak for them. ###### From: Will Salt Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 18 Jul 2000 23:29:32 +0100 Organization: Pretty disorganised Lines: 27 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> <8l0cvt$e50$1@freenet9.carleton.ca> NNTP-Posting-Host: host62-7-109-219.btinternet.com X-Newsreader: Gnus v5.5/XEmacs 20.4 - "Emerald" Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!newsfeed.online.be!diablo.netcom.net.uk!netcom.net.uk!peer.news.th.u-net.net!u-net!btnet-peer0!btnet!mendelevium.btinternet.com!candle.btinternet.com!nobody Xref: chonsp.franklin.ch alt.folklore.computers:60068 ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) writes: > Will Salt (wpsalt@btinternet.com) writes: > > > .... > > In the 1980s, British Rail developed solid-state equipment for doing > > this, based (originally) on Motorola CPUs. These systems are called > > Integrated Electronic Signalling Centres, and can work automatically > > as long as the trains run to time :-) There are about 10 such centres > > currently working; the first opened in 1987. The safety-critical core > > of there centres uses 3 processors running in parallel and comparing > > results. The less safety-critical parts (such as the Automatic Route > > Setting and Train Timetable Processor subsystems) are completely > > seperate from the interlocking processors themselves. Railway > > signalling engineers never install anything unless they are *sure* it > > cannot produce wrong-side failure. > > > So has the cause of that horrid crash last year been determined yet? Not yet; the official enquiry is still ongoing. Its website is ; incidentally, the evidence presented so far contains, IIRC, quite a lot of info on how the solid-state control systems described above work and are operated. -- Will Salt ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l0lvh$quk$1@freenet9.carleton.ca> From: Ric Werme X-Newsreader: NN version 6.5.0 CURRENT #119 Lines: 25 Message-ID: Date: Tue, 18 Jul 2000 23:40:20 GMT NNTP-Posting-Host: 24.91.12.15 X-Complaints-To: abuse@mediaone.net X-Trace: typhoon.ne.mediaone.net 963963620 24.91.12.15 (Tue, 18 Jul 2000 19:40:20 EDT) NNTP-Posting-Date: Tue, 18 Jul 2000 19:40:20 EDT Organization: Road Runner Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news-koe1.dfn.de!news-was.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.cwix.com!chnws02.mediaone.net!chnws05.ne.mediaone.net!24.128.8.70!typhoon.ne.mediaone.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60072 ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) writes: > Although I've never heard of it, (process control being a secretive > black art, if one believes the priests^Wpractioners), why was it shut > down? In other words, which MicroSoft Excell or Word product replaced it? It's not so much that it's secretive, is more that computer science tended to ignore it. One of may father's hires wrote a well regarded book on the subject that he dedicated to Dad. As for why it was shut down, I seem to recall the power plant was shut down itself, but I may well be wrong. In the latter years I heard that the plant's operator programmed the 756 via a spreadsheet that did the rotational optimization good drum programmers had to do. If the machine were replaced, it could well have been with a PC or special purpose system. My neighbor does embedded system work, and often the code runs on PCs. He uses MS-DOS so he doesn't have to compete with all the clock interrupt and other problems that comes with running Windows. He's had to train a few customers that "Shutdown to DOS" isn't good enough - windows is still running, fielding clock interrupts, and generally getting in the way. -- Ric Werme | werme@nospam.mediaone.net http://people.ne.mediaone.net/werme | ^^^^^^^ delete ###### From: Rui Pedro Mendes Salgueiro Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 19 Jul 2000 13:25:08 +0100 Organization: Universidade de Coimbra Lines: 16 Message-ID: <8l46n4$j6o$1@rena.mat.uc.pt> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8kv4bv$ju4$1@news.kersur.net> <3973B271.763C405@prescienttech.com> <3974935b$0$4933@reader3> NNTP-Posting-Host: localhost X-Trace: rena.mat.uc.pt 964009512 19673 127.0.0.1 (19 Jul 2000 12:25:11 GMT) X-Complaints-To: usenet@rena.mat.uc.pt NNTP-Posting-Date: 19 Jul 2000 12:25:11 GMT User-Agent: tin/pre-1.4-981114 ("The Watchman") (UNIX) (FreeBSD/2.2.7-RELEASE (i386)) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!rccn.net!rena.mat.uc.pt!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60013 Frits wrote: > I've always wondered, what kind of system they used in the Shuttle. I happen to have here a copie of an article about that: "Architecture of the Space Shuttle Primary Avionics Software System" Gene D. Carlow Comunications of the ACM, September 1984, Volume 27, Number 9, pages 926-936. This is a reprint of an article published in 1980-1981. -- http://www.mat.uc.pt/~rps/f1/ an half-tifoso until Canada 2000 Mark Sandman - Morphine, RIP (1952-1999/07/03, Italy) .pt is Portugal| `Whom the gods love die young'-Menander (342-292 BC) Europe | Villeneuve 50-82, Toivonen 56-86, Senna 60-94 ###### Message-ID: <39763807.A76430E4@mail.bcpl.net> From: Ken McMonigal X-Mailer: Mozilla 4.61 [en] (Win98; I) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l0lvh$quk$1@freenet9.carleton.ca> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 40 Date: Wed, 19 Jul 2000 19:21:43 -0400 NNTP-Posting-Host: 204.255.212.65 X-Complaints-To: abuse@bcpl.net X-Trace: news.abs.net 964048776 204.255.212.65 (Wed, 19 Jul 2000 19:19:36 EDT) NNTP-Posting-Date: Wed, 19 Jul 2000 19:19:36 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!unlisys!news.snafu.de!diablo.theplanet.net!europa.netcrusader.net!207.114.4.11!nntp.abs.net!news.abs.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60092 Ric Werme wrote: > ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) writes: > > > Although I've never heard of it, (process control being a secretive > > black art, if one believes the priests^Wpractioners), why was it shut > > down? In other words, which MicroSoft Excell or Word product replaced it? > > It's not so much that it's secretive, is more that computer science tended > to ignore it. One of may father's hires wrote a well regarded book on > the subject that he dedicated to Dad. > > As for why it was shut down, I seem to recall the power plant was shut down > itself, but I may well be wrong. In the latter years I heard that the > plant's operator programmed the 756 via a spreadsheet that did the > rotational optimization good drum programmers had to do. > > If the machine were replaced, it could well have been with a PC or > special purpose system. My neighbor does embedded system work, and > often the code runs on PCs. He uses MS-DOS so he doesn't have to > compete with all the clock interrupt and other problems that comes > with running Windows. He's had to train a few customers that > "Shutdown to DOS" isn't good enough - windows is still running, > fielding clock interrupts, and generally getting in the way. > -- > Ric Werme | werme@nospam.mediaone.net > http://people.ne.mediaone.net/werme | ^^^^^^^ delete In the 70's, alot of process control was accomplished by programmable (logic) controllers that were configured to replace relay racks. Note: I believe Ric alluded to Lake Erie effect snowstorms earlier and has now mentioned that his father worked at Bailey Meter. For those of you at home, that means Cleveland (Euclid), Ohio. Collect your bets now. ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l0lvh$quk$1@freenet9.carleton.ca> <39763807.A76430E4@mail.bcpl.net> From: Ric Werme X-Newsreader: NN version 6.5.0 CURRENT #119 Lines: 29 Message-ID: <%sud5.50540$jE1.316690@typhoon.ne.mediaone.net> Date: Thu, 20 Jul 2000 03:36:27 GMT NNTP-Posting-Host: 24.91.12.15 X-Complaints-To: abuse@mediaone.net X-Trace: typhoon.ne.mediaone.net 964064187 24.91.12.15 (Wed, 19 Jul 2000 23:36:27 EDT) NNTP-Posting-Date: Wed, 19 Jul 2000 23:36:27 EDT Organization: Road Runner Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!newsfeed1.online.no!newsfeed.online.no!news.powertech.no!uninett.no!news.maxwell.syr.edu!newsfeed.cwix.com!chnws02.mediaone.net!chnws05.ne.mediaone.net!24.128.8.70!typhoon.ne.mediaone.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60133 Ken McMonigal writes: >In the 70's, alot of process control was accomplished by >programmable (logic) controllers that were configured to >replace relay racks. I got something in a April 1 CMU campus calendar to the effect of "Panel discussion on replacing the Science (now Wein) Hall elevator controller with a PDP-8 or other minicomputer." I invented a room number that would have been 50 feet beyond the end of the hall above Panther Hollow. I wasn't around that day, but apparently several people came looking for the meeting! >Note: I believe Ric alluded to Lake Erie effect snowstorms earlier >and has now mentioned that his father worked at Bailey Meter. >For those of you at home, that means Cleveland (Euclid), Ohio. >Collect your bets now. I suppose I should wait a day or two for the bets to accumulate, but you're close enough. Dad was at the Wickliffe site which was mostly R&D. I think Euclid was manufacturing. We lived in Painesville, about 25 miles east of Cleveland. (And in the snow belt.) So, what's your connection to that area? -Ric Werme -- Ric Werme | werme@nospam.mediaone.net http://people.ne.mediaone.net/werme | ^^^^^^^ delete ###### Message-ID: <39768D52.7D0A6928@mail.bcpl.net> From: Ken McMonigal X-Mailer: Mozilla 4.61 [en] (Win98; I) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l0lvh$quk$1@freenet9.carleton.ca> <39763807.A76430E4@mail.bcpl.net> <%sud5.50540$jE1.316690@typhoon.ne.mediaone.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 45 Date: Thu, 20 Jul 2000 01:25:38 -0400 NNTP-Posting-Host: 204.255.212.65 X-Complaints-To: abuse@bcpl.net X-Trace: news.abs.net 964070610 204.255.212.65 (Thu, 20 Jul 2000 01:23:30 EDT) NNTP-Posting-Date: Thu, 20 Jul 2000 01:23:30 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.datacomm.ch!newsmaster-01.atnet.at!atnet.at!newsrouter.chello.at!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.abs.net!news.abs.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60095 Ric Werme wrote: > Ken McMonigal writes: > > >In the 70's, alot of process control was accomplished by > >programmable (logic) controllers that were configured to > >replace relay racks. > > I got something in a April 1 CMU campus calendar to the effect of > "Panel discussion on replacing the Science (now Wein) Hall elevator > controller with a PDP-8 or other minicomputer." I invented a room > number that would have been 50 feet beyond the end of the hall above > Panther Hollow. I wasn't around that day, but apparently several > people came looking for the meeting! > > >Note: I believe Ric alluded to Lake Erie effect snowstorms earlier > >and has now mentioned that his father worked at Bailey Meter. > >For those of you at home, that means Cleveland (Euclid), Ohio. > >Collect your bets now. > > I suppose I should wait a day or two for the bets to accumulate, but > you're close enough. Dad was at the Wickliffe site which was mostly > R&D. I think Euclid was manufacturing. We lived in Painesville, > about 25 miles east of Cleveland. (And in the snow belt.) > > So, what's your connection to that area? > > -Ric Werme > -- > Ric Werme | werme@nospam.mediaone.net > http://people.ne.mediaone.net/werme | ^^^^^^^ delete 1) Not really a hack, but, in the 70's while logged on at Case-10, did a Who on CMA-10 and found a user with no password, logged remotely on as that user and listed files. 2) While I worked at Ailin' Badly (uh ... Allen Bradley), people moved back and forth between there and Bailey. ###### From: djb0x1312fca@scream.org (Dan) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 20 Jul 2000 09:04:50 GMT Lines: 33 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> Reply-To: djb0x1312fca@scream.org NNTP-Posting-Host: v2000.scream.org X-Trace: mochi.lava.net 964083890 903 198.151.213.3 (20 Jul 2000 09:04:50 GMT) X-Complaints-To: abuse@lava.net NNTP-Posting-Date: 20 Jul 2000 09:04:50 GMT User-Agent: slrn/0.9.6.2 (Linux) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!newsfeed.direct.ca!look.ca!newsfeed1.earthlink.net!newsfeed2.earthlink.net!newsfeed.earthlink.net!lsanca1-snf1!news.gtei.net!news.lava.net!djb0x1312fca Xref: chonsp.franklin.ch alt.folklore.computers:60131 Tim Shoppa wrote: > How about systems that haven't been *rebooted* in 17 years? > > There's a VAX/VMS system running railroad switching in Ireland > that has been running VMS 3.x continuously since it was last > rebooted in 1983. I have to be a little concerned about this one, since old versions of VAX/VMS had an "end-of-the-epoch" problem (somewhat similar to UNIX's year-2038 limitation) on or about May 19, 1997. 10,000 days from the start of the epoch, which was 01/01/1970, if I recall. I believe the problem cropped up when one tried to measure the difference in time between two dates, and one or both of them were outside the epoch. Of course, this only affected systems that dealt with dates, so if the railroad folks are doing an application that's unconcerned with dates (which railroad switching may very well be), it might simply not be an issue for them. The vast majority of old Vaxen of which I'm personally aware were rebooted right around May 19, 1997, as various second-tier schools scurried to upgrade before The End Of The World. (The first-tier ones, of course, had abandoned VAX/VMS by then... ;) -Dan -- Dan Birchall - Palolo Valley, Honolulu HI - http://dan.scream.org Get paid for your opinions - http://epinions.scream.org/join.html My address expires - take out the hex stamp if your reply bounces ###### Message-ID: <3976D79B.D2943F34@netinsight.se> From: Johnny Billquist Organization: Netinsight AB X-Mailer: Mozilla 4.5 [en] (X11; I; Linux 2.2.14 i686) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Lines: 60 Date: Thu, 20 Jul 2000 10:42:36 GMT NNTP-Posting-Host: 194.16.221.33 X-Complaints-To: abuse@telia.com X-Trace: newsc.telia.net 964089756 194.16.221.33 (Thu, 20 Jul 2000 12:42:36 MET DST) NNTP-Posting-Date: Thu, 20 Jul 2000 12:42:36 MET DST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!newsfeedZ.netscum.dQ!netscum.int!uninett.no!newsfeed1.enitel.no!masternews.telia.net!newsc.telia.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60102 Dan wrote: > > Tim Shoppa wrote: > > > How about systems that haven't been *rebooted* in 17 years? > > > > There's a VAX/VMS system running railroad switching in Ireland > > that has been running VMS 3.x continuously since it was last > > rebooted in 1983. > > I have to be a little concerned about this one, since old versions > of VAX/VMS had an "end-of-the-epoch" problem (somewhat similar to > UNIX's year-2038 limitation) on or about May 19, 1997. 10,000 > days from the start of the epoch, which was 01/01/1970, if I recall. This isn't really a VMS problem as such. It's a problem with C-programs under VMS, which uses the Unix time calls. Those measure time as a delta time from the Unix epoch, which reached 10,000 days in 1997. VMS itself counts ticks since ??-NOV-1876(?) (or whatever it is). The 10,000 days limit on delta times was well documented, it just didn't occur to people that this meant impeding doom for Unix programs when compiled on VMS. (Once again, Unix rears its ugly head... :-) > I believe the problem cropped up when one tried to measure the > difference in time between two dates, and one or both of them > were outside the epoch. Actually, the problem is/was delta times larger than 10,000 days, and since Unix measures things since 1970, the VMS C RTL choose to use delta times since 1970, which in itself was obviously a pretty simple solution. > Of course, this only affected systems that dealt with dates, so > if the railroad folks are doing an application that's unconcerned > with dates (which railroad switching may very well be), it might > simply not be an issue for them. Nah. VMS itself don't have any such problems. VMS dates *will* run out sometime, but you can rest assured that we're far from there yet. Any other language on VMS continues to work just fine, it was only C that was affected. > The vast majority of old Vaxen of which I'm personally aware were > rebooted right around May 19, 1997, as various second-tier schools > scurried to upgrade before The End Of The World. (The first-tier > ones, of course, had abandoned VAX/VMS by then... ;) Probably some of those reboots were because of uninformed system managers, and so on. :-) Johnny -- Johnny Billquist | johnny.billquist@netinsight.net Net Insight AB | phone: +46 8 685 04 88 Västberga Allé 9 | fax: +46 8 685 04 20 Box 42093 | SE-126 30 STOCKHOLM, Sweden | http://www.netinsight.net ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> <3976D79B.D2943F34@netinsight.se> Organization: Daedalus Consulting X-Newsreader: trn 4.0-test72 (19 April 1999) From: don@news.daedalus.co.nz (Don Stokes) Message-ID: <964099762.86210@shelley.paradise.net.nz> Cache-Post-Path: shelley.paradise.net.nz!unknown@203-96-144-16.cable.paradise.net.nz X-Cache: nntpcache 2.4.0b5 (see http://www.nntpcache.org/) Lines: 20 Date: Thu, 20 Jul 2000 13:29:41 GMT NNTP-Posting-Host: 203.96.152.26 X-Complaints-To: newsadmin@xtra.co.nz X-Trace: news.xtra.co.nz 964099781 203.96.152.26 (Fri, 21 Jul 2000 01:29:41 NZST) NNTP-Posting-Date: Fri, 21 Jul 2000 01:29:41 NZST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!feeder.via.net!enews.sgi.com!news.xtra.co.nz!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60122 Johnny Billquist wrote: >Nah. VMS itself don't have any such problems. VMS dates *will* run >out sometime, but you can rest assured that we're far from there yet. >Any other language on VMS continues to work just fine, it was only >C that was affected. First sign of VMS dates running out is when the year stops fitting into 4 digits. I believe the next failure is in the year 32768. >Probably some of those reboots were because of uninformed system >managers, and so on. :-) Indeed. VAXCRTL can be replaced without rebooting the system, if you hold your mouth right. Anyway, VAXCRTL wasn't a part of the VMS distribution in 3.x... (C was latecomer to the VAX/VMS language stable, and never really fitted in very well.) -- don ###### Message-ID: <397711EB.CC1C5723@netinsight.se> From: Johnny Billquist Organization: Netinsight AB X-Mailer: Mozilla 4.5 [en] (X11; I; Linux 2.2.14 i686) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> <3976D79B.D2943F34@netinsight.se> <964099762.86210@shelley.paradise.net.nz> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Lines: 49 Date: Thu, 20 Jul 2000 14:51:44 GMT NNTP-Posting-Host: 194.16.221.33 X-Complaints-To: abuse@telia.com X-Trace: newsc.telia.net 964104704 194.16.221.33 (Thu, 20 Jul 2000 16:51:44 MET DST) NNTP-Posting-Date: Thu, 20 Jul 2000 16:51:44 MET DST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!newsxfer.eecs.umich.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!uninett.no!newsfeed1.enitel.no!masternews.telia.net!newsc.telia.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60100 Don Stokes wrote: > > Johnny Billquist wrote: > >Nah. VMS itself don't have any such problems. VMS dates *will* run > >out sometime, but you can rest assured that we're far from there yet. > >Any other language on VMS continues to work just fine, it was only > >C that was affected. > > First sign of VMS dates running out is when the year stops fitting into > 4 digits. I believe the next failure is in the year 32768. Hmmm, no, it's most likely not 32768. 32768+1900 is the first real hard problem for RSX though. VMS counts in 100ns(?) increments in a 64 bit counter. If I rember that correctly, it gives VMS a range of 29227 years, this added to 1876 would give VMS breaking sometime in the year 31103, or possibly 31104. But you're right, when the years start using 5 digits, much software will break, so that is probably when we'd need the first upgrade. Still beats Unix, which only have 36 years left. :-) > >Probably some of those reboots were because of uninformed system > >managers, and so on. :-) > > Indeed. VAXCRTL can be replaced without rebooting the system, if you > hold your mouth right. > > Anyway, VAXCRTL wasn't a part of the VMS distribution in 3.x... > (C was latecomer to the VAX/VMS language stable, and never really > fitted in very well.) You tell me. NUL-terminated strings. Pointers to buffers instead of buffer descriptors (ever heard of security issues because of buffer overflows?), every assuming that you have the Unix system calls, and so on... Oh well... Johnny -- Johnny Billquist | johnny.billquist@netinsight.net Net Insight AB | phone: +46 8 685 04 88 Västberga Allé 9 | fax: +46 8 685 04 20 Box 42093 | SE-126 30 STOCKHOLM, Sweden | http://www.netinsight.net ###### Path: chonsp.franklin.ch!not-for-mail From: Neil Franklin Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 20 Jul 2000 21:05:55 +0200 Organization: My own Private Self Lines: 50 Message-ID: <6ur98ohbak.fsf@chonsp.franklin.ch> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> <3976D79B.D2943F34@netsight.se> NNTP-Posting-Host: chonsp.franklin.ch X-Trace: chonsp.franklin.ch 964119955 455 10.0.3.2 (20 Jul 2000 19:05:55 GMT) X-Complaints-To: news@chonsp.franklin.ch NNTP-Posting-Date: 20 Jul 2000 19:05:55 GMT X-Newsreader: Gnus v5.7/Emacs 20.4 Xref: chonsp.franklin.ch alt.folklore.computers:60152 Johnny Billquist writes: > Dan wrote: > > > > I have to be a little concerned about this one, since old versions > > of VAX/VMS had an "end-of-the-epoch" problem (somewhat similar to > > UNIX's year-2038 limitation) on or about May 19, 1997. 10,000 > > days from the start of the epoch, which was 01/01/1970, if I recall. > > The 10,000 days limit on delta times was well documented, it just didn't > occur to people that this meant impeding doom for Unix programs when Perhaps because no one writing an program on Unix would expect such an limit (which Unix does not have)? And Unix programmers (even those few who have heard of VMS) tend to not read VMS docs when writing their stuff (for Unix). Now people porting stuff from Unix to VMS... time_t is 32bit and defined as seconds... > compiled on VMS. (Once again, Unix rears its ugly head... :-) Unix (and C) count in _seconds_ since 1970, which is why they go wrong on 2^31 seconds = in 2038. There is nowhere in Unix (or C) anythong about 10'000 days, that is just an VAXCRTL nonsense. Anyway: a) What clutz calculates in days? b) What clutz puts in 10^x boundries? The blame here lies fully on the dimwits who made an C RTL which simply does not behave the way C is expected to. > > Of course, this only affected systems that dealt with dates, so > > if the railroad folks are doing an application that's unconcerned > > with dates (which railroad switching may very well be), it might > > simply not be an issue for them. > > Nah. VMS itself don't have any such problems. VMS dates *will* run > out sometime, but you can rest assured that we're far from there yet. Room for 2^64 ticks of 100ns each. -- Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ Nerd, Geek, Hacker, Unix Guru, Sysadmin, Roleplayer, LARPer, Mystic ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3976D79B.D2943F34@netsight.se> <6ur98ohbak.fsf@chonsp.franklin.ch> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <3977751c$1@news.ucsc.edu> Date: 20 Jul 2000 14:54:36 -0800 X-Trace: 20 Jul 2000 14:54:36 -0800, sundance.cse.ucsc.edu Lines: 10 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.datacomm.ch!newsmaster-01.atnet.at!atnet.at!newsrouter.chello.at!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.primenet.com!nntp.gblx.net!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:60252 In article <6ur98ohbak.fsf@chonsp.franklin.ch>, Neil Franklin wrote: >Anyway: >a) What clutz calculates in days? The Earth. Every one else merely wants to keep up. I guess we had better not tell you about the Millenium Clock, then. ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3976D79B.D2943F34@netinsight.se> <964099762.86210@shelley.paradise.net.nz> <397711EB.CC1C5723@netinsight.se> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <397776d6$1@news.ucsc.edu> Date: 20 Jul 2000 15:01:58 -0800 X-Trace: 20 Jul 2000 15:01:58 -0800, sundance.cse.ucsc.edu Lines: 19 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.datacomm.ch!newsmaster-01.atnet.at!atnet.at!newsrouter.chello.at!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-hog.berkeley.edu!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:60257 In article <397711EB.CC1C5723@netinsight.se>, Johnny Billquist wrote: >VMS counts in 100ns(?) increments in a 64 bit counter. If I remember >that correctly, it gives VMS a range of 29227 years, this added to >1876 would give VMS breaking sometime in the year 31103, or possibly 31104. Oh, this is interesting. Compaq changed this? My foggy memory had a 60 Hz to 50/100 Hz (max) clock which lots of people confused. Which SYS$call? I used to write little clock synchronization loops to see how much work I could do during 1 clock tick waiting to sync. I remembering doing as many as 33 complex instructions in a clock tick, but that was long ago. I really need a cycle time clock like on a Cray (which has many other advantages but not as many as the formerly planned SS-1). 100 ns on VMS would in my opinion raise the benchmarking quality of the machine from a C to a B which is where Convex was in 1984. ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3973B271.763C405@prescienttech.com> <3974935b$0$4933@reader3> <8l26rp$1fpk$1@ausnews.austin.ibm.com> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <39777772$1@news.ucsc.edu> Date: 20 Jul 2000 15:04:34 -0800 X-Trace: 20 Jul 2000 15:04:34 -0800, sundance.cse.ucsc.edu Lines: 6 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!news-hog.berkeley.edu!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:60250 In article <8l26rp$1fpk$1@ausnews.austin.ibm.com>, wrote: >the shuttle computers are IBM AP-101 machines, formerly manufactured >by IBM FSD (Federal Service Division) in Owego, before FSD was sold Federal Systems Divison Oswego ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8kv4bv$ju4$1@news.kersur.net> <3973B271.763C405@prescienttech.com> <3974935b$0$4933@reader3> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <39777916$1@news.ucsc.edu> Date: 20 Jul 2000 15:11:34 -0800 X-Trace: 20 Jul 2000 15:11:34 -0800, sundance.cse.ucsc.edu Lines: 31 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.stanford.edu!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:60249 >Carl R. Friend heeft geschreven in bericht ><3973B271.763C405@prescienttech.com>... >> Agreed. For all the scorn that's heaped upon the military for >>its adherence to "antique" hardware, one needs to remember that the >>old stuff has been proven to work. Heck, even the "Space Shuttle" >>used good old ferrite-core memory until a few years ago. We had a General at JPL who took the "if it ain't broke don't fix it" routine. People like that are best by passed. The context is clearly important. Zero G and rad-hard are not the concerns of most programmers. It also makes for silly edicts like those made by the JSC CIO who was quoted in Datamation about getting rid of Macs and standardizing on Win95 as if it were going to be bug free. In article <3974935b$0$4933@reader3>, wrote: >I've always wondered, what kind of system they used in the Shuttle. >Vague rumors about triple redundancy come to mind, but does anyone >know exactly what computer(s) were used on the first mission? >And, what are they using now? IBM AP-101Bs. It's an old sci/space FAQ question. There's a cited CACM article and other articles about it. If you find the old sci.space FAQ (Leech is at SGI now). Astronauts are issued personalized Think-Pads for email. There's a human in the loop to minimize the waste of the limited uplink bandwidth. I keep a few of the down link bits which I have received. Surely there's a #5 ESS sitting somewhere running longer than other machines. Maybe a Tandem some place. ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <964099762.86210@shelley.paradise.net.nz> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> Organization: Daedalus Consulting X-Newsreader: trn 4.0-test72 (19 April 1999) From: don@news.daedalus.co.nz (Don Stokes) Message-ID: <964148894.748064@shelley.paradise.net.nz> Cache-Post-Path: shelley.paradise.net.nz!unknown@203-96-144-16.cable.paradise.net.nz X-Cache: nntpcache 2.4.0b5 (see http://www.nntpcache.org/) Lines: 21 Date: Fri, 21 Jul 2000 03:08:33 GMT NNTP-Posting-Host: 203.96.152.26 X-Complaints-To: newsadmin@xtra.co.nz X-Trace: news.xtra.co.nz 964148913 203.96.152.26 (Fri, 21 Jul 2000 15:08:33 NZST) NNTP-Posting-Date: Fri, 21 Jul 2000 15:08:33 NZST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!fu-berlin.de!news.maxwell.syr.edu!feeder.via.net!enews.sgi.com!news.xtra.co.nz!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60235 In article <397776d6$1@news.ucsc.edu>, Eugene Miya wrote: >In article <397711EB.CC1C5723@netinsight.se>, >Johnny Billquist wrote: >>VMS counts in 100ns(?) increments in a 64 bit counter. If I remember >>that correctly, it gives VMS a range of 29227 years, this added to >>1876 would give VMS breaking sometime in the year 31103, or possibly 31104. > >Oh, this is interesting. Compaq changed this? >My foggy memory had a 60 Hz to 50/100 Hz (max) clock which lots of >people confused. Which SYS$call? 100 ns is the resolution of the time format. The actual resolution of the VAX/VMS clock is (usually) 10 ms, ie that's how often the clock interrupt goes off. IIRC, Some VAXes had higher resolution interval timers that could (in theory at least) update more frequently. The clock increments by 100 ticks every timer interrupt. I'm not sure whether this was improved in Alpha VMS. -- don ###### From: bmeyer@cs.monash.edu.au Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Organization: This is innd taking over... Lines: 16 Sender: bmeyer@bruce.cs.monash.edu.au Message-ID: <8l8kaa$3i5$1@wombat.cs.monash.edu.au> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <964099762.86210@shelley.paradise.net.nz> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <964148894.748064@shelley.paradise.net.nz> Reply-To: bmeyer@cs.monash.edu.au Date: 21 Jul 2000 14:41:46 +1000 NNTP-Posting-Host: 130.194.64.217 X-Trace: news0.optus.net.au 964184198 130.194.64.217 (Fri, 21 Jul 2000 22:56:38 EST) NNTP-Posting-Date: Fri, 21 Jul 2000 22:56:38 EST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!news.rhrz.uni-bonn.de!RRZ.Uni-Koeln.DE!news.netcologne.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!intgwpad.nntp.telstra.net!news1.optus.net.au!optus!news0.optus.net.au!wombat.cs.monash.edu.au!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60230 don@news.daedalus.co.nz (Don Stokes) writes: >100 ns is the resolution of the time format. The actual resolution of >the VAX/VMS clock is (usually) 10 ms, ie that's how often the clock >interrupt goes off. [...] >The clock increments by 100 ticks every timer interrupt. That would make the clock be somewhat slow ;-) Bernie -- Thomas --- Jefferson --- still surv-- John Adams 2nd President of the US Last words, 4 July 1826 ###### From: Alistair@caribsurf.com (Alistair Gale) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 21 Jul 2000 09:03:10 GMT Organization: http://extra.newsguy.com Lines: 8 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8kv4bv$ju4$1@news.kersur.net> <3973B271.763C405@prescienttech.com> <3974935b$0$4933@reader3> <39777916$1@news.ucsc.edu> Reply-To: alistair@caribsurf.com NNTP-Posting-Host: p-919.newsdawg.com User-Agent: slrn/0.9.5.7 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!informatik.tu-muenchen.de!news.csl-gmbh.net!blackbush.xlink.net!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.stanford.edu!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!Alistair Xref: chonsp.franklin.ch alt.folklore.computers:60188 On 20 Jul 2000 15:11:34 -0800, Eugene Miya wrote: >Surely there's a #5 ESS sitting somewhere running longer than other machines. >Maybe a Tandem some place. Don't you mean a #1ESS? -- alistair ###### Message-ID: <3978151E.493527D5@netinsight.se> From: Johnny Billquist Organization: Netinsight AB X-Mailer: Mozilla 4.5 [en] (X11; I; Linux 2.2.14 i686) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> <3976D79B.D2943F34@netsight.se> <6ur98ohbak.fsf@chonsp.franklin.ch> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Lines: 61 Date: Fri, 21 Jul 2000 09:17:18 GMT NNTP-Posting-Host: 194.16.221.33 X-Complaints-To: abuse@telia.com X-Trace: newsc.telia.net 964171038 194.16.221.33 (Fri, 21 Jul 2000 11:17:18 MET DST) NNTP-Posting-Date: Fri, 21 Jul 2000 11:17:18 MET DST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!news.rhrz.uni-bonn.de!RRZ.Uni-Koeln.DE!news.netcologne.de!skynet.be!newsfeedZ.netscum.dQ!netscum.int!uninett.no!newsfeed1.enitel.no!masternews.telia.net!newsc.telia.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60186 Neil Franklin wrote: > > Johnny Billquist writes: > > > The 10,000 days limit on delta times was well documented, it just didn't > > occur to people that this meant impeding doom for Unix programs when > > Perhaps because no one writing an program on Unix would expect such an > limit (which Unix does not have)? And Unix programmers (even those few > who have heard of VMS) tend to not read VMS docs when writing their > stuff (for Unix). Of course not. My complaint here was not with the Unix programmers, but with the VAXCRTL implementors. (But Unix programmers don't particularly expect the 2038 limit either...) > > compiled on VMS. (Once again, Unix rears its ugly head... :-) > > Unix (and C) count in _seconds_ since 1970, which is why they go wrong > on 2^31 seconds = in 2038. There is nowhere in Unix (or C) anythong > about 10'000 days, that is just an VAXCRTL nonsense. Yes. But it is rather funny that we have so short limits in *both* cases, while not any such problems with other language/os combinations. (I only count rather currently maintaned OSes, and Microsoft don't count at all... :-) > Anyway: > > a) What clutz calculates in days? > b) What clutz puts in 10^x boundries? > > The blame here lies fully on the dimwits who made an C RTL which > simply does not behave the way C is expected to. Yes. The blame lies heavily on the people implementing the C RTL. But perhaps someone can elaborate on why VMS delta times had a limit of 10,000 days? I really don't know. Internally, delta times are represented as negative time values, which are just 100ns increments, so the 10,000 days limit seem rather arbitrary. Probably some ASCII conversion routine didn't want to have more than four digits for the days, but apart from that? > > Nah. VMS itself don't have any such problems. VMS dates *will* run > > out sometime, but you can rest assured that we're far from there yet. > > Room for 2^64 ticks of 100ns each. Almost. 2^63 actually, since the value is signed, and negative values are delta times (unless my memory fails me). Johnny -- Johnny Billquist | johnny.billquist@netinsight.net Net Insight AB | phone: +46 8 685 04 88 Västberga Allé 9 | fax: +46 8 685 04 20 Box 42093 | SE-126 30 STOCKHOLM, Sweden | http://www.netinsight.net ###### From: cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 21 Jul 2000 09:51:17 GMT Organization: teabag Message-ID: <8l96el$kad$2@teabag.demon.co.uk> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> <3976D79B.D2943F34@netsight.se> <6ur98ohbak.fsf@chonsp.franklin.ch> <3978151E.493527D5@netinsight.se> NNTP-Posting-Host: teabag.demon.co.uk X-NNTP-Posting-Host: teabag.demon.co.uk:193.237.4.110 X-Trace: news.demon.co.uk 964174040 nnrp-10:10787 NO-IDENT teabag.demon.co.uk:193.237.4.110 X-Complaints-To: abuse@demon.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 21 Jul 2000 09:51:17 GMT X-Newsreader: knews 1.0b.0 Lines: 16 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!fu-berlin.de!news-fra1.dfn.de!news0.de.colt.net!colt.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!teabag.demon.co.uk!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60214 In article <3978151E.493527D5@netinsight.se>, Johnny Billquist writes: > (But Unix programmers don't particularly expect the 2038 limit either...) I wouldn't be so sure, most I've known are all too well aware of it; a possible problem is that some expect it to transparently disappear upon introduction of a longer time_t, which of course doesn't take into account either less experienced programmers continuing to code it as a 32-bit signed int or people diddling with existing code and breaking it. Personally, I hope to have retired by then (but knowing my luck and/ or stupidity I'll still be going) and just intend to sit back and smugly have a good laugh to myself... :) Chris. ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397776d6$1@news.ucsc.edu> <964148894.748064@shelley.paradise.net.nz> <8l8kaa$3i5$1@wombat.cs.monash.edu.au> Organization: Daedalus Consulting X-Newsreader: trn 4.0-test72 (19 April 1999) From: don@news.daedalus.co.nz (Don Stokes) Message-ID: <964186064.134834@shelley.paradise.net.nz> Cache-Post-Path: shelley.paradise.net.nz!unknown@203-96-144-16.cable.paradise.net.nz X-Cache: nntpcache 2.4.0b5 (see http://www.nntpcache.org/) Lines: 15 Date: Fri, 21 Jul 2000 13:28:03 GMT NNTP-Posting-Host: 203.96.152.26 X-Complaints-To: newsadmin@xtra.co.nz X-Trace: news.xtra.co.nz 964186083 203.96.152.26 (Sat, 22 Jul 2000 01:28:03 NZST) NNTP-Posting-Date: Sat, 22 Jul 2000 01:28:03 NZST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!informatik.tu-muenchen.de!news.csl-gmbh.net!blackbush.xlink.net!newsfeed.germany.net!newscore.gigabell.net!newsfeeds.belnet.be!news.belnet.be!newsfeedZ.netscum.dQ!netscum.int!enews.sgi.com!news.xtra.co.nz!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60232 In article <8l8kaa$3i5$1@wombat.cs.monash.edu.au>, wrote: >don@news.daedalus.co.nz (Don Stokes) writes: > >>100 ns is the resolution of the time format. The actual resolution of >>the VAX/VMS clock is (usually) 10 ms, ie that's how often the clock >>interrupt goes off. >[...] >>The clock increments by 100 ticks every timer interrupt. > >That would make the clock be somewhat slow ;-) Bugger. Hey, 'ns' and 'ms' look pretty similar to me ... 8-/ -- don ###### From: Eric Sosman Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Fri, 21 Jul 2000 10:04:53 -0400 Organization: Sun Microsystems Lines: 17 Message-ID: <39785885.CBE33EE8@east.sun.com> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> <3976D79B.D2943F34@netsight.se> <6ur98ohbak.fsf@chonsp.franklin.ch> NNTP-Posting-Host: tardis.east.sun.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!fu-berlin.de!newsfeed.ision.net!ision!newsfeed.icl.net!newspeer.clara.net!news.clara.net!diablo.netcom.net.uk!netcom.net.uk!easynet-melon!easynet.net!btnet-feed2!btnet!carbon.eu.sun.com!new-usenet.uk.sun.com!eastnews1.east.sun.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60180 Neil Franklin wrote: > > Johnny Billquist writes: > > > Nah. VMS itself don't have any such problems. VMS dates *will* run > > out sometime, but you can rest assured that we're far from there yet. > > Room for 2^64 ticks of 100ns each. No; only 2^63 ticks. VMS time is a signed quantity, non-negative values representing absolute times and negative values representing delta times. Rather convenient, actually: you use the exact same system calls and library routines to specify "17.04 seconds from NOW" as you do to specify "at 13:52:57.23 tomorrow." -- Eric.Sosman@east.sun.com ###### From: glass2@glass2.lexington.ibm.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 21 Jul 2000 15:19:50 GMT Organization: IBM Austin Lines: 14 Message-ID: <8l9pmm$jk2$1@ausnews.austin.ibm.com> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3973B271.763C405@prescienttech.com> <3974935b$0$4933@reader3> <8l26rp$1fpk$1@ausnews.austin.ibm.com> <39777772$1@news.ucsc.edu> Reply-To: wa4qal@vnet.ibm.com NNTP-Posting-Host: glass2.cv.lexington.ibm.com X-Newsreader: IBM NewsReader/2 2.0 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!nmaster.kpnqwest.net!blackbush.xlink.net!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!znr.news.ans.net!news.chips.ibm.com!newsfeed.btv.ibm.com!news2atm.raleigh.ibm.com!ausnews.austin.ibm.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60234 In <39777772$1@news.ucsc.edu>, eugene@cse.ucsc.edu (Eugene Miya) writes: >In article <8l26rp$1fpk$1@ausnews.austin.ibm.com>, > wrote: >>the shuttle computers are IBM AP-101 machines, formerly manufactured >>by IBM FSD (Federal Service Division) in Owego, before FSD was sold > Federal Systems Divison Oswego Oh, phooey! I knew that didn't sound right when I typed it. Thanks for correcting me and keeping me honest. Dave P.S. Standard Disclaimer: I work for them, but I don't speak for them. ###### Message-ID: <3978463B.66DD8871@trailing-edge.com> Date: Fri, 21 Jul 2000 12:46:51 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.2 AlphaServer 1200 5/533 4MB) MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> <3976D79B.D2943F34@netinsight.se> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 45 NNTP-Posting-Host: 63.73.218.130 X-Trace: reader2.news.uu.net 964198012 10531 63.73.218.130 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!rz.uni-karlsruhe.de!schlund.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.onemain.com!feed1.onemain.com!uunet!ffx.uu.net!spool1.news.uu.net!spool0.news.uu.net!reader2.news.uu.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60213 Johnny Billquist wrote: > > Dan wrote: > > > > Tim Shoppa wrote: > > > > > How about systems that haven't been *rebooted* in 17 years? > > > > > > There's a VAX/VMS system running railroad switching in Ireland > > > that has been running VMS 3.x continuously since it was last > > > rebooted in 1983. > > > > I have to be a little concerned about this one, since old versions > > of VAX/VMS had an "end-of-the-epoch" problem (somewhat similar to > > UNIX's year-2038 limitation) on or about May 19, 1997. 10,000 > > days from the start of the epoch, which was 01/01/1970, if I recall. > > This isn't really a VMS problem as such. It's a problem with C-programs > under VMS, which uses the Unix time calls. Those measure time as a > delta time from the Unix epoch, which reached 10,000 days in 1997. Exactly correct. A VMS 3.x system would be almost (completely?) C-free, so completely immune to any bugs in the C Run-time-library. The Day-10000 bug hit some late VMS V6.x and early V7.x releases. > > Of course, this only affected systems that dealt with dates, so > > if the railroad folks are doing an application that's unconcerned > > with dates (which railroad switching may very well be), it might > > simply not be an issue for them. > > Nah. VMS itself don't have any such problems. VMS dates *will* run > out sometime, but you can rest assured that we're far from there yet. > Any other language on VMS continues to work just fine, it was only > C that was affected. And any ssytem components that used the C RTL. In the case of VMS 3.x, there were no such components, so it was safe. The VMS End-of-epoch is 31-Jul-31086 2:48:05.47. Note that current VMS versions only support 4-digit years, but according to an SPR response from DEC "This is expected to be corrected sometime prior to 31-DEC-9999". Tim. ###### Message-ID: <39784803.44BB656@trailing-edge.com> Date: Fri, 21 Jul 2000 12:54:27 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.2 AlphaServer 1200 5/533 4MB) MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3976D79B.D2943F34@netinsight.se> <964099762.86210@shelley.paradise.net.nz> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 22 NNTP-Posting-Host: 63.73.218.130 X-Trace: reader2.news.uu.net 964198467 10531 63.73.218.130 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!news.rhrz.uni-bonn.de!news-koe1.dfn.de!news-fra1.dfn.de!news0.de.colt.net!colt.net!diablo.netcom.net.uk!netcom.net.uk!newspump.monmouth.com!newspeer.monmouth.com!uunet!ffx.uu.net!spool1.news.uu.net!spool0.news.uu.net!reader2.news.uu.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60207 Eugene Miya wrote: > > In article <397711EB.CC1C5723@netinsight.se>, > Johnny Billquist wrote: > >VMS counts in 100ns(?) increments in a 64 bit counter. If I remember > >that correctly, it gives VMS a range of 29227 years, this added to > >1876 would give VMS breaking sometime in the year 31103, or possibly 31104. > > Oh, this is interesting. Compaq changed this? > My foggy memory had a 60 Hz to 50/100 Hz (max) clock which lots of > people confused. Which SYS$call? Eugene, you've proved the old adage "the best way to get correct information on Usenet is to post incorrect information" yet again :-). No, Compaq did not change this; the granularity always has been, and will probably remain, 100 nanoseconds inside VMS. See the famous Y10K SPR at http://www.trailing-edge.com/www/y10k.txt Tim. ###### Message-ID: <397848A1.424F06D7@trailing-edge.com> Date: Fri, 21 Jul 2000 12:57:05 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.2 AlphaServer 1200 5/533 4MB) MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> <3976D79B.D2943F34@netsight.se> <6ur98ohbak.fsf@chonsp.franklin.ch> <3978151E.493527D5@netinsight.se> <8l96el$kad$2@teabag.demon.co.uk> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 17 NNTP-Posting-Host: 63.73.218.130 X-Trace: reader2.news.uu.net 964198625 10531 63.73.218.130 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!news.rhrz.uni-bonn.de!RRZ.Uni-Koeln.DE!news.netcologne.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!152.163.239.129!portc01.blue.aol.com!peerfeed.news.psi.net!uunet!ffx.uu.net!spool1.news.uu.net!spool0.news.uu.net!reader2.news.uu.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60212 Chris Hedley wrote: > > In article <3978151E.493527D5@netinsight.se>, > Johnny Billquist writes: > > (But Unix programmers don't particularly expect the 2038 limit either...) > > I wouldn't be so sure, most I've known are all too well aware of it; > a possible problem is that some expect it to transparently disappear > upon introduction of a longer time_t, which of course doesn't take > into account either less experienced programmers continuing to code > it as a 32-bit signed int or people diddling with existing code and > breaking it. Or, more importantly, existing data structures (such inconsequential things as, say, filesystems) which only have room for 32 bits. Tim. ###### From: cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 21 Jul 2000 17:08:37 GMT Organization: teabag Message-ID: <8la02l$116k$1@teabag.demon.co.uk> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-edge.com> <3976D79B.D2943F34@netsight.se> <6ur98ohbak.fsf@chonsp.franklin.ch> <3978151E.493527D5@netinsight.se> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> NNTP-Posting-Host: teabag.demon.co.uk X-NNTP-Posting-Host: teabag.demon.co.uk:193.237.4.110 X-Trace: news.demon.co.uk 964199917 nnrp-01:11840 NO-IDENT teabag.demon.co.uk:193.237.4.110 X-Complaints-To: abuse@demon.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 21 Jul 2000 17:08:37 GMT X-Newsreader: knews 1.0b.0 Lines: 18 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!diablo.theplanet.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!teabag.demon.co.uk!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60209 In article <397848A1.424F06D7@trailing-edge.com>, Tim Shoppa writes: > Or, more importantly, existing data structures (such inconsequential > things as, say, filesystems) which only have room for 32 bits. Well, call me naive if you like, but I expect that filesystems in vogue in the next 30-odd years will have had time_t extended to 64 bits or have some other means of storing timestamps. Some may worry that filesystems from ancient discs and archives will be problematic but as long as these are considered read-only archives which nobody will attempt to add post-2038 dated files to then, as long as the relevant utilities recognise the magic no of semi-ancient (I believe the term "ancient" has already been reserved for PDP-11 stuff) data it should still be accessible without any spurious date errors. Chris. ###### From: Eric Chomko Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 21 Jul 2000 18:20:14 GMT Organization: IDT Internet Services Lines: 34 Message-ID: <8la48u$h1q@nnrp2.farm.idt.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l146o$14ca@edrn.newsguy.com> NNTP-Posting-Host: u3.farm.idt.net X-Newsreader: TIN [UNIX 1.3 unoff BETA release 961025] Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!rz.uni-karlsruhe.de!schlund.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!netnews.com!newsin.iconnet.net!news.idt.net!nntp.farm.idt.net!u3.farm.idt.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60202 Charles Eicher wrote: [...] : From the early computer era, 30 years doesn't seem to be so rare. One of my : CompSci professors worked on design of the Univac 1, back in the late 1970s he : said he was actively maintaining one of the first Univac 1 models, which was : still in daily use running AP and Payroll. At that time, the CPU would easily : have been at least 25 years old. : Of course, this professor used to proudly describe this Univac 1 to each : incoming class of students, because there would always be some smart aleck who : declare it would be cheaper to set up a cheap PC and a spreadsheet. And the : professor would reply, "Well, what's the fun in THAT?" But the main point he was : trying to make, companies with working systems are reluctant to change, as long : as the current system works adequately (even if it's just BARELY adequate). : Well, that was over 20 years ago, somehow I doubt that this Univac 1 is still : operational, especially since the professor retired, and I heard he died several : years ago. I would like to believe that the machine outlived the professor, : doing accurate computations until it eventually blew a tube and then he was no : longer around to repair it. The last of the Univac repairmen, keeping alive the : last of the Univacs.... Touching story... Conjures up thoughts of old soldiers not dying and such a la Gen. MacArthur. Hey, how about old computer repairmen don't die, they tool away. Or, OCRDD, they get shifted into the bit bucket. Or, OCRDD, they become part of the machine. Or,... Eric ###### Sender: jrm@IVAN-IV Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l146o$14ca@edrn.newsguy.com> <8la48u$h1q@nnrp2.farm.idt.net> From: Joe Marshall Message-ID: Lines: 12 User-Agent: Gnus/5.0803 (Gnus v5.8.3) XEmacs/21.2 (Molpe) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: 21 Jul 2000 14:48:42 -0400 NNTP-Posting-Host: 206.166.172.54 X-Complaints-To: abuse@verio.net X-Trace: iad-read.news.verio.net 964205323 206.166.172.54 (Fri, 21 Jul 2000 18:48:43 GMT) NNTP-Posting-Date: Fri, 21 Jul 2000 18:48:43 GMT Organization: Verio Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sjc-peer.news.verio.net!news.verio.net!iad-read.news.verio.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60199 Eric Chomko writes: > Hey, how about old computer repairmen don't die, they tool away. > > Or, OCRDD, they get shifted into the bit bucket. > > Or, OCRDD, they become part of the machine. > > Or,... Old computer repairmen don't die, their MTTR just increases without bound. ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l146o$14ca@edrn.newsguy.com> <8la48u$h1q@nnrp2.farm.idt.net> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 17 Message-ID: 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 X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Fri, 21 Jul 2000 16:23:04 EDT Date: Fri, 21 Jul 2000 20:23:04 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-koe1.dfn.de!news-was.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!152.163.239.131!portc03.blue.aol.com!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!news-west.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60244 In article , wrote: >>Charles Eicher wrote: >>[...] >>: At that time, the CPU would easily have been at least 25 years old. >... > >That seems kind of silly except for as a museum. A $500 computer from >best buy could easilly hold all the software, their database, and run >an interpreter to run their software. It probably wouldn't run without breaking for 25 years. -- Kragen Sitaker Perilous to all of us are the devices of an art deeper than we ourselves possess. -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"] ###### From: "P Linnane" Newsgroups: alt.folklore.computers References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l146o$14ca@edrn.newsguy.com> <8la48u$h1q@nnrp2.farm.idt.net> Subject: Re: Longest running computer Lines: 42 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.6600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Organization: Ye 'Ol Disorganized NNTPCache groupie Message-ID: <964215436.300480@server16.cable.com> Cache-Post-Path: server16.cable.com!unknown@vp1-92.xcessible.com X-Cache: nntpcache 2.4.0b2 (see http://www.nntpcache.org/) Date: Fri, 21 Jul 2000 17:33:12 -0400 NNTP-Posting-Host: 204.92.64.17 X-Trace: nnrp1.uunet.ca 964215151 204.92.64.17 (Fri, 21 Jul 2000 17:32:31 EDT) NNTP-Posting-Date: Fri, 21 Jul 2000 17:32:31 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!rz.uni-karlsruhe.de!schlund.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sunqbc.risq.qc.ca!news.uunet.ca!nnrp1.uunet.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60203 wrote in message news:slrn8nh88a.3lf.TheCentralScrutinizer.199@C298344-A.arvada1.co.home.com. .. > >Charles Eicher wrote: > >[...] > >: From the early computer era, 30 years doesn't seem to be so rare. One of my > >: CompSci professors worked on design of the Univac 1, back in the late 1970s he > >: said he was actively maintaining one of the first Univac 1 models, which was > >: still in daily use running AP and Payroll. At that time, the CPU would easily > >: have been at least 25 years old. > ... > > That seems kind of silly except for as a museum. A $500 computer from > best buy could easilly hold all the software, their database, and run > an interpreter to run their software. Your wisdom amazes me, so let me get this straight: For $500 I can get:: Complete AP/Payroll and database packages for a medium/large-sized company PC-Interpreter for Univac OS and (presumably COBOL or Fortran) applications PC-compatible terminals x? PC-compatible printers PC-compatible networking Complete conversion of existing system Client re-training 25 year warranty. . . . and you'll even throw in a piece-of-crap computer ? ? Rick I spend an inordinate amount of time avoiding people like you. ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l26rp$1fpk$1@ausnews.austin.ibm.com> <39777772$1@news.ucsc.edu> <8l9pmm$jk2$1@ausnews.austin.ibm.com> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <3978c473$1@news.ucsc.edu> Date: 21 Jul 2000 14:45:23 -0800 X-Trace: 21 Jul 2000 14:45:23 -0800, sundance.cse.ucsc.edu Lines: 16 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-koe1.dfn.de!news-was.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.stanford.edu!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:60294 In article <8l9pmm$jk2$1@ausnews.austin.ibm.com>, wrote: >In <39777772$1@news.ucsc.edu>, eugene@cse.ucsc.edu (Eugene Miya) writes: >>In article <8l26rp$1fpk$1@ausnews.austin.ibm.com>, >> wrote: >>>the shuttle computers are IBM AP-101 machines, formerly manufactured >>>by IBM FSD (Federal Service Division) in Owego, before FSD was sold >> Federal Systems Divison Oswego > >Oh, phooey! I knew that didn't sound right when I typed it. Thanks >for correcting me and keeping me honest. FSD is a common acronym with firms doing business with the government, not just IBM, but also SUN and others. And I have friends who live in Oswego. ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3974935b$0$4933@reader3> <39777916$1@news.ucsc.edu> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <3978c5ea$1@news.ucsc.edu> Date: 21 Jul 2000 14:51:38 -0800 X-Trace: 21 Jul 2000 14:51:38 -0800, sundance.cse.ucsc.edu Lines: 12 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:60265 In article , Alistair Gale wrote: >On 20 Jul 2000 15:11:34 -0800, Eugene Miya wrote: >>Surely there's a #5 ESS sitting somewhere running longer than other machines. > >Don't you mean a #1ESS? There's a tradeoff between size and functionality. I prefer collecting a smaller 5 than a humongous 1. But there might be a 1 running longer, much less all the other intervening machines. 8^) ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <964148894.748064@shelley.paradise.net.nz> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <3978ca32$1@news.ucsc.edu> Date: 21 Jul 2000 15:09:54 -0800 X-Trace: 21 Jul 2000 15:09:54 -0800, sundance.cse.ucsc.edu Lines: 37 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!news.rhrz.uni-bonn.de!news-koe1.dfn.de!news-was.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feeder.via.net!enews.sgi.com!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:60259 In article <397711EB.CC1C5723@netinsight.se>, Johnny Billquist wrote: >>>VMS counts in 100ns(?) increments in a 64 bit counter. If I remember >>>that correctly, it gives VMS a range of 29227 years, this added to >>>1876 would give VMS breaking sometime in the year 31103, or possibly 31104. In article <397776d6$1@news.ucsc.edu>, Eugene Miya wrote: >>Oh, this is interesting. Compaq changed this? >>My foggy memory had a 60 Hz to 50/100 Hz (max) clock which lots of >>people confused. Which SYS$call? In article <964148894.748064@shelley.paradise.net.nz>, Don Stokes wrote: >100 ns is the resolution of the time format. Yes, there's a different between the format (like the BSD time.h structure [sec and usec]) and the function of the clock. >The actual resolution of the VAX/VMS clock is (usually) 10 ms, ie that's >how often the clock interrupt goes off. That's 100 Hz. >IIRC, Some VAXes had higher resolution interval >timers that could (in theory at least) update more frequently. I have heard this, too. Also w.r.t. to IBM 360/370s/etc. Like in the case of the u-VAX based M31/Andromeda system. But I have yet to really see one and program it. I saw the announcement of the book title on the Andromeda, but never got a book from Digital Press. >The clock increments by 100 ticks every timer interrupt. > >I'm not sure whether this was improved in Alpha VMS. I'd doubt it. ###### Message-ID: <3978E801.8E390A1A@mail.bcpl.net> From: Ken McMonigal X-Mailer: Mozilla 4.61 [en] (Win98; I) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3976D79B.D2943F34@netinsight.se> <964099762.86210@shelley.paradise.net.nz> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 29 Date: Fri, 21 Jul 2000 20:17:05 -0400 NNTP-Posting-Host: 208.242.126.195 X-Complaints-To: abuse@bcpl.net X-Trace: news.abs.net 964224895 208.242.126.195 (Fri, 21 Jul 2000 20:14:55 EDT) NNTP-Posting-Date: Fri, 21 Jul 2000 20:14:55 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!news.rhrz.uni-bonn.de!news-koe1.dfn.de!news-was.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.abs.net!news.abs.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60271 Tim Shoppa wrote: > Eugene Miya wrote: > > > > In article <397711EB.CC1C5723@netinsight.se>, > > Johnny Billquist wrote: > > >VMS counts in 100ns(?) increments in a 64 bit counter. If I remember > > >that correctly, it gives VMS a range of 29227 years, this added to > > >1876 would give VMS breaking sometime in the year 31103, or possibly 31104. > > > > Oh, this is interesting. Compaq changed this? > > My foggy memory had a 60 Hz to 50/100 Hz (max) clock which lots of > > people confused. Which SYS$call? > > Eugene, you've proved the old adage "the best way to get correct > information on Usenet is to post incorrect information" yet again :-). > > No, Compaq did not change this; the granularity always has been, > and will probably remain, 100 nanoseconds inside VMS. See the famous > Y10K SPR at > > http://www.trailing-edge.com/www/y10k.txt > > Tim. Which means Arthur C. Clarke will be on book # ? ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 22 Jul 2000 01:06:00 GMT Organization: Neosoft (using Airnews.net!) Lines: 6 Message-ID: <48A6A565160289F7.B6AB1B2AC2E18C44.8D36189F69B3CB75@lp.airnews.net> X-Orig-Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <964099762.86210@shelley.paradise.net.nz> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <964148894.748064@shelley.paradise.net.nz> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: 204.181.96.50 NNTP-Posting-Time: Fri Jul 21 20:06:00 2000 NNTP-Posting-Host: !b$U)1k-V\"['6R (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!newsfeed.Austria.EU.net!nmaster.kpnqwest.net!npeer.kpnqwest.net!nntp.primenet.com!nntp.gblx.net!feeder.via.net!HSNX.atgi.net!cabal11.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal1.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:60272 On Fri, 21 Jul 2000 03:08:33 GMT, Don Stokes wrote: >The clock increments by 100 ticks every timer interrupt. >I'm not sure whether this was improved in Alpha VMS. Good question. The Alpha's certainly capable of it; the PALcode for Linux (and, I believe, Tru64 Unix) produces 1024 ticks/second. ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <964099762.86210@shelley.paradise.net.nz> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <964148894.748064@shelley.paradise.net.nz> <48A6A565160289F7.B6AB1B2AC2E18C44.8D36189F69B3CB75@lp.airnews.net> From: Ric Werme X-Newsreader: NN version 6.5.0 CURRENT #119 Lines: 28 Message-ID: Date: Sat, 22 Jul 2000 01:54:35 GMT NNTP-Posting-Host: 24.91.12.15 X-Complaints-To: abuse@mediaone.net X-Trace: typhoon.ne.mediaone.net 964230875 24.91.12.15 (Fri, 21 Jul 2000 21:54:35 EDT) NNTP-Posting-Date: Fri, 21 Jul 2000 21:54:35 EDT Organization: Road Runner Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!rz.uni-karlsruhe.de!schlund.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.cwix.com!chnws02.mediaone.net!chnws05.ne.mediaone.net!24.128.8.70!typhoon.ne.mediaone.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60285 jmaynard@thebrain.conmicro.cx (Jay Maynard) writes: >On Fri, 21 Jul 2000 03:08:33 GMT, Don Stokes wrote: >>The clock increments by 100 ticks every timer interrupt. >>I'm not sure whether this was improved in Alpha VMS. >Good question. The Alpha's certainly capable of it; the PALcode for Linux >(and, I believe, Tru64 Unix) produces 1024 ticks/second. Well, I can't speak for VMS (or Tru64 for that matter), but there's a barely known tweak to Tru64. If you enable the MICRO_TIME option in the /sys/conf/NAME file (and I thing NTP_TIME or somesuch) and rebuild your kernel, gettimeofday() will adjust the time based on the CPU's cycle counter to give you near microsecond resolution. If you need anything finer, you can always read the cycle counter with the RPCC instruction or library routine. (Too lazy to check from home.) I liked Alliant's FX/2800 system. An I/O board (one i860 and 2 VME busses) had logic to increment a memory location every 1 or 10 usec. Concentrix had it do so to a page that was mapped into all programs, so anyone could do high resolution timing without having to deal with cycle counter resynchs if their program moves from CPU to CPU (or ran on several CPUs at once). -Ric Werme -- Ric Werme | werme@nospam.mediaone.net http://people.ne.mediaone.net/werme | ^^^^^^^ delete ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 22 Jul 2000 02:16:21 GMT Organization: Neosoft (using Airnews.net!) Lines: 23 Message-ID: <441EF078B6B19400.7B2F5BFE6C01266F.68C4AF216247D0AA@lp.airnews.net> X-Orig-Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <964099762.86210@shelley.paradise.net.nz> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <964148894.748064@shelley.paradise.net.nz> <48A6A565160289F7.B6AB1B2AC2E18C44.8D36189F69B3CB75@lp.airnews.net> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: 204.181.96.50 NNTP-Posting-Time: Fri Jul 21 21:16:22 2000 NNTP-Posting-Host: ![/p:1k-XU7@X"p (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!news-out.nntp.airnews.net.MISMATCH!cabal10.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal1.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:60270 On Sat, 22 Jul 2000 01:54:35 GMT, Ric Werme wrote: >Well, I can't speak for VMS (or Tru64 for that matter), but there's a >barely known tweak to Tru64. If you enable the MICRO_TIME option in >the /sys/conf/NAME file (and I thing NTP_TIME or somesuch) and rebuild >your kernel, gettimeofday() will adjust the time based on the CPU's >cycle counter to give you near microsecond resolution. If you need >anything finer, you can always read the cycle counter with the RPCC >instruction or library routine. (Too lazy to check from home.) Hmmm. Wonder if that's easy to add to Linux. ...and if there's a similar thing that can be added to Intel platforms. I'm running into problems with Hercules' emulation of the 370 time-of-day clock, which is defined as a 64-bit value where bit 51 (0=MSB) is incremented every microsecond...and the standard Linux timing facilities only provide 1/100 second resolution. If a program does things like try to time the CPU by reading the TOD clock, running a 50,000 instruction loop, reading the TOD clock again, and dividing, on a fast enough machine, that'll get a divide-by-zero program check instead of a result. (As Hercules gets faster, the bar for "fast enough machine" gets lower, and lower...) My current plan is to use the Linux /dev/rtc to provide a faster tick (by default, 1024/second, but settable to 8192/second). If there's a reasonably portable way to get faster, I'll use it. ###### Newsgroups: alt.folklore.computers From: Terry Kennedy Subject: Re: Longest running computer X-Complaints-To: Email abuse@spc.edu if this posting is inappropriate User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (BSD/OS/4.1 (i386)) NNTP-Posting-Host: gate.tmk.com Organization: St. Peter's College, US Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <964148894.748064@shelley.paradise.net.nz> <3978ca32$1@news.ucsc.edu> X-Trace: spcuna.spc.edu 964243177 7374 terry [204.141.35.61] Date: Sat, 22 Jul 2000 05:19:38 GMT Lines: 15 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!newsfeedZ.netscum.dQ!netscum.int!cpk-news-hub1.bbnplanet.com!crtntx1-snh1.gtei.net!news.gtei.net!dfw-peer.news.verio.net!ord-feed.news.verio.net!news.verio.net!news.spc.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60269 Eugene Miya writes: > In article <397711EB.CC1C5723@netinsight.se>, > Johnny Billquist wrote: >>The clock increments by 100 ticks every timer interrupt. >>I'm not sure whether this was improved in Alpha VMS. > > I'd doubt it. Actually, it has. Instead of 10ms, the resolution is now 1ms. On VAXen, you could install a KW11-P clock and write a driver for it if you wanted higher- resolution timing, but that would be unrelated to the system timekeeping func- tion. Terry Kennedy http://www.tmk.com terry@tmk.com Jersey City, NJ USA ###### Path: chonsp.franklin.ch!not-for-mail From: Neil Franklin Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 22 Jul 2000 17:49:52 +0200 Organization: My own Private Self Lines: 38 Message-ID: <6ur98mmafz.fsf@chonsp.franklin.ch> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-dge.com> <3976D79B.D2943F34@netsight.se> <6ur98ohbak.fsf@chonsp.franklin.ch> <3978151E.493527D5@netinsight.se> NNTP-Posting-Host: chonsp.franklin.ch X-Trace: chonsp.franklin.ch 964280992 508 10.0.3.2 (22 Jul 2000 15:49:52 GMT) X-Complaints-To: news@chonsp.franklin.ch NNTP-Posting-Date: 22 Jul 2000 15:49:52 GMT X-Newsreader: Gnus v5.7/Emacs 20.4 Xref: chonsp.franklin.ch alt.folklore.computers:60297 Johnny Billquist writes: > Neil Franklin wrote: > > > Of course not. My complaint here was not with the Unix programmers, > but with the VAXCRTL implementors. OK, I missunderstood you because of: > > > compiled on VMS. (Once again, Unix rears its ugly head... :-) So I thought you were attributiong it to Unix (which is to blame for 2038, but not for 10000 days). > > Unix (and C) count in _seconds_ since 1970, which is why they go wrong > > on 2^31 seconds = in 2038. There is nowhere in Unix (or C) anythong > > about 10'000 days, that is just an VAXCRTL nonsense. > > Yes. But it is rather funny that we have so short limits in *both* cases, 68 years must have looked like eternity in 1970. :-) > > > Nah. VMS itself don't have any such problems. VMS dates *will* run > > > out sometime, but you can rest assured that we're far from there yet. > > > > Room for 2^64 ticks of 100ns each. and negative values > are delta times (unless my memory fails me). Overloading the top bit for deltas. A hidden flag. Hmmm. Well that halves the time range then. -- Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ Nerd, Geek, Hacker, Unix Guru, Sysadmin, Roleplayer, LARPer, Mystic ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Sat, 22 Jul 00 11:14:54 GMT Organization: UltraNet Communications, Inc. Lines: 24 Message-ID: <8lca47$5co$7@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> <964273344.185144@shelley.paradise.net.nz> X-Trace: ymY/EQv1F6ulx/fp+mX2MKeIJk9gdxOffB6rYNVKgPw= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 22 Jul 2000 14:12:23 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!rz.uni-karlsruhe.de!schlund.de!newsfeed01.sul.t-online.de!t-online.de!blackbush.xlink.net!news0.de.colt.net!colt.net!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-245-24 Xref: chonsp.franklin.ch alt.folklore.computers:60302 In article <964273344.185144@shelley.paradise.net.nz>, don@news.daedalus.co.nz (Don Stokes) wrote: >In article <8lc76n$5co$3@bob.news.rcn.net>, wrote: >>Why spend time when a NVMS$;-3DIRSX$EX$$ will do? Answering SPRs >>were not a favorite pasttime for most people. Actually, I can't >>think of a single person who liked to answer them. > >Yeah, but the SPR response completely answers the RSX Y2K problem that >is the actual subject of the SPR (albeit far from satisfactorily, >basically, "bother us again in 1999"), then purely as an aside, appends >the RSXified VMS Y2K leap year SPR response. In fact, it refers to >the attached material as "a rather famous SPR regarding why VMS treats >2000 as a leap year." > >Weird. I don't know how VMS did their SPR answer screening. If it was anything similar to ours, it was probably the only was the engineer could get a real answer out and still pass the screening rules. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Sat, 22 Jul 00 09:11:09 GMT Organization: UltraNet Communications, Inc. Lines: 56 Message-ID: <8lc2s7$g9e$1@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l146o$14ca@edrn.newsguy.com> <8la48u$h1q@nnrp2.farm.idt.net> <964215436.300480@server16.cable.com> X-Trace: 2qX4riJ0NhNfrPEPNPrI67x4hTYYhmQ8LHHMzp1eDeI= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 22 Jul 2000 12:08:39 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!uni-erlangen.de!newsfeeds.belnet.be!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-245-52 Xref: chonsp.franklin.ch alt.folklore.computers:60305 In article <964215436.300480@server16.cable.com>, "P Linnane" wrote: > > wrote in message >news:slrn8nh88a.3lf.TheCentralScrutinizer.199@C298344-A.arvada1.co.home.co m. >... >> >Charles Eicher wrote: >> >[...] >> >: From the early computer era, 30 years doesn't seem to be so rare. One >of my >> >: CompSci professors worked on design of the Univac 1, back in the late >1970s he >> >: said he was actively maintaining one of the first Univac 1 models, >which was >> >: still in daily use running AP and Payroll. At that time, the CPU would >easily >> >: have been at least 25 years old. >> ... >> >> That seems kind of silly except for as a museum. A $500 computer from >> best buy could easilly hold all the software, their database, and run >> an interpreter to run their software. > >Your wisdom amazes me, so let me get this straight: > >For $500 I can get:: > >Complete AP/Payroll and database packages for a medium/large-sized company >PC-Interpreter for Univac OS and (presumably COBOL or Fortran) applications >PC-compatible terminals x? >PC-compatible printers >PC-compatible networking >Complete conversion of existing system >Client re-training >25 year warranty. > >.. . . and you'll even throw in a piece-of-crap computer ? >? Very nice :-). My reaction was, "Kids!". > >Rick > >I spend an inordinate amount of time avoiding people like you. Are you successful? If you are, I know a couple of people who would pay for tips of avoidance success. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Sat, 22 Jul 00 10:25:03 GMT Organization: UltraNet Communications, Inc. Lines: 33 Message-ID: <8lc76n$5co$3@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> X-Trace: LevY+eoa5inXVBiStqbQ2LvdYrt0B7BDEjuPXti2tTg= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 22 Jul 2000 13:22:31 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!europa.netcrusader.net!207.172.3.37!feed1.news.rcn.net!rcn!207-172-245-24 Xref: chonsp.franklin.ch alt.folklore.computers:60306 In article <964269064.98678@shelley.paradise.net.nz>, don@news.daedalus.co.nz (Don Stokes) wrote: >In article <39784803.44BB656@trailing-edge.com>, >Tim Shoppa wrote: >>No, Compaq did not change this; the granularity always has been, >>and will probably remain, 100 nanoseconds inside VMS. See the famous >>Y10K SPR at >> >> http://www.trailing-edge.com/www/y10k.txt > >On the subject of date SPRs, there's also: > > http://www.openvms.digital.com/openvms/products/year-2000/leap.html > >a copy of which was forwarded to me by a friendly Digit many moons >ago, appended to an SPR from 1989 about two-digit RSX dates and Y2K, >which talked about year problems in OS/8 and OS/78, before copying the >above. > >Oddly enough, the appended copy refers to RSX, not VMS, although the >wording is otherwise identical. Why spend time when a NVMS$;-3DIRSX$EX$$ will do? Answering SPRs were not a favorite pasttime for most people. Actually, I can't think of a single person who liked to answer them. /BAH /BAH Subtract a hundred and four for e-mail. ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> Organization: Daedalus Consulting X-Newsreader: trn 4.0-test72 (19 April 1999) From: don@news.daedalus.co.nz (Don Stokes) Message-ID: <964269064.98678@shelley.paradise.net.nz> Cache-Post-Path: shelley.paradise.net.nz!unknown@203-96-144-16.cable.paradise.net.nz X-Cache: nntpcache 2.4.0b5 (see http://www.nntpcache.org/) Lines: 24 Date: Sat, 22 Jul 2000 12:31:23 GMT NNTP-Posting-Host: 203.96.152.26 X-Complaints-To: newsadmin@xtra.co.nz X-Trace: news.xtra.co.nz 964269083 203.96.152.26 (Sun, 23 Jul 2000 00:31:23 NZST) NNTP-Posting-Date: Sun, 23 Jul 2000 00:31:23 NZST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!nyc-news-feed1.bbnplanet.com!news.gtei.net!colt.net!news.maxwell.syr.edu!feeder.via.net!enews.sgi.com!news.xtra.co.nz!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60321 In article <39784803.44BB656@trailing-edge.com>, Tim Shoppa wrote: >No, Compaq did not change this; the granularity always has been, >and will probably remain, 100 nanoseconds inside VMS. See the famous >Y10K SPR at > > http://www.trailing-edge.com/www/y10k.txt On the subject of date SPRs, there's also: http://www.openvms.digital.com/openvms/products/year-2000/leap.html a copy of which was forwarded to me by a friendly Digit many moons ago, appended to an SPR from 1989 about two-digit RSX dates and Y2K, which talked about year problems in OS/8 and OS/78, before copying the above. Oddly enough, the appended copy refers to RSX, not VMS, although the wording is otherwise identical. Unfortunately, the copyright notice forbids redistribution, and although I can find plenty of copies of the VMS leap year SPR on the web, I can't find another copy of the RSX Y2K SPR. -- don ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> Organization: Daedalus Consulting X-Newsreader: trn 4.0-test72 (19 April 1999) From: don@news.daedalus.co.nz (Don Stokes) Message-ID: <964273344.185144@shelley.paradise.net.nz> Cache-Post-Path: shelley.paradise.net.nz!unknown@203-96-144-16.cable.paradise.net.nz X-Cache: nntpcache 2.4.0b5 (see http://www.nntpcache.org/) Lines: 15 Date: Sat, 22 Jul 2000 13:42:43 GMT NNTP-Posting-Host: 203.96.152.26 X-Complaints-To: newsadmin@xtra.co.nz X-Trace: news.xtra.co.nz 964273363 203.96.152.26 (Sun, 23 Jul 2000 01:42:43 NZST) NNTP-Posting-Date: Sun, 23 Jul 2000 01:42:43 NZST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!news-hog.berkeley.edu!ucberkeley!enews.sgi.com!news.xtra.co.nz!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60324 In article <8lc76n$5co$3@bob.news.rcn.net>, wrote: >Why spend time when a NVMS$;-3DIRSX$EX$$ will do? Answering SPRs >were not a favorite pasttime for most people. Actually, I can't >think of a single person who liked to answer them. Yeah, but the SPR response completely answers the RSX Y2K problem that is the actual subject of the SPR (albeit far from satisfactorily, basically, "bother us again in 1999"), then purely as an aside, appends the RSXified VMS Y2K leap year SPR response. In fact, it refers to the attached material as "a rather famous SPR regarding why VMS treats 2000 as a leap year." Weird. -- don ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> Organization: Plethora . Net - More Net, Less Spam! X-Newsreader: trn 4.0-test72 (19 April 1999) From: seebs@plethora.net (Peter Seebach) Date: 22 Jul 2000 15:01:41 GMT Lines: 27 Message-ID: <3979b755$0$8317$3c090ad1@news.plethora.net> NNTP-Posting-Host: e46c5b61.news.plethora.net X-Trace: 964278101 gemini.plethora.net 8317 seebs@205.166.146.8 X-Complaints-To: abuse@plethora.net Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!feed2.onemain.com!feed1.onemain.com!newsfeed.wirehub.nl!hermes.visi.com!news-out.visi.com!gemini-int.visi.com.MISMATCH!gemini.plethora.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60323 In article <8la02l$116k$1@teabag.demon.co.uk>, Chris Hedley wrote: >Some may worry that filesystems from ancient discs and archives >will be problematic but as long as these are considered read-only >archives which nobody will attempt to add post-2038 dated files >to then, as long as the relevant utilities recognise the magic no >of semi-ancient (I believe the term "ancient" has already been >reserved for PDP-11 stuff) data it should still be accessible >without any spurious date errors. Indeed. I believe NetBSD had a period during which it could read (in read-only mode) 32-bit Berkeley FFS, and could read and write 64-bit FFS, and fsck had an option to upgrade a filesystem to the new structures. I tested this because I was converting from SVR4, which was using the old FFS. Mounted filesystems, looked around, unmounted, prayed, ran fsck, mounted filesystems. That easy. Why would it be harder with times instead of file sizes? -s -- Copyright 2000, All rights reserved. Peter Seebach / seebs@plethora.net C/Unix wizard, Pro-commerce radical, Spam fighter. Boycott Spamazon! Consulting & Computers: http://www.plethora.net/ ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 22 Jul 2000 15:09:20 GMT Organization: Neosoft (using Airnews.net!) Lines: 5 Message-ID: X-Orig-Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Sat Jul 22 10:09:20 2000 NNTP-Posting-Host: !Z\!]1k-W+GSnUr (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feeder.via.net!HSNX.atgi.net!cabal11.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal1.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:60299 On 22 Jul 2000 15:01:41 GMT, Peter Seebach wrote: >Mounted filesystems, looked around, unmounted, prayed, ran fsck, mounted >filesystems. That easy. What, no backup? ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> Organization: Plethora . Net - More Net, Less Spam! X-Newsreader: trn 4.0-test72 (19 April 1999) From: seebs@plethora.net (Peter Seebach) Date: 22 Jul 2000 15:19:29 GMT Lines: 16 Message-ID: <3979bb81$0$8303$3c090ad1@news.plethora.net> NNTP-Posting-Host: e46c5b61.news.plethora.net X-Trace: 964279169 gemini.plethora.net 8303 seebs@205.166.146.8 X-Complaints-To: abuse@plethora.net Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newscore.univie.ac.at!uninett.no!news.tele.dk!newsfeedZ.netscum.dQ!netscum.int!hermes.visi.com!news-out.visi.com!gemini-int.visi.com.MISMATCH!gemini.plethora.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60327 In article , Jay Maynard wrote: >On 22 Jul 2000 15:01:41 GMT, Peter Seebach wrote: >>Mounted filesystems, looked around, unmounted, prayed, ran fsck, mounted >>filesystems. That easy. >What, no backup? I had one, but there were long and complicated hardware issues... In any event, I didn't need it. -s -- Copyright 2000, All rights reserved. Peter Seebach / seebs@plethora.net C/Unix wizard, Pro-commerce radical, Spam fighter. Boycott Spamazon! Consulting & Computers: http://www.plethora.net/ ###### Path: chonsp.franklin.ch!not-for-mail From: Neil Franklin Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 22 Jul 2000 20:40:44 +0200 Organization: My own Private Self Lines: 21 Message-ID: <6uhf9im2j7.fsf@chonsp.franklin.ch> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> NNTP-Posting-Host: chonsp.franklin.ch X-Trace: chonsp.franklin.ch 964291244 870 10.0.3.2 (22 Jul 2000 18:40:44 GMT) X-Complaints-To: news@chonsp.franklin.ch NNTP-Posting-Date: 22 Jul 2000 18:40:44 GMT X-Newsreader: Gnus v5.7/Emacs 20.4 Xref: chonsp.franklin.ch alt.folklore.computers:60331 seebs@plethora.net (Peter Seebach) writes: > Chris Hedley wrote: > >to then, as long as the relevant utilities recognise the magic no > >of semi-ancient ... data it should still be accessible > >without any spurious date errors. > > Indeed. I believe NetBSD had a period during which it could read (in > read-only mode) 32-bit Berkeley FFS, and could read and write 64-bit FFS, > and fsck had an option to upgrade a filesystem to the new structures. > > Why would it be harder with times instead of file sizes? File contents. Files written by user software that contain 32 bit date fields. The same problem an with 2 digit year numbers. Y2K tidy up code audits all over again. -- Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ Nerd, Geek, Hacker, Unix Guru, Sysadmin, Roleplayer, LARPer, Mystic ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Sun, 23 Jul 00 09:48:53 GMT Organization: UltraNet Communications, Inc. Lines: 31 Message-ID: <8lepf8$gfh$6@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> X-Trace: lpECzgFOK4nc4q6yvyfQnvtQ2Uu1YM70k8PUtt/TtHg= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 23 Jul 2000 12:46:32 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!rz.uni-karlsruhe.de!schlund.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!diablo.theplanet.net!newsfeed.icl.net!netnews.com!feed1.news.rcn.net!rcn!207-172-245-230 Xref: chonsp.franklin.ch alt.folklore.computers:60342 In article <8lehhe$g74$1@flywheel.innovative.iinet.net.au>, bernie@innovative.iinet.net.au (Bernd Felsche) wrote: >Alistair@caribsurf.com (Alistair Gale) writes: > >>On 22 Jul 2000 20:40:44 +0200, Neil Franklin wrote: >>>File contents. Files written by user software that contain 32 bit date >>>fields. The same problem an with 2 digit year numbers. Y2K tidy up >>>code audits all over again. > >>So which database programs use Unix time stamps as opposed to >>Y/M/D? Is this just a red herring? > >Most don't use Unix time() (or timeval() stamp but a derived value. >This is because it's common to require the storage of dates prior to >1970 and beyond 2038. > >Timestamps in commercial databases are AFAIK in the same internal >storage format as the data date-time. > >That also provides platform independence. > >Some mainframe (DB2?) databases store dates as packed digits >equivalent to CYYMMDD where "C" is a century indicator, usually "0" >for 1900 to 1999 and "1" for 2000 to 2099. I've never seen that before. Has it been in use for a long time? Or is this just another quick and dirty fix for Y2K. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Sun, 23 Jul 00 09:47:01 GMT Organization: UltraNet Communications, Inc. Lines: 26 Message-ID: <8lepbo$gfh$5@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8lc76n$5co$3@bob.news.rcn.net> <964273344.185144@shelley.paradise.net.nz> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> X-Trace: jXwlPlWefScswpmkSGAA8/gYKh6Ws12Azl66LPHySAg= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 23 Jul 2000 12:44:40 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!news.rhrz.uni-bonn.de!RRZ.Uni-Koeln.DE!news.netcologne.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-245-230 Xref: chonsp.franklin.ch alt.folklore.computers:60345 In article <964298714.421191@shelley.paradise.net.nz>, don@news.daedalus.co.nz (Don Stokes) wrote: >Tim Shoppa wrote: >>This why I laugh when Unix weenies tell me that to fix the Y2038 problem >>all they have to do is change to a 64 bit time_t and recompile. >>It's so easy, but so very wrong, to ignore back-compatibility with >>existing data structures! > >Well, time_t is a *signed* 32 bit int. For file systems, barring >idiocy, there aren't any file dates < 1970, Since I started contributing to social security in 1962, I would certainly hope that there is software that can read file dates before 1970. > ... so they can get an extra 68 >years without a new file system by the simple expedient of declaring the >time stamps to be unsigned. Cobine it with a 64 bit internal time_t (or >declare time_t to be unsigned as well, which I'm sure would be bug food) >and watch your file system problem get deferred until 2106. It's thinking like this that caused all the Y2K expenditures. /BAH Subtract a hundred and four for e-mail. ###### Message-ID: <3979B287.3EA2B030@trailing-edge.com> Date: Sat, 22 Jul 2000 14:41:11 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.2 AlphaServer 1200 5/533 4MB) MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> <964273344.185144@shelley.paradise.net.nz> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 22 NNTP-Posting-Host: 63.73.218.130 X-Trace: reader2.news.uu.net 964291271 17108 63.73.218.130 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!rz.uni-karlsruhe.de!schlund.de!news-fra.pop.de!ams.uu.net!ffx.uu.net!spool0.news.uu.net!reader2.news.uu.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60348 Don Stokes wrote: > > In article <8lc76n$5co$3@bob.news.rcn.net>, wrote: > >Why spend time when a NVMS$;-3DIRSX$EX$$ will do? Answering SPRs > >were not a favorite pasttime for most people. Actually, I can't > >think of a single person who liked to answer them. > > Yeah, but the SPR response completely answers the RSX Y2K problem that > is the actual subject of the SPR (albeit far from satisfactorily, > basically, "bother us again in 1999") They leave out another big problem: ODS-1 filesystem Y2K support. This required more work than all the other basic Y2K issues in the kernel of RSX. (And the resulting fix, while functional, is hardly pretty.) This why I laugh when Unix weenies tell me that to fix the Y2038 problem all they have to do is change to a 64 bit time_t and recompile. It's so easy, but so very wrong, to ignore back-compatibility with existing data structures! Tim. ###### From: cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 22 Jul 2000 18:59:37 GMT Organization: teabag Message-ID: <8lcqup$8cn$1@teabag.demon.co.uk> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> <964273344.185144@shelley.paradise.net.nz> <3979B287.3EA2B030@trailing-edge.com> NNTP-Posting-Host: teabag.demon.co.uk X-NNTP-Posting-Host: teabag.demon.co.uk:193.237.4.110 X-Trace: news.demon.co.uk 964293346 nnrp-09:3125 NO-IDENT teabag.demon.co.uk:193.237.4.110 X-Complaints-To: abuse@demon.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 22 Jul 2000 18:59:37 GMT X-Newsreader: knews 1.0b.0 Lines: 16 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-mue1.dfn.de!news-was.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!dispose.news.demon.net!demon!news.demon.co.uk!demon!teabag.demon.co.uk!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60349 In article <3979B287.3EA2B030@trailing-edge.com>, Tim Shoppa writes: > This why I laugh when Unix weenies tell me that to fix the Y2038 problem > all they have to do is change to a 64 bit time_t and recompile. > It's so easy, but so very wrong, to ignore back-compatibility with > existing data structures! If that was in response to my comment in another thread ("Unix Weenie," indeed, I've been using VMS longer) it is, in essence, that simple, provided those doing the recompine have the presence of mind to build the longer-timestamped filesystem as a different version and keep the old driver, preferably which forces read-only access at some point prior to 2038, intact so it can still read all those old volumes and give the correct time & datestamps. Chris. ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8lc76n$5co$3@bob.news.rcn.net> <964273344.185144@shelley.paradise.net.nz> <3979B287.3EA2B030@trailing-edge.com> Organization: Daedalus Consulting X-Newsreader: trn 4.0-test72 (19 April 1999) From: don@news.daedalus.co.nz (Don Stokes) Message-ID: <964298714.421191@shelley.paradise.net.nz> Cache-Post-Path: shelley.paradise.net.nz!unknown@203-96-144-16.cable.paradise.net.nz X-Cache: nntpcache 2.4.0b5 (see http://www.nntpcache.org/) Lines: 14 Date: Sat, 22 Jul 2000 20:45:33 GMT NNTP-Posting-Host: 203.96.152.26 X-Complaints-To: newsadmin@xtra.co.nz X-Trace: news.xtra.co.nz 964298733 203.96.152.26 (Sun, 23 Jul 2000 08:45:33 NZST) NNTP-Posting-Date: Sun, 23 Jul 2000 08:45:33 NZST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!rz.uni-karlsruhe.de!schlund.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-hog.berkeley.edu!ucberkeley!enews.sgi.com!news.xtra.co.nz!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60359 Tim Shoppa wrote: >This why I laugh when Unix weenies tell me that to fix the Y2038 problem >all they have to do is change to a 64 bit time_t and recompile. >It's so easy, but so very wrong, to ignore back-compatibility with >existing data structures! Well, time_t is a *signed* 32 bit int. For file systems, barring idiocy, there aren't any file dates < 1970, so they can get an extra 68 years without a new file system by the simple expedient of declaring the time stamps to be unsigned. Cobine it with a 64 bit internal time_t (or declare time_t to be unsigned as well, which I'm sure would be bug food) and watch your file system problem get deferred until 2106. -- don ###### From: Alistair@caribsurf.com (Alistair Gale) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 23 Jul 2000 00:02:58 GMT Organization: http://extra.newsguy.com Lines: 12 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> Reply-To: alistair@caribsurf.com NNTP-Posting-Host: p-515.newsdawg.com User-Agent: slrn/0.9.5.7 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-mue1.dfn.de!news-nue1.dfn.de!uni-erlangen.de!newsfeed.germany.net!newsfeed.berkeley.edu!ucberkeley!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!Alistair Xref: chonsp.franklin.ch alt.folklore.computers:60335 On 22 Jul 2000 20:40:44 +0200, Neil Franklin wrote: >File contents. Files written by user software that contain 32 bit date >fields. The same problem an with 2 digit year numbers. Y2K tidy up >code audits all over again. So which database programs use Unix time stamps as opposed to Y/M/D? Is this just a red herring? -- alistair BA: "Ohh, I want to bed down with a dozen concubines." Baldrick:" Wouldn't that be all prickly?" ###### From: Erno Palonheimo Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 23 Jul 2000 05:59:37 +0300 Organization: Helsinki University of Technology, Finland Lines: 16 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> NNTP-Posting-Host: vipunen.hut.fi X-Trace: nntp.hut.fi 964321177 21972 130.233.249.7 (23 Jul 2000 02:59:37 GMT) X-Complaints-To: usenet@nntp.hut.fi NNTP-Posting-Date: 23 Jul 2000 02:59:37 GMT X-Newsreader: Gnus v5.3/Emacs 19.33 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!nyc-news-feed1.bbnplanet.com!news.gtei.net!colt.net!newspeer.highwayone.net!newsfeed1.telenordia.se!algonet!news.wineasy.se!newsfeed.wineasy.se!news.clinet.fi!news.cs.hut.fi!nntp.hut.fi!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60352 "Dennis Eldridge" writes: > So here's an interesting topic: Old systems still in production use. I'd be > very interested to hear stories of systems which are either still in use or > recently retired, which have lasted at least 15 years. We have a IBM 370/XA-compatible mainframe, Hitachi AS/8040, with its original disks, tape drives and all in production use. I've been told that it was installed in 1984. The only part that has ever broken is the 8" floppy containing the CPU microcode (not the drive but the actual floppy). The machine will be in service at least until 2001, I hope. It is running a large database containing student records. -- --- Erno Palonheimo -- Helsinki University of Technology ----- esp@cc.hut.fi -- Administrative Manouvers in the Dark ###### From: bernie@innovative.iinet.net.au (Bernd Felsche) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 23 Jul 2000 18:31:10 +0800 Organization: Innovative Reckoning Message-ID: <8lehhe$g74$1@flywheel.innovative.iinet.net.au> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> X-Newsreader: NN version 6.5.1 (NOV) Lines: 25 NNTP-Posting-Host: innovative.iinet.net.au X-Trace: news.iinet.net.au 964348846 8843 203.59.144.24 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!awabi.library.ucla.edu!216.65.16.3!news-out.nibble.net!news-in.nibble.net!feed.newsfeeds.com!newsfeeds.com!newsfeed.iinet.net.au!news.iinet.net.au!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60334 Alistair@caribsurf.com (Alistair Gale) writes: >On 22 Jul 2000 20:40:44 +0200, Neil Franklin wrote: >>File contents. Files written by user software that contain 32 bit date >>fields. The same problem an with 2 digit year numbers. Y2K tidy up >>code audits all over again. >So which database programs use Unix time stamps as opposed to >Y/M/D? Is this just a red herring? Most don't use Unix time() (or timeval() stamp but a derived value. This is because it's common to require the storage of dates prior to 1970 and beyond 2038. Timestamps in commercial databases are AFAIK in the same internal storage format as the data date-time. That also provides platform independence. Some mainframe (DB2?) databases store dates as packed digits equivalent to CYYMMDD where "C" is a century indicator, usually "0" for 1900 to 1999 and "1" for 2000 to 2099. -- Bernd Felsche - Innovative Reckoning Perth, Western Australia ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> Organization: Daedalus Consulting X-Newsreader: trn 4.0-test72 (19 April 1999) From: don@news.daedalus.co.nz (Don Stokes) Message-ID: <964358954.300056@shelley.paradise.net.nz> Cache-Post-Path: shelley.paradise.net.nz!unknown@203-96-144-16.cable.paradise.net.nz X-Cache: nntpcache 2.4.0b5 (see http://www.nntpcache.org/) Lines: 29 Date: Sun, 23 Jul 2000 13:29:33 GMT NNTP-Posting-Host: 203.96.152.26 X-Complaints-To: newsadmin@xtra.co.nz X-Trace: news.xtra.co.nz 964358973 203.96.152.26 (Mon, 24 Jul 2000 01:29:33 NZST) NNTP-Posting-Date: Mon, 24 Jul 2000 01:29:33 NZST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!news.xtra.co.nz!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60358 In article <8lepbo$gfh$5@bob.news.rcn.net>, wrote: > don@news.daedalus.co.nz (Don Stokes) wrote: >>Well, time_t is a *signed* 32 bit int. For file systems, barring >>idiocy, there aren't any file dates < 1970, > > Since I started contributing to social security >in 1962, I would certainly hope that there is software that >can read file dates before 1970. I'm talking filesystems only. Eg, if I mount a UFS or SysV filesystem, I can treat the 32 bit dates as unsigned instead of signed. I'd read the numbers into a 64 bit int for manipulation, and time_t would be 64 bits. The translation to/from 32 bits only happens when dealling with these old filesystems, and only for file dates. Hopefully, these things will all be dead by 2106. Mind you, there seems to be an assumption by those exposed only to personal computers that there's a natural progression 4-8-16-32 bits in word sizes which will continue through 64 and beyond, forgetting that word sizes in the 32-48 bit kind of range have been the norm for most real computing work since the very beginning. Smaller word sizes, first 12-18 bits in minicomputers then 4-16 bits, turned up as temporary abberations when new technologies were developed to build rudimentary computers with much better price performance than their grown up bretheren. As each technology has matured, the word size has tended back toward thirty-something bits. -- don ###### Message-ID: <397AD791.1A2C1265@trailing-edge.com> Date: Sun, 23 Jul 2000 11:31:29 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.2 AlphaServer 1200 5/533 4MB) MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> <8lepf8$gfh$6@bob.news.rcn.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 70 NNTP-Posting-Host: 63.73.218.130 X-Trace: reader2.news.uu.net 964366290 23650 63.73.218.130 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!newsfeedZ.netscum.dQ!netscum.int!hermes.visi.com!news-out.visi.com!uunet!ffx.uu.net!spool0.news.uu.net!reader2.news.uu.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60347 jmfbahciv@aol.com wrote: > > In article <8lehhe$g74$1@flywheel.innovative.iinet.net.au>, > bernie@innovative.iinet.net.au (Bernd Felsche) wrote: > >Alistair@caribsurf.com (Alistair Gale) writes: > > > >>On 22 Jul 2000 20:40:44 +0200, Neil Franklin wrote: > >>>File contents. Files written by user software that contain 32 bit date > >>>fields. The same problem an with 2 digit year numbers. Y2K tidy up > >>>code audits all over again. > > > >>So which database programs use Unix time stamps as opposed to > >>Y/M/D? Is this just a red herring? > > > >Most don't use Unix time() (or timeval() stamp but a derived value. > >This is because it's common to require the storage of dates prior to > >1970 and beyond 2038. > > > >Timestamps in commercial databases are AFAIK in the same internal > >storage format as the data date-time. > > > >That also provides platform independence. > > > >Some mainframe (DB2?) databases store dates as packed digits > >equivalent to CYYMMDD where "C" is a century indicator, usually "0" > >for 1900 to 1999 and "1" for 2000 to 2099. > > I've never seen that before. Has it been in use for a long time? > Or is this just another quick and dirty fix for Y2K. This is very similar to ANSI standard tape labels. These contained, before the Y2K worry, dates (creation and expiration) in 80-byte headers of the form YYDDD where DDD is the so=called "Julian day", 1-365 (or 1-366 on a leap year) and YY was the two-digit year. The the Y2K worries came along. So someone looked at the ANSI tape label standard, and noticed that, by pure random luck, there was a before each YYDDD field. They declared this the "century" value, so that each date field now starts one byte earlier, and is of the form CYYDDD C=0 means years are to be interpreted as 20YY, and C= means years are to be as interpreted as 19YY. The extension through the year 2999 is obvious, though I don't believe it formally became part of the new ANSI date standard. The extension didn't matter to the extended RT-11 date word interpretation, which uses a 7-bit value for the year and a base year of 1972, so, it naturally stops in the year 2099 :-). There are other filesystem limits coming upon us real soon. For instance, DOS-11 tapes and disks use a 16-bit integer value in the range 1-32767 to represent year and Julian day as YYDDD YY is year relative to 1970. As you can see, the sign bit of the 16-bit value will become set in the year 2003. Various DOS-11 derived filesystems interpret this sign bit differently, the most common interpretation being that the file is "write protected" if the sign bit is set. Tim. ###### Path: chonsp.franklin.ch!not-for-mail From: Neil Franklin Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 23 Jul 2000 21:38:39 +0200 Organization: My own Private Self Lines: 47 Message-ID: <6ubszohc1s.fsf@chonsp.franklin.ch> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> NNTP-Posting-Host: chonsp.franklin.ch X-Trace: chonsp.franklin.ch 964381121 1023 10.0.3.2 (23 Jul 2000 19:38:41 GMT) X-Complaints-To: news@chonsp.franklin.ch NNTP-Posting-Date: 23 Jul 2000 19:38:41 GMT X-Newsreader: Gnus v5.7/Emacs 20.4 Xref: chonsp.franklin.ch alt.folklore.computers:60373 bernie@innovative.iinet.net.au (Bernd Felsche) writes: > Alistair@caribsurf.com (Alistair Gale) writes: > > >On 22 Jul 2000 20:40:44 +0200, Neil Franklin wrote: > >>File contents. Files written by user software that contain 32 bit date > >>fields. The same problem an with 2 digit year numbers. Y2K tidy up > >>code audits all over again. > > >So which database programs use Unix time stamps as opposed to > >Y/M/D? Finding out that is exactly what will require code audit II. :-) > Is this just a red herring? Just like Y2K was believed to be by many. :-) One I know: squid (a web cache): root@chonsp 21:32:49 ~# tail -1 /sda4/old_debian_1.2/var/spool/squid/log /var/spool/squid/35/1135 http://www.pbm.com/%7Elindahl/rialto/p-tents-art.html 1070106094 880760504 10174 root@chonsp 21:33:11 ~# Them 1070106094 and 880760504 are time_t values (for expiry time and time the page was fetched). > Most don't use Unix time() (or timeval() stamp but a derived value. > This is because it's common to require the storage of dates prior to > 1970 and beyond 2038. time_t is _signed_ int. That makes 1970+-68.something. Good for 1902-2038. Definitely better than YY which was only good for 1900-2000. People did the later. You can bet on the other. > Timestamps in commercial databases are AFAIK in the same internal > storage format as the data date-time. And in all them other programmers code? -- Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ Nerd, Geek, Hacker, Unix Guru, Sysadmin, Roleplayer, LARPer, Mystic ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Mon, 24 Jul 00 08:13:46 GMT Organization: UltraNet Communications, Inc. Lines: 69 Message-ID: <8lh895$iui$3@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <8lh601$orq$1@flywheel.innovative.iinet.net.au> X-Trace: 4Xk0cTI5eX9Vp8klw6Z4hvkKO8RiRdhUMNuWWztxSFM= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 24 Jul 2000 11:11:33 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-216-96 Xref: chonsp.franklin.ch alt.folklore.computers:60401 In article <8lh601$orq$1@flywheel.innovative.iinet.net.au>, bernie@innovative.iinet.net.au (Bernd Felsche) wrote: >kragen@dnaco.net (Kragen Sitaker) writes: > >>In article <8lepbo$gfh$5@bob.news.rcn.net>, wrote: >>>In article <964298714.421191@shelley.paradise.net.nz>, >>> don@news.daedalus.co.nz (Don Stokes) wrote: >>>>Well, time_t is a *signed* 32 bit int. For file systems, barring >>>>idiocy, there aren't any file dates < 1970, >>> >>> Since I started contributing to social security >>>in 1962, I would certainly hope that there is software that >>>can read file dates before 1970. > >>Unix filesystems don't have files created, last modified, or last >>accessed before 1970, because no Unix filesystems created before 1970 >>still exist. > >Try this on Linux/Unix: :-) > $ TZ=GMT-1 touch -t 195809102030 born > $ ls -l born > -rw------- 1 bernie users 0 Sep 11 1958 born > >Never assume that somebody won't exploit the signed nature of the >timestamp. > >Old filesystems should be OK to mount read-only at any point in the >future. The filesystem type identified by the superblock tells you >about the structure anyway so you can make allowances for antique >media. Since you're insisting that the old formatted media be mounted read-only, how in the world are you going to mark it with a new datum type (assuming that this datum gets defined as part of a new file system design). Don't these files have last accessed date/time? I can't begin to imagine not having that very important piece of information for debugging analysis. > >>>>Cobine it with a 64 bit internal time_t (or >>>>declare time_t to be unsigned as well, which I'm sure would be bug food) >>>>and watch your file system problem get deferred until 2106. >>> >>>It's thinking like this that caused all the Y2K expenditures. > >>I do have to wonder when I see thinking like this; will it be easier or >>harder to fix 100-year-old bugs than 20-year-old bugs? > >My experience has been that a 6-month old bug is just as easy to fix >as a 5-year-old bug. You were very, very lucky, then. We had problems finding sources and build procedures for stuff only 3 years old that had lain fallow. We almost didn't have new RSX-20F releases because of that. > So a 20-year-old bug is probably no easier to >fix than a 100-year-old one. > >Code-fester seems to decline unless new 'talent' or specifications >are injected. It's like the bugs need new food, and being further >away from the core working components seems to slow down the >festering process. /BAH Subtract a hundred and four for e-mail. ###### ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Mon, 24 Jul 00 09:25:40 GMT Organization: UltraNet Communications, Inc. Lines: 85 Message-ID: <8lhcfv$e33$2@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> <8lepf8$gfh$6@bob.news.rcn.net> <397AD791.1A2C1265@trailing-edge.com> X-Trace: b3jMS7s31E5l/pAmP8lcvcNvQcNcIIgpAy4S0YePyMY= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 24 Jul 2000 12:23:27 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!216-164-246-95 Xref: chonsp.franklin.ch alt.folklore.computers:60402 In article <397AD791.1A2C1265@trailing-edge.com>, Tim Shoppa wrote: >jmfbahciv@aol.com wrote: >> >> In article <8lehhe$g74$1@flywheel.innovative.iinet.net.au>, >> bernie@innovative.iinet.net.au (Bernd Felsche) wrote: >> >Alistair@caribsurf.com (Alistair Gale) writes: >> > >> >>On 22 Jul 2000 20:40:44 +0200, Neil Franklin wrote: >> >>>File contents. Files written by user software that contain 32 bit date >> >>>fields. The same problem an with 2 digit year numbers. Y2K tidy up >> >>>code audits all over again. >> > >> >>So which database programs use Unix time stamps as opposed to >> >>Y/M/D? Is this just a red herring? >> > >> >Most don't use Unix time() (or timeval() stamp but a derived value. >> >This is because it's common to require the storage of dates prior to >> >1970 and beyond 2038. >> > >> >Timestamps in commercial databases are AFAIK in the same internal >> >storage format as the data date-time. >> > >> >That also provides platform independence. >> > >> >Some mainframe (DB2?) databases store dates as packed digits >> >equivalent to CYYMMDD where "C" is a century indicator, usually "0" >> >for 1900 to 1999 and "1" for 2000 to 2099. >> >> I've never seen that before. Has it been in use for a long time? >> Or is this just another quick and dirty fix for Y2K. > >This is very similar to ANSI standard tape labels. Ah. I had a 20 foot bamboo pole for ANSI :-). The guy who implemented ANSI labels almost didn't survive. > These contained, >before the Y2K worry, dates (creation and expiration) in 80-byte >headers of the form > > YYDDD > >where DDD is the so=called "Julian day", 1-365 (or 1-366 on a leap >year) and YY was the two-digit year. I hated Julian day...especialy after 29-Feb-nn. > >The the Y2K worries came along. So someone looked at the ANSI tape >label standard, and noticed that, by pure random luck, there was >a before each YYDDD field. They declared this the "century" >value, so that each date field now starts one byte earlier, and >is of the form > > CYYDDD > >C=0 means years are to be interpreted as 20YY, and C= means >years are to be as interpreted as 19YY. > >The extension through the year 2999 is obvious, though I don't >believe it formally became part of the new ANSI date standard. >The extension didn't matter to the extended RT-11 date word >interpretation, which uses a 7-bit value for >the year and a base year of 1972, so, it naturally stops >in the year 2099 :-). > >There are other filesystem limits coming upon us real soon. For >instance, DOS-11 tapes and disks use a 16-bit integer value >in the range 1-32767 to represent year and Julian day as > > YYDDD > >YY is year relative to 1970. As you can see, the sign bit of the >16-bit value will become set in the year 2003. Various DOS-11 >derived filesystems interpret this sign bit differently, the most >common interpretation being that the file is "write protected" >if the sign bit is set. Thank you :-). /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Mon, 24 Jul 00 08:19:37 GMT Organization: UltraNet Communications, Inc. Lines: 20 Message-ID: <8lh8k3$iui$5@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> <6ubszohc1s.fsf@chonsp.franklin.ch> <8lfj1i$404$1@teabag.cbhnet> X-Trace: Uigl6ULoibsbd++pEHjUaBGHoAek71VYFQaFJdL7Kpw= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 24 Jul 2000 11:17:23 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-216-96 Xref: chonsp.franklin.ch alt.folklore.computers:60403 In article <8lfj1i$404$1@teabag.cbhnet>, cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) wrote: >In article <6ubszohc1s.fsf@chonsp.franklin.ch>, > Neil Franklin writes: >> Just like Y2K was believed to be by many. :-) > >It wasn't only believed to be a red-herring but was proven to be >so in most cases. > >Smugly, > >Chris. I wouldn't get too smug since there seems to be people who are writing Y2K bugs today. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Mon, 24 Jul 00 08:18:16 GMT Organization: UltraNet Communications, Inc. Lines: 24 Message-ID: <8lh8hi$iui$4@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> <397B7B0A.F680ED9B@ev1.net> <5DNe5.35033$I7.719214@news-west.usenetserver.com> X-Trace: Uigl6ULoibto0WUXe6wcl1Fz1IOv97NAE32zeiPDsVc= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 24 Jul 2000 11:16:02 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-216-96 Xref: chonsp.franklin.ch alt.folklore.computers:60406 In article <5DNe5.35033$I7.719214@news-west.usenetserver.com>, swaim@nol.net wrote: >Charles Richmond wrote: >> IMHO the Oracle database has an excellent time/date >>storage facility. It >> handles dates back to 2000-something B.C., and also >>way, way into the future. >> This time/date storage should be made the ANSI standard format, and then >> back-filled into VMS/UNIX/whatever...and the time/date problems will be >> put to bed. It would make all this discussion of >>whose date runs out when >> quite irrelevent. > > Oddly enough, all of the Y2K problems I encountered involved Oracle and >dates. By default, Oracle rejects years with 4 digits. (We tried going 4 >digits with one of our databases, but that broke the links to all of the >other databases. Sigh.) > There. There. Another computer rule is "Damned if you do; damned if you don't." Murphy always did love irony. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Mon, 24 Jul 00 08:06:01 GMT Organization: UltraNet Communications, Inc. Lines: 48 Message-ID: <8lh7qj$iui$1@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> X-Trace: Scb7YjMaLCP4v9aqHBVFWK7JO6XGGfJmr97nnPqkYYQ= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 24 Jul 2000 11:03:47 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-216-96 Xref: chonsp.franklin.ch alt.folklore.computers:60407 In article , kragen@dnaco.net (Kragen Sitaker) wrote: >In article <8lepbo$gfh$5@bob.news.rcn.net>, wrote: >>In article <964298714.421191@shelley.paradise.net.nz>, >> don@news.daedalus.co.nz (Don Stokes) wrote: >>>Well, time_t is a *signed* 32 bit int. For file systems, barring >>>idiocy, there aren't any file dates < 1970, >> >> Since I started contributing to social security >>in 1962, I would certainly hope that there is software that >>can read file dates before 1970. > >Unix filesystems don't have files created, last modified, or last >accessed before 1970, because no Unix filesystems created before 1970 >still exist. Sigh! Are you saying that Unix will not let me access a file that was created under another OS before 1970? I don't believe that. > >>>Cobine it with a 64 bit internal time_t (or >>>declare time_t to be unsigned as well, which I'm sure would be bug food) >>>and watch your file system problem get deferred until 2106. >> >>It's thinking like this that caused all the Y2K expenditures. > >I do have to wonder when I see thinking like this; will it be easier or >harder to fix 100-year-old bugs than 20-year-old bugs? Much, much harder. Think about lost sources, non-existant hardware, and the complexity that is going to be around with little bits and pieces of everything located on many machines hidden away in dark corners of this world. Now, think about Java and what that design implies. People are not saving "systems" all in one place these days. I had problems with this ISP (not AOL) last summer. They couldn't even locate the modems that I was dialing into until I suggested that they find the guy who has the key to the closet....that got their attention. /BAH /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Mon, 24 Jul 00 08:08:42 GMT Organization: UltraNet Communications, Inc. Lines: 44 Message-ID: <8lh7vk$iui$2@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <964358954.300056@shelley.paradise.net.nz> X-Trace: TxQXMXzqfdQa/mtV875f8c5IkQNfLVmlircs8L9OAOg= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 24 Jul 2000 11:06:28 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!feed2.onemain.com!feed1.onemain.com!europa.netcrusader.net!207.172.3.37!feed1.news.rcn.net!rcn!207-172-216-96 Xref: chonsp.franklin.ch alt.folklore.computers:60409 In article <964358954.300056@shelley.paradise.net.nz>, don@news.daedalus.co.nz (Don Stokes) wrote: >In article <8lepbo$gfh$5@bob.news.rcn.net>, wrote: >> don@news.daedalus.co.nz (Don Stokes) wrote: >>>Well, time_t is a *signed* 32 bit int. For file systems, barring >>>idiocy, there aren't any file dates < 1970, >> >> Since I started contributing to social security >>in 1962, I would certainly hope that there is software that >>can read file dates before 1970. > >I'm talking filesystems only. So am I. Now think about it. > Eg, if I mount a UFS or SysV filesystem, >I can treat the 32 bit dates as unsigned instead of signed. I'd read >the numbers into a 64 bit int for manipulation, and time_t would be >64 bits. The translation to/from 32 bits only happens when dealling >with these old filesystems, and only for file dates. Hopefully, these >things will all be dead by 2106. Hopefully? HOPEFULLY? [very excited emoticon here] That word only belongs in fairy tales. > >Mind you, there seems to be an assumption by those exposed only to >personal computers that there's a natural progression 4-8-16-32 bits in >word sizes which will continue through 64 and beyond, forgetting that >word sizes in the 32-48 bit kind of range have been the norm for most >real computing work since the very beginning. > >Smaller word sizes, first 12-18 bits in minicomputers then 4-16 bits, >turned up as temporary abberations when new technologies were developed >to build rudimentary computers with much better price performance than >their grown up bretheren. As each technology has matured, the word >size has tended back toward thirty-something bits. Think about data flow, not hardware and you'll begin to understand what I'm trying to say. /BAH Subtract a hundred and four for e-mail. ###### From: Alistair@caribsurf.com (Alistair Gale) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 23 Jul 2000 19:03:26 GMT Organization: http://extra.newsguy.com Lines: 21 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> Reply-To: alistair@caribsurf.com NNTP-Posting-Host: p-892.newsdawg.com User-Agent: slrn/0.9.5.7 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeed.berkeley.edu!ucberkeley!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!Alistair Xref: chonsp.franklin.ch alt.folklore.computers:60395 >Some mainframe (DB2?) databases store dates as packed digits >equivalent to CYYMMDD where "C" is a century indicator, usually "0" >for 1900 to 1999 and "1" for 2000 to 2099. >-- >Bernd Felsche - Innovative Reckoning >Perth, Western Australia Hola Bruce^WBerndt: My point was that even though the Unix OS has *file* timestamps that run out in 2038, this has nothing to do with dates embedded in applications data on those systems. So there shouldn't be any "Y-2038" problem as long as you port your database to a 64 bit Unix within the next 37 years. What I'm trying to say is that Y2K or Y2038 are not OS issues, but rather application issues (BIG HINT TO PREVIOUS POSTER) __ alistair BA: "Ohh, I want to bed down with a dozen concubines." Baldrick:" Wouldn't that be all prickly?" ###### From: cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 23 Jul 2000 20:02:58 GMT Organization: teabag Message-ID: <8lfj1i$404$1@teabag.cbhnet> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> <6ubszohc1s.fsf@chonsp.franklin.ch> NNTP-Posting-Host: teabag.cbhnet X-NNTP-Posting-Host: teabag.demon.co.uk:193.237.4.110 X-Trace: news.demon.co.uk 964383621 nnrp-08:2344 NO-IDENT teabag.demon.co.uk:193.237.4.110 X-Complaints-To: abuse@demon.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Date: 23 Jul 2000 20:02:58 GMT X-Newsreader: knews 1.0b.0 Lines: 10 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!rz.uni-karlsruhe.de!schlund.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!colt.net!dispose.news.demon.net!demon!news.demon.co.uk!demon!teabag.demon.co.uk!teabag.cbhnet!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60380 In article <6ubszohc1s.fsf@chonsp.franklin.ch>, Neil Franklin writes: > Just like Y2K was believed to be by many. :-) It wasn't only believed to be a red-herring but was proven to be so in most cases. Smugly, Chris. ###### From: Charles Richmond Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Sun, 23 Jul 2000 15:53:16 -0700 Organization: Cannine Computer Center Lines: 22 Message-ID: <397B775D.3FAA20B9@ev1.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8kv4bv$ju4$1@news.kersur.net> <3973B271.763C405@prescienttech.com> <3974935b$0$4933@reader3> <8l46n4$j6o$1@rena.mat.uc.pt> Reply-To: richmond@ev1.net X-Complaints-To: newsabuse@supernews.com X-Mailer: Mozilla 4.7C-CCK-MCD {C-UDP; EBM-APPLE} (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-mue1.dfn.de!news-nue1.dfn.de!uni-erlangen.de!newsfeeds.belnet.be!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!telocity-west!TELOCITY!sn-xit-01!supernews.com!sn-inject-01!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60445 Rui Pedro Mendes Salgueiro wrote: > > Frits wrote: > > I've always wondered, what kind of system they used in the Shuttle. > > I happen to have here a copie of an article about that: > > "Architecture of the Space Shuttle Primary Avionics Software System" > Gene D. Carlow > > Comunications of the ACM, September 1984, Volume 27, Number 9, pages > 926-936. This is a reprint of an article published in 1980-1981. > This article is talking about the computers that run the cockpit displays and such...I think what others are posting about is the flight computers. The avionics computers are IBM SP-0 computers, I believe... -- +-------------------------------------------------------------+ | Charles and Francis Richmond | +-------------------------------------------------------------+ ###### From: Charles Richmond Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Sun, 23 Jul 2000 16:08:57 -0700 Organization: Cannine Computer Center Lines: 30 Message-ID: <397B7B0A.F680ED9B@ev1.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> Reply-To: richmond@ev1.net X-Complaints-To: newsabuse@supernews.com X-Mailer: Mozilla 4.7C-CCK-MCD {C-UDP; EBM-APPLE} (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.cwix.com!news.maxwell.syr.edu!telocity-west!TELOCITY!sn-xit-01!supernews.com!sn-inject-01!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60447 Alistair Gale wrote: > > >Some mainframe (DB2?) databases store dates as packed digits > >equivalent to CYYMMDD where "C" is a century indicator, usually "0" > >for 1900 to 1999 and "1" for 2000 to 2099. > >-- > >Bernd Felsche - Innovative Reckoning > >Perth, Western Australia > > Hola Bruce^WBerndt: > My point was that even though the Unix OS has *file* timestamps that > run out in 2038, this has nothing to do with dates embedded in > applications data on those systems. So there shouldn't > be any "Y-2038" problem as long as you port your database to a > 64 bit Unix within the next 37 years. > > What I'm trying to say is that Y2K or Y2038 are not OS issues, but rather > application issues (BIG HINT TO PREVIOUS POSTER) > IMHO the Oracle database has an excellent time/date storage facility. It handles dates back to 2000-something B.C., and also way, way into the future. This time/date storage should be made the ANSI standard format, and then back-filled into VMS/UNIX/whatever...and the time/date problems will be put to bed. It would make all this discussion of whose date runs out when quite irrelevent. -- +-------------------------------------------------------------+ | Charles and Francis Richmond | +-------------------------------------------------------------+ ###### From: David Scheidt Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 24 Jul 2000 00:08:20 GMT Organization: EnterAct Corp. Lines: 16 Message-ID: <8lg1dk$2csq$1@news.enteract.com> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8kv4bv$ju4$1@news.kersur.net> <3973B271.763C405@prescienttech.com> <3974935b$0$4933@reader3> <39777916$1@news.ucsc.edu> NNTP-Posting-Host: shell-3.enteract.com X-Trace: news.enteract.com 964397300 78746 207.229.143.42 (24 Jul 2000 00:08:20 GMT) X-Complaints-To: abuse@enteract.com NNTP-Posting-Date: 24 Jul 2000 00:08:20 GMT User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (FreeBSD/3.5-STABLE (i386)) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-mue1.dfn.de!news.augsburg.net!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.enteract.com!news.enteract.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60417 Eugene Miya wrote: : Surely there's a #5 ESS sitting somewhere running longer than other machines. : Maybe a Tandem some place. Depends on how you are going to calculate uptime. If you mean the whole system, maybe. If you mean just the the 3B21, then no. It would have had to have been upgraded to put a Y2K certified generic on it, which likely would involve a UNIX upgrade. -- David Scheidt The presumption of the flamers is, I assume, that folks barging in to AFU with tired old stories and off-topic drivel have already ignored the polite and subtle clues, and require a thwack with the clue-by-four. -- Andrew Reid ###### From: swaim@nol.net Subject: Re: Longest running computer Newsgroups: alt.folklore.computers References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> <397B7B0A.F680ED9B@ev1.net> User-Agent: tin/pre-1.4-19990413 ("Endemoniada") (UNIX) (FreeBSD/3.2-RELEASE (i386)) Lines: 16 Message-ID: <5DNe5.35033$I7.719214@news-west.usenetserver.com> X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Sun, 23 Jul 2000 22:13:21 EDT Organization: WebUseNet Corp http://www.usenetserver.com - Home of the fastest NNTP servers on the Net. Date: Mon, 24 Jul 2000 02:13:21 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-mue1.dfn.de!news.augsburg.net!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!portc01.blue.aol.com!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!news-west.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60452 Charles Richmond wrote: > IMHO the Oracle database has an excellent time/date storage facility. It > handles dates back to 2000-something B.C., and also way, way into the future. > This time/date storage should be made the ANSI standard format, and then > back-filled into VMS/UNIX/whatever...and the time/date problems will be > put to bed. It would make all this discussion of whose date runs out when > quite irrelevent. Oddly enough, all of the Y2K problems I encountered involved Oracle and dates. By default, Oracle rejects years with 4 digits. (We tried going 4 digits with one of our databases, but that broke the links to all of the other databases. Sigh.) -- Mike Swaim, Avatar of Chaos: Disclaimer:I sometimes lie. Home: swaim at nol * net Quote: "Boingie"^4 Y,W&D ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 27 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Sun, 23 Jul 2000 23:03:43 EDT Date: Mon, 24 Jul 2000 03:03:43 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!news-west.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60376 In article <8lepbo$gfh$5@bob.news.rcn.net>, wrote: >In article <964298714.421191@shelley.paradise.net.nz>, > don@news.daedalus.co.nz (Don Stokes) wrote: >>Well, time_t is a *signed* 32 bit int. For file systems, barring >>idiocy, there aren't any file dates < 1970, > > Since I started contributing to social security >in 1962, I would certainly hope that there is software that >can read file dates before 1970. Unix filesystems don't have files created, last modified, or last accessed before 1970, because no Unix filesystems created before 1970 still exist. >>Cobine it with a 64 bit internal time_t (or >>declare time_t to be unsigned as well, which I'm sure would be bug food) >>and watch your file system problem get deferred until 2106. > >It's thinking like this that caused all the Y2K expenditures. I do have to wonder when I see thinking like this; will it be easier or harder to fix 100-year-old bugs than 20-year-old bugs? -- Kragen Sitaker Perilous to all of us are the devices of an art deeper than we ourselves possess. -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"] ###### From: bernie@innovative.iinet.net.au (Bernd Felsche) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 24 Jul 2000 18:20:45 +0800 Organization: Innovative Reckoning Message-ID: <8lh59t$oq1$1@flywheel.innovative.iinet.net.au> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> <8lepf8$gfh$6@bob.news.rcn.net> X-Newsreader: NN version 6.5.1 (NOV) Lines: 36 NNTP-Posting-Host: innovative.iinet.net.au X-Trace: news.iinet.net.au 964434951 15551 203.59.144.24 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!news.mel.connect.com.au!news.per.connect.com.au!newsfeed.iinet.net.au!news.iinet.net.au!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60379 jmfbahciv@aol.com writes: >In article <8lehhe$g74$1@flywheel.innovative.iinet.net.au>, > bernie@innovative.iinet.net.au (Bernd Felsche) wrote: >>Alistair@caribsurf.com (Alistair Gale) writes: >> >>>On 22 Jul 2000 20:40:44 +0200, Neil Franklin wrote: >>>>File contents. Files written by user software that contain 32 bit date >>>>fields. The same problem an with 2 digit year numbers. Y2K tidy up >>>>code audits all over again. >> >>>So which database programs use Unix time stamps as opposed to >>>Y/M/D? Is this just a red herring? >> >>Most don't use Unix time() (or timeval() stamp but a derived value. >>This is because it's common to require the storage of dates prior to >>1970 and beyond 2038. >> >>Timestamps in commercial databases are AFAIK in the same internal >>storage format as the data date-time. >> >>That also provides platform independence. >> >>Some mainframe (DB2?) databases store dates as packed digits >>equivalent to CYYMMDD where "C" is a century indicator, usually "0" >>for 1900 to 1999 and "1" for 2000 to 2099. >I've never seen that before. Has it been in use for a long time? Could be CYYDDD - I always check the manual before touching the keboard. And no; relatively new; been around maybe 30-something years >Or is this just another quick and dirty fix for Y2K. -- Bernd Felsche - Innovative Reckoning Perth, Western Australia ###### From: bernie@innovative.iinet.net.au (Bernd Felsche) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 24 Jul 2000 18:32:33 +0800 Organization: Innovative Reckoning Message-ID: <8lh601$orq$1@flywheel.innovative.iinet.net.au> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> X-Newsreader: NN version 6.5.1 (NOV) Lines: 49 NNTP-Posting-Host: innovative.iinet.net.au X-Trace: news.iinet.net.au 964434952 15551 203.59.144.24 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!newspump.monmouth.com!newspeer.monmouth.com!intgwpad.nntp.telstra.net!news1.optus.net.au!optus!news.mel.connect.com.au!news.per.connect.com.au!newsfeed.iinet.net.au!news.iinet.net.au!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60378 kragen@dnaco.net (Kragen Sitaker) writes: >In article <8lepbo$gfh$5@bob.news.rcn.net>, wrote: >>In article <964298714.421191@shelley.paradise.net.nz>, >> don@news.daedalus.co.nz (Don Stokes) wrote: >>>Well, time_t is a *signed* 32 bit int. For file systems, barring >>>idiocy, there aren't any file dates < 1970, >> >> Since I started contributing to social security >>in 1962, I would certainly hope that there is software that >>can read file dates before 1970. >Unix filesystems don't have files created, last modified, or last >accessed before 1970, because no Unix filesystems created before 1970 >still exist. Try this on Linux/Unix: :-) $ TZ=GMT-1 touch -t 195809102030 born $ ls -l born -rw------- 1 bernie users 0 Sep 11 1958 born Never assume that somebody won't exploit the signed nature of the timestamp. Old filesystems should be OK to mount read-only at any point in the future. The filesystem type identified by the superblock tells you about the structure anyway so you can make allowances for antique media. >>>Cobine it with a 64 bit internal time_t (or >>>declare time_t to be unsigned as well, which I'm sure would be bug food) >>>and watch your file system problem get deferred until 2106. >> >>It's thinking like this that caused all the Y2K expenditures. >I do have to wonder when I see thinking like this; will it be easier or >harder to fix 100-year-old bugs than 20-year-old bugs? My experience has been that a 6-month old bug is just as easy to fix as a 5-year-old bug. So a 20-year-old bug is probably no easier to fix than a 100-year-old one. Code-fester seems to decline unless new 'talent' or specifications are injected. It's like the bugs need new food, and being further away from the core working components seems to slow down the festering process. -- Bernd Felsche - Innovative Reckoning Perth, Western Australia ###### Message-ID: <397C37FB.8DF69EEC@netinsight.se> From: Johnny Billquist Organization: Netinsight AB X-Mailer: Mozilla 4.5 [en] (X11; I; Linux 2.2.14 i686) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> <964273344.185144@shelley.paradise.net.nz> <3979B287.3EA2B030@trailing-edge.com> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Lines: 39 Date: Mon, 24 Jul 2000 12:35:07 GMT NNTP-Posting-Host: 194.16.221.33 X-Complaints-To: abuse@telia.com X-Trace: newsc.telia.net 964442107 194.16.221.33 (Mon, 24 Jul 2000 14:35:07 MET DST) NNTP-Posting-Date: Mon, 24 Jul 2000 14:35:07 MET DST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!uninett.no!newsfeed1.enitel.no!masternews.telia.net!newsc.telia.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60400 Tim Shoppa wrote: > > Don Stokes wrote: > > > > In article <8lc76n$5co$3@bob.news.rcn.net>, wrote: > > >Why spend time when a NVMS$;-3DIRSX$EX$$ will do? Answering SPRs > > >were not a favorite pasttime for most people. Actually, I can't > > >think of a single person who liked to answer them. > > > > Yeah, but the SPR response completely answers the RSX Y2K problem that > > is the actual subject of the SPR (albeit far from satisfactorily, > > basically, "bother us again in 1999") > > They leave out another big problem: ODS-1 filesystem Y2K support. > This required more work than all the other basic Y2K issues in the > kernel of RSX. (And the resulting fix, while functional, is hardly > pretty.) Ah, yes. It is rather sad that they decided on such a stupid format for dates in ODS-1. :-/ The RSX kernel itself actually didn't need any Y2K modifications. F11ACP and assorted utilities were the things affected, I believe. > This why I laugh when Unix weenies tell me that to fix the Y2038 problem > all they have to do is change to a 64 bit time_t and recompile. > It's so easy, but so very wrong, to ignore back-compatibility with > existing data structures! Definitely. The problematic thing is that in the Unix case, backward compatibility will not be that easy... Johnny -- Johnny Billquist | johnny.billquist@netinsight.net Net Insight AB | phone: +46 8 685 04 88 Västberga Allé 9 | fax: +46 8 685 04 20 Box 42093 | SE-126 30 STOCKHOLM, Sweden | http://www.netinsight.net ###### From: viro@weyl.math.psu.edu (Alexander Viro) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 24 Jul 2000 08:35:50 -0400 Organization: -ENOENT Lines: 47 Message-ID: <8lhd76$lmq@weyl.math.psu.edu> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8lh601$orq$1@flywheel.innovative.iinet.net.au> <8lh895$iui$3@bob.news.rcn.net> NNTP-Posting-Host: weyl.math.psu.edu Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!news.cis.ohio-state.edu!news.ems.psu.edu!news3.cac.psu.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60405 In article <8lh895$iui$3@bob.news.rcn.net>, wrote: >>Old filesystems should be OK to mount read-only at any point in the >>future. The filesystem type identified by the superblock tells you >>about the structure anyway so you can make allowances for antique >>media. > >Since you're insisting that the old formatted media be mounted >read-only, how in the world are you going to mark it with a new >datum type (assuming that this datum gets defined as part of >a new file system design). Don't these files have last >accessed date/time? I can't begin to imagine not having that >very important piece of information for debugging analysis. Huh? Assuming Linux - in fs/foo/super.c::foo_read_super() ... disk_sb = (struct foo_super_block*)bh->b_data; if (disk_sb->foo_magic == OLD_FOO_MAGIC) { printk("Old filesystem, refusing to mount r/w" "(running fsck.foo is recommended for upgrading it)"); sb->s_flags |= MS_RDONLY; sb->u.foo_sb.type = FOOFS_OLD; } else if (disk_sb->foo_magic != NEW_FOO_MAGIC) { if (!silent) printk("Can't find foofs on device %s\n", bdevname(sb->s_dev); goto failed_mount; } else sb->u.foo_sb.type = FOOFS_NEW; ... In other words, attempt to mount the old filesystem will a) be detected b) will be forcibly made read-only c) you will get a good advice in logs d) manpage of fsck.foo is going to give you details on transform-to-new-format invocation. fsck.foo can open the device read/write even though fs is mouted r/o - that how it normally works. Do it and get the atime back as soon as fs is converted and may be mounted r/w. Pretty similar trick will go for 4.4BSD and its kin, indeed. -- "You're one of those condescending Unix computer users!" "Here's a nickel, kid. Get yourself a better computer" - Dilbert. ###### Message-ID: <397C3B7D.BC1AF21@netinsight.se> From: Johnny Billquist Organization: Netinsight AB X-Mailer: Mozilla 4.5 [en] (X11; I; Linux 2.2.14 i686) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3972A82B.17177CCC@trailing-dge.com> <3976D79B.D2943F34@netsight.se> <6ur98ohbak.fsf@chonsp.franklin.ch> <3978151E.493527D5@netinsight.se> <6ur98mmafz.fsf@chonsp.franklin.ch> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Lines: 55 Date: Mon, 24 Jul 2000 12:50:06 GMT NNTP-Posting-Host: 194.16.221.33 X-Complaints-To: abuse@telia.com X-Trace: newsc.telia.net 964443006 194.16.221.33 (Mon, 24 Jul 2000 14:50:06 MET DST) NNTP-Posting-Date: Mon, 24 Jul 2000 14:50:06 MET DST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!news1.ebone.net!news.ebone.net!newsfeed.sunet.se!news01.sunet.se!news-sto.telia.net!masternews.telia.net!newsc.telia.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60390 Neil Franklin wrote: > > Johnny Billquist writes: > > > Neil Franklin wrote: > > > > > Of course not. My complaint here was not with the Unix programmers, > > but with the VAXCRTL implementors. > > OK, I missunderstood you because of: > > > > > compiled on VMS. (Once again, Unix rears its ugly head... :-) > > So I thought you were attributiong it to Unix (which is to blame for > 2038, but not for 10000 days). Understandable. I could probably have expressed myself better. What I meant by that remark was jsut that we seem to be hitting problems with code that in any way is associated with Unix. The fault was the VAXCRTL implementors however, in the 10,000-days case. > > > Unix (and C) count in _seconds_ since 1970, which is why they go wrong > > > on 2^31 seconds = in 2038. There is nowhere in Unix (or C) anythong > > > about 10'000 days, that is just an VAXCRTL nonsense. > > > > Yes. But it is rather funny that we have so short limits in *both* cases, > > 68 years must have looked like eternity in 1970. :-) Perhaps. But some other OSes from that same time didn't go for such short timeframes. (But then again, those other OSes were OSes, and not glorified video games... :-) (Hmmm, there is another OS-making company which that epitet can be used for nowadays... :-) > > > Room for 2^64 ticks of 100ns each. and negative values > > are delta times (unless my memory fails me). > > Overloading the top bit for deltas. A hidden flag. Hmmm. Hidden... Nah. :-) > Well that halves the time range then. Yup. But I guess that going in to the 310:th century is enough. Johnny -- Johnny Billquist | johnny.billquist@netinsight.net Net Insight AB | phone: +46 8 685 04 88 Västberga Allé 9 | fax: +46 8 685 04 20 Box 42093 | SE-126 30 STOCKHOLM, Sweden | http://www.netinsight.net ###### From: "P Linnane" Newsgroups: alt.folklore.computers References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> <8lepf8$gfh$6@bob.news.rcn.net> <8lh59t$oq1$1@flywheel.innovative.iinet.net.au> Subject: Re: Longest running computer Lines: 45 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.6600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Organization: Ye 'Ol Disorganized NNTPCache groupie Message-ID: <964459352.967749@server16.cable.com> Cache-Post-Path: server16.cable.com!unknown@vp1-18.xcessible.com X-Cache: nntpcache 2.4.0b2 (see http://www.nntpcache.org/) Date: Mon, 24 Jul 2000 13:18:15 -0400 NNTP-Posting-Host: 204.92.64.17 X-Trace: nnrp1.uunet.ca 964459049 204.92.64.17 (Mon, 24 Jul 2000 13:17:29 EDT) NNTP-Posting-Date: Mon, 24 Jul 2000 13:17:29 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-mue1.dfn.de!news-was.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!east1.newsfeed.sprint-canada.net!news.uunet.ca!nnrp1.uunet.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60408 "Bernd Felsche" wrote in message news:8lh59t$oq1$1@flywheel.innovative.iinet.net.au... > jmfbahciv@aol.com writes: > > >In article <8lehhe$g74$1@flywheel.innovative.iinet.net.au>, > > bernie@innovative.iinet.net.au (Bernd Felsche) wrote: > >>Alistair@caribsurf.com (Alistair Gale) writes: > >> > >>>On 22 Jul 2000 20:40:44 +0200, Neil Franklin wrote: > >>>>File contents. Files written by user software that contain 32 bit date > >>>>fields. The same problem an with 2 digit year numbers. Y2K tidy up > >>>>code audits all over again. > >> > >>>So which database programs use Unix time stamps as opposed to > >>>Y/M/D? Is this just a red herring? > >> > >>Most don't use Unix time() (or timeval() stamp but a derived value. > >>This is because it's common to require the storage of dates prior to > >>1970 and beyond 2038. > >> > >>Timestamps in commercial databases are AFAIK in the same internal > >>storage format as the data date-time. > >> > >>That also provides platform independence. > >> > >>Some mainframe (DB2?) databases store dates as packed digits > >>equivalent to CYYMMDD where "C" is a century indicator, usually "0" > >>for 1900 to 1999 and "1" for 2000 to 2099. > > >I've never seen that before. Has it been in use for a long time? > > Could be CYYDDD - I always check the manual before touching the nah, then you'd need another packed byte, might as well CCYYMMDD - Rick > keboard. And no; relatively new; been around maybe 30-something years > > >Or is this just another quick and dirty fix for Y2K. > -- > Bernd Felsche - Innovative Reckoning > Perth, Western Australia ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Tue, 25 Jul 00 09:12:36 GMT Organization: UltraNet Communications, Inc. Lines: 32 Message-ID: <8lk03p$o2c$6@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> <397D7A22.A2108963@netinsight.se> X-Trace: CA/Z/EIJmXLz/vywTOwV+w0cof/6pigwRn6U5YrjnXg= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 25 Jul 2000 12:10:33 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!news.rhrz.uni-bonn.de!RRZ.Uni-Koeln.DE!news.netcologne.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-245-120 Xref: chonsp.franklin.ch alt.folklore.computers:60488 In article <397D7A22.A2108963@netinsight.se>, Johnny Billquist wrote: >Jim Thomas wrote: >> >> >>>>> "/BAH" == jmfbahciv writes: >> >> /BAH> Why spend time when a NVMS$;-3DIRSX$EX$$ will do? >> >> Ummm, I think you meant EX$$ :-) > >That won't work either. The EX$$ will not be executed, since >the macro is aborted once the search fails. I don't remember >the modifier to return a code instead of aborting on fail, >but I'm sure anyone here can correct the code, given a few >minutes and a manual. :-) Once a debugger, always a debugger. However, the command works as written. If there's an error, then the editor types a ^C leaving all the old modified, access, and created date/time stamps unchanged. One of the unsung techniques of a rule "If it ain't broke, don't change it." One of the Laws of Computing is that, for every time a source or document is opened for an edit, a typo occurs in a piece of the file that is remote from that edit....especially if the file is to be submitted to the SDC for First Customer Ship. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Tue, 25 Jul 00 09:16:07 GMT Organization: UltraNet Communications, Inc. Lines: 28 Message-ID: <8lk0ac$o2c$7@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> X-Trace: H43QAOf22fMzDsKrCzf9QeHLt3iBcpdafRs8gcNUklI= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 25 Jul 2000 12:14:04 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!rz.uni-karlsruhe.de!schlund.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-245-120 Xref: chonsp.franklin.ch alt.folklore.computers:60490 In article , djb0x1312fca@scream.org (Dan) wrote: >Neil Franklin wrote: > >> File contents. Files written by user software that contain 32 bit date >> fields. The same problem an with 2 digit year numbers. Y2K tidy up >> code audits all over again. > >Ah, but... As long as the start of the epoch is 1/1/1970, wouldn't a >system with a 64-bit time_t still read a 32-bit "legacy" date, count >the appropriate number of seconds, and render the correct date? The >numbers themselves are not going to abruptly stop being valid, right? How are you going to know that number you have came from a 32-bit field rather than a 64-bit field _without setting some kind of flag that means "new format"_. And how many "new formats" can one have? > >Presuming, of course, that the system was coded in such a way that it >knew to Do The Right Thing if it encountered a 32-bit date. And that's a huge presumption when the coder is only working and debugging on a newer system and has never seen an old system. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Tue, 25 Jul 00 09:07:05 GMT Organization: UltraNet Communications, Inc. Lines: 24 Message-ID: <8ljvpf$o2c$5@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> X-Trace: ZY7Lx+YlNFM43uT6WcdYkymopAz2BPp6Ft+TM9I4ngk= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 25 Jul 2000 12:05:03 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-koe1.dfn.de!news-was.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-245-120 Xref: chonsp.franklin.ch alt.folklore.computers:60497 In article , Jim Thomas wrote: >>>>>> "/BAH" == jmfbahciv writes: > > /BAH> Why spend time when a NVMS$;-3DIRSX$EX$$ will do? > >Ummm, I think you meant EX$$ :-) > > /BAH> Answering SPRs > /BAH> were not a favorite pasttime for most people. Actually, I can't > /BAH> think of a single person who liked to answer them. > >Stand up and be counted. I learned a lot doing so :-) Oh, I didn't mean it that way. One always learned a lot from working on an SPR. I literally meant _answering_ the SPR. That is, writing an answer in English ASCII that would pass the reviewer and not get us into any lawsuit corners. /BAH Subtract a hundred and four for e-mail. ###### From: Jim Thomas Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 24 Jul 2000 14:12:10 -1000 Organization: Canada France Hawai`i Telescope Lines: 13 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> NNTP-Posting-Host: atlas.cfht.hawaii.edu X-Trace: news.hawaii.edu 964483930 6119 128.171.80.135 (25 Jul 2000 00:12:10 GMT) X-Complaints-To: usenet@hawaii.edu NNTP-Posting-Date: 25 Jul 2000 00:12:10 GMT X-Newsreader: Gnus v5.7/Emacs 20.6 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!newsfeed.axxsys.net!hammer.uoregon.edu!news.hawaii.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60530 >>>>> "/BAH" == jmfbahciv writes: /BAH> Why spend time when a NVMS$;-3DIRSX$EX$$ will do? Ummm, I think you meant EX$$ :-) /BAH> Answering SPRs /BAH> were not a favorite pasttime for most people. Actually, I can't /BAH> think of a single person who liked to answer them. Stand up and be counted. I learned a lot doing so :-) Nothead ###### From: "P Linnane" Newsgroups: alt.folklore.computers References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> <8lepf8$gfh$6@bob.news.rcn.net> <8lh59t$oq1$1@flywheel.innovative.iinet.net.au> <964459352.967749@server16.cable.com> Subject: Re: Longest running computer Lines: 8 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.6600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Organization: Ye 'Ol Disorganized NNTPCache groupie Message-ID: <964501250.780388@server16.cable.com> Cache-Post-Path: server16.cable.com!unknown@vp1-34.xcessible.com X-Cache: nntpcache 2.4.0b2 (see http://www.nntpcache.org/) Date: Tue, 25 Jul 2000 00:56:31 -0400 NNTP-Posting-Host: 204.92.64.17 X-Trace: nnrp1.uunet.ca 964500944 204.92.64.17 (Tue, 25 Jul 2000 00:55:44 EDT) NNTP-Posting-Date: Tue, 25 Jul 2000 00:55:44 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-koe1.dfn.de!news-was.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.cwix.com!prairie.attcanada.net!newsfeed.attcanada.net!142.77.1.188!news.uunet.ca!nnrp1.uunet.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60495 > > > > Could be CYYDDD - I always check the manual before touching the > > nah, then you'd need another packed byte, might as well CCYYMMDD - Rick > duh, change that to CCYYDDD - Rick ###### From: djb0x1312fca@scream.org (Dan) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 25 Jul 2000 09:24:49 GMT Lines: 20 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> Reply-To: djb0x1312fca@scream.org NNTP-Posting-Host: v2000.scream.org X-Trace: mochi.lava.net 964517089 13721 198.151.213.3 (25 Jul 2000 09:24:49 GMT) X-Complaints-To: abuse@lava.net NNTP-Posting-Date: 25 Jul 2000 09:24:49 GMT User-Agent: slrn/0.9.6.2 (Linux) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!feeder.via.net!news.he.net!news.lava.net!djb0x1312fca Xref: chonsp.franklin.ch alt.folklore.computers:60546 Neil Franklin wrote: > File contents. Files written by user software that contain 32 bit date > fields. The same problem an with 2 digit year numbers. Y2K tidy up > code audits all over again. Ah, but... As long as the start of the epoch is 1/1/1970, wouldn't a system with a 64-bit time_t still read a 32-bit "legacy" date, count the appropriate number of seconds, and render the correct date? The numbers themselves are not going to abruptly stop being valid, right? Presuming, of course, that the system was coded in such a way that it knew to Do The Right Thing if it encountered a 32-bit date. -Dan -- Dan Birchall - Palolo Valley, Honolulu HI - http://dan.scream.org Get paid for your opinions - http://epinions.scream.org/join.html My address expires - take out the hex stamp if your reply bounces ###### Message-ID: <397D7A22.A2108963@netinsight.se> From: Johnny Billquist Organization: Netinsight AB X-Mailer: Mozilla 4.5 [en] (X11; I; Linux 2.2.14 i686) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Lines: 22 Date: Tue, 25 Jul 2000 11:29:38 GMT NNTP-Posting-Host: 194.16.221.33 X-Complaints-To: abuse@telia.com X-Trace: newsc.telia.net 964524578 194.16.221.33 (Tue, 25 Jul 2000 13:29:38 MET DST) NNTP-Posting-Date: Tue, 25 Jul 2000 13:29:38 MET DST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!uninett.no!newsfeed1.enitel.no!masternews.telia.net!newsc.telia.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60481 Jim Thomas wrote: > > >>>>> "/BAH" == jmfbahciv writes: > > /BAH> Why spend time when a NVMS$;-3DIRSX$EX$$ will do? > > Ummm, I think you meant EX$$ :-) That won't work either. The EX$$ will not be executed, since the macro is aborted once the search fails. I don't remember the modifier to return a code instead of aborting on fail, but I'm sure anyone here can correct the code, given a few minutes and a manual. :-) Johnny -- Johnny Billquist | johnny.billquist@netinsight.net Net Insight AB | phone: +46 8 685 04 88 Västberga Allé 9 | fax: +46 8 685 04 20 Box 42093 | SE-126 30 STOCKHOLM, Sweden | http://www.netinsight.net ###### From: bernie@innovative.iinet.net.au (Bernd Felsche) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 25 Jul 2000 20:36:53 +0800 Organization: Innovative Reckoning Distribution: local Message-ID: <8lk1l5$uah$1@flywheel.innovative.iinet.net.au> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <8lh601$orq$1@flywheel.innovative.iinet.net.au> <8lh895$iui$3@bob.news.rcn.net> X-Newsreader: NN version 6.5.1 (NOV) Lines: 70 NNTP-Posting-Host: innovative.iinet.net.au X-Trace: news.iinet.net.au 964533698 15544 203.59.144.24 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!diablo.theplanet.net!newsfeed.icl.net!skynet.be!newsfeed.iinet.net.au!news.iinet.net.au!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60474 jmfbahciv@aol.com writes: >In article <8lh601$orq$1@flywheel.innovative.iinet.net.au>, > bernie@innovative.iinet.net.au (Bernd Felsche) wrote: >>kragen@dnaco.net (Kragen Sitaker) writes: >>>In article <8lepbo$gfh$5@bob.news.rcn.net>, wrote: >>>>In article <964298714.421191@shelley.paradise.net.nz>, >>>> don@news.daedalus.co.nz (Don Stokes) wrote: >>>>>Well, time_t is a *signed* 32 bit int. For file systems, barring >>>>>idiocy, there aren't any file dates < 1970, >>>> >>>> Since I started contributing to social security >>>>in 1962, I would certainly hope that there is software that >>>>can read file dates before 1970. >> >>>Unix filesystems don't have files created, last modified, or last >>>accessed before 1970, because no Unix filesystems created before 1970 >>>still exist. >> >>Try this on Linux/Unix: :-) >> $ TZ=GMT-1 touch -t 195809102030 born >> $ ls -l born >> -rw------- 1 bernie users 0 Sep 11 1958 born >> >>Never assume that somebody won't exploit the signed nature of the >>timestamp. >>Old filesystems should be OK to mount read-only at any point in the >>future. The filesystem type identified by the superblock tells you >>about the structure anyway so you can make allowances for antique >>media. >Since you're insisting that the old formatted media be mounted >read-only, how in the world are you going to mark it with a new >datum type (assuming that this datum gets defined as part of >a new file system design). Don't these files have last >accessed date/time? I can't begin to imagine not having that >very important piece of information for debugging analysis. 1. you don't have those timestamps on CD-ROMs 2. you can copy/convert the filesystem to a newer type; hint: mainly the inodes need restructuring. (Somebody already suggested fsck for the task.) >>>I do have to wonder when I see thinking like this; will it be easier or >>>harder to fix 100-year-old bugs than 20-year-old bugs? >> >>My experience has been that a 6-month old bug is just as easy to fix >>as a 5-year-old bug. >You were very, very lucky, then. We had problems finding sources >and build procedures for stuff only 3 years old that had lain >fallow. We almost didn't have new RSX-20F releases because of >that. Who said anything about lucky? The "documentation" was impenetrable and/or lost after 6 months. Some source code was "lost" (version "compression" had lost the version used to build the application initially), fortunately, a later version of the same code worked. It doesn't get much more difficult. The next level of un-debugability is when competent programmers and analysts of the language have retired or died. We're just lucky that C is still in fashion after about 30 years. >> So a 20-year-old bug is probably no easier to >>fix than a 100-year-old one. -- Bernd Felsche - Innovative Reckoning Perth, Western Australia ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Wed, 26 Jul 00 09:38:26 GMT Organization: UltraNet Communications, Inc. Lines: 28 Message-ID: <8lmm0h$l8b$5@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> <8ljvpf$o2c$5@bob.news.rcn.net> X-Trace: 8dNV3v88TAvaJGCzqN1B6aaEaUOWDrSFaHQ4hJyPPYg= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 26 Jul 2000 12:36:33 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!newsfeeds.belnet.be!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!216-164-247-65 Xref: chonsp.franklin.ch alt.folklore.computers:60579 In article , Jim Thomas wrote: >>>>>> "/BAH" == jmfbahciv writes: > > /BAH> In article , > /BAH> Jim Thomas wrote: > >>>>>>> "/BAH" == jmfbahciv writes: > > >> /BAH> Answering SPRs > >> /BAH> were not a favorite pasttime for most people. Actually, I can't > >> /BAH> think of a single person who liked to answer them. > >> > >> Stand up and be counted. I learned a lot doing so :-) > > /BAH> Oh, I didn't mean it that way. One always learned a lot from > /BAH> working on an SPR. I literally meant _answering_ the SPR. > /BAH> That is, writing an answer in English ASCII that would pass > /BAH> the reviewer and not get us into any lawsuit corners. > >Maybe we didn't have as many lawyers yet when I was there :-) Perhaps. Some of us got very stupid...and stayed. Did you have to fill out those project time cards? /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Wed, 26 Jul 00 09:56:31 GMT Organization: UltraNet Communications, Inc. Lines: 28 Message-ID: <8lmn2e$l8b$8@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> <397D7A22.A2108963@netinsight.se> X-Trace: h/1P41Y7PEBfije9dP+9IL5ucl+yx6x+jGZfHhTdYb0= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 26 Jul 2000 12:54:38 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!216-164-247-65 Xref: chonsp.franklin.ch alt.folklore.computers:60581 In article , Jim Thomas wrote: >>>>>> "Johnny" == Johnny Billquist writes: > > Johnny> Jim Thomas wrote: > >> > >> >>>>> "/BAH" == jmfbahciv writes: > >> > >> /BAH> Why spend time when a NVMS$;-3DIRSX$EX$$ will do? > >> > >> Ummm, I think you meant EX$$ :-) > > Johnny> That won't work either. The EX$$ will not be executed, since > Johnny> the macro is aborted once the search fails. I don't remember > Johnny> the modifier to return a code instead of aborting on fail, > Johnny> but I'm sure anyone here can correct the code, given a few > Johnny> minutes and a manual. :-) > >? The ";" causes the search failure to terminate the loop. It does not >cause the discarding of the rest of the command. The EX$$ happens. I don't remember anymore. This is one of those editing commands that the fingers knew what the brain had forgotten. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Wed, 26 Jul 00 09:42:11 GMT Organization: UltraNet Communications, Inc. Lines: 12 Message-ID: <8lmm7i$l8b$6@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <964099762.86210@shelley.paradise.net.nz> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <964148894.748064@shelley.paradise.net.nz> <48A6A565160289F7.B6AB1B2AC2E18C44.8D36189F69B3CB75@lp.airnews.net> <441EF078B6B19400.7B2F5BFE6C01266F.68C4AF216247D0AA@lp.airnews.net> <82mnnsk34vsng2u857c46h5lb5br0q5k1i@4ax.com> X-Trace: mCRDMn4yVlgI8+PRwVQQaT0Z1PUh/cI0P9zPAzOAj2U= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 26 Jul 2000 12:40:18 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!diablo.theplanet.net!europa.netcrusader.net!207.172.3.37!feed1.news.rcn.net!rcn!216-164-247-65 Xref: chonsp.franklin.ch alt.folklore.computers:60582 In article <82mnnsk34vsng2u857c46h5lb5br0q5k1i@4ax.com>, Brian Inglis wrote: >On 22 Jul 2000 02:16:21 GMT, jmaynard@thebrain.conmicro.cx (Jay >Maynard) wrote: > >>On Sat, 22 Jul 2000 01:54:35 GMT, Ric Werme wrote: Definition request. What's an "escape tape"? /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Wed, 26 Jul 00 11:16:58 GMT Organization: UltraNet Communications, Inc. Lines: 20 Message-ID: <8lmrp9$ob4$3@bob.news.rcn.net> References: <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <964148894.748064@shelley.paradise.net.nz> <48A6A565160289F7.B6AB1B2AC2E18C44.8D36189F69B3CB75@lp.airnews.net> <441EF078B6B19400.7B2F5BFE6C01266F.68C4AF216247D0AA@lp.airnews.net> <82mnnsk34vsng2u857c46h5lb5br0q5k1i@4ax.com> <8lmm7i$l8b$6@bob.news.rcn.net> <191EF92D5E4E9A14.0B76D0AABBC43419.531001220794130E@lp.airnews.net> X-Trace: gLv2i3K+LIdkKP02G8an6jQO7jQxdIqpwv2cZLm8YpE= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 26 Jul 2000 14:15:05 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newsfeed.icl.net!newsfeed.icl.net!netnews.com!feed1.news.rcn.net!rcn!209-122-234-217 Xref: chonsp.franklin.ch alt.folklore.computers:60583 In article <191EF92D5E4E9A14.0B76D0AABBC43419.531001220794130E@lp.airnews.net>, jmaynard@thebrain.conmicro.cx (Jay Maynard) wrote: >On Wed, 26 Jul 00 09:42:11 GMT, jmfbahciv@aol.com wrote: >> Definition request. What's an "escape tape"? > >It's a tape you build of stuff you want to take with you >when you change jobs, typically created in the short >interval between finding out you've got >the new job and telling your boss you're leaving. >The term is common in IBM >shops. (Perhaps should be added to the Jargon File?) Aahh!!! Thank you :-). Silly me. I was thinking of escape key, escape sequence..... /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Wed, 26 Jul 00 09:51:28 GMT Organization: UltraNet Communications, Inc. Lines: 72 Message-ID: <8lmmov$l8b$7@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <8lksuj$qr7$1@news.iastate.edu> X-Trace: rsePcoHIpN1D1oFNP6WY5RpAKfV96lMJxq855RnpBQM= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 26 Jul 2000 12:49:35 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!newsxfer.eecs.umich.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!216-164-247-65 Xref: chonsp.franklin.ch alt.folklore.computers:60589 In article <8lksuj$qr7$1@news.iastate.edu>, rhawkins@iastate.edu (Rick Hawkins) wrote: >In article <8lepbo$gfh$5@bob.news.rcn.net>, wrote: >>In article <964298714.421191@shelley.paradise.net.nz>, >> don@news.daedalus.co.nz (Don Stokes) wrote: > >>>Well, time_t is a *signed* 32 bit int. For file systems, barring >>>idiocy, there aren't any file dates < 1970, > >> Since I started contributing to social security >>in 1962, I would certainly hope that there is software that >>can read file dates before 1970. > >*giggle* > >Yikes, BAH, you must be *old* if you're expecting your social security >contributions to make a difference :) Not the contributions...the fact that I did them. That's going to be key rather than amounts. > >I ask my students each semester how many expect to see anything out >of social security. A couple of years ago it was still around 5%. >Not only is it dropping, but now they *laugh*. I haven't expected that since the congress critters revamped social security...when was it...late 70s or early 80s. What it's going to be used for is access to medical insurance. One will need an OK from social security to get access to a group medical insurance; one will need access to a group medical insurance to get an appointment with a doctor. One will need both access to medical insurance _and_ a doctor to get any drug, unless one lives on an international border. > >>> ... so they can get an extra 68 >>>years without a new file system by the simple expedient of declaring the >>>time stamps to be unsigned. Cobine it with a 64 bit internal time_t (or >>>declare time_t to be unsigned as well, which I'm sure would be bug food) >>>and watch your file system problem get deferred until 2106. > >>It's thinking like this that caused all the Y2K expenditures. > >However, someone did an analysis on what it would have cost to >*avoid* the Y2K changes, and found that that it dwarfed the cleanup >costs . . . those extra digits (and entry time) added up . . . Entry time wouldn't have taken time. Damn..I almost remembered the column punch that started a dup field.. But you are right. What I was talking about was the guy just making a decision without knowing the consequences. Most of the people I worked with, who made those 2-digit vs. 4-digit decisions, actually spent time thinking about the pros and cons of not using a complete data field. > >hawk, who doesn't agree that contributing to social security is equivalent >to flushing the money down the toilet--at least that way he gets >to watch it spin . . . But that's to prevent you from getting dizzy, Hawk. The government is there to protect you. I guess I'd better put a grinning emoticon here before those nuts decide I'm making a serious statement. /BAH Subtract a hundred and four for e-mail. ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <397dd579$1@news.ucsc.edu> Date: 25 Jul 2000 10:59:21 -0800 X-Trace: 25 Jul 2000 10:59:21 -0800, sundance.cse.ucsc.edu Lines: 44 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:60558 In article <39784803.44BB656@trailing-edge.com>, Tim Shoppa wrote: >Eugene Miya wrote: >> In article <397711EB.CC1C5723@netinsight.se>, >> Johnny Billquist wrote: >> >VMS counts in 100ns(?) increments in a 64 bit counter. If I remember >> >that correctly, it gives VMS a range of 29227 years, this added to >> >1876 would give VMS breaking sometime in the year 31103, or possibly 31104. >> >> Oh, this is interesting. Compaq changed this? >> My foggy memory had a 60 Hz to 50/100 Hz (max) clock which lots of >> people confused. Which SYS$call? > >Eugene, you've proved the old adage "the best way to get correct >information on Usenet is to post incorrect information" yet again :-). Oh, true. Trust me, I have refined the art of network interogation. I've got it down to a way of characterizing classes of questions into Boolean forms: What simpler question, for instance, could one ask: T or F? For the kid asking homework on the net like 1+1=? So T or F? 1+1=0? 1+1=1? 1+1=2? Those are easy ones. 1+1=1024? This is perceived as a ridiculous one. But what about this one? 1+1=10? This can be seen as an ambiguous one. The ambiguous questions are the best. Engineers and scientists need always be mindful of the engineer and gillotine joke. 8^) Now which SYS$call? 8^) >No, Compaq did not change this; the granularity always has been, >and will probably remain, 100 nanoseconds inside VMS. See the famous >Y10K SPR at > > http://www.trailing-edge.com/www/y10k.txt Are you telling me that 11/780s had 0.1 ms clocks? If no, when did the change occur? ###### From: rhawkins@iastate.edu (Rick Hawkins) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 25 Jul 2000 19:39:57 GMT Organization: Debian Linux site Lines: 35 Message-ID: <8lkqed$qn5$1@news.iastate.edu> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8la48u$h1q@nnrp2.farm.idt.net> <964215436.300480@server16.cable.com> NNTP-Posting-Host: pv2087.vincent.iastate.edu Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-koe1.dfn.de!news-was.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!logbridge.uoregon.edu!newsrelay.iastate.edu!news.iastate.edu!rhawkins Xref: chonsp.franklin.ch alt.folklore.computers:60591 In article <964215436.300480@server16.cable.com>, P Linnane wrote: > wrote in message >news:slrn8nh88a.3lf.TheCentralScrutinizer.199@C298344-A.arvada1.co.home.com. >> That seems kind of silly except for as a museum. A $500 computer from >> best buy could easilly hold all the software, their database, and run >> an interpreter to run their software. >Your wisdom amazes me, so let me get this straight: >For $500 I can get:: >Complete AP/Payroll and database packages for a medium/large-sized company >PC-Interpreter for Univac OS and (presumably COBOL or Fortran) applications >PC-compatible terminals x? >PC-compatible printers >PC-compatible networking >Complete conversion of existing system >Client re-training >25 year warranty. >. . . and you'll even throw in a piece-of-crap computer ? Even better--he could sell you this in the early 70's :) hawk -- Richard E. Hawkins dochawk@psu.edu [regardless of where the message says it comes from] These opinions will not be those of Penn State until they pay my retainer. ###### From: rhawkins@iastate.edu (Rick Hawkins) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 25 Jul 2000 20:22:43 GMT Organization: Debian Linux site Lines: 42 Message-ID: <8lksuj$qr7$1@news.iastate.edu> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> NNTP-Posting-Host: pv2087.vincent.iastate.edu Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-koe1.dfn.de!RRZ.Uni-Koeln.DE!fu-berlin.de!logbridge.uoregon.edu!newsrelay.iastate.edu!news.iastate.edu!rhawkins Xref: chonsp.franklin.ch alt.folklore.computers:60594 In article <8lepbo$gfh$5@bob.news.rcn.net>, wrote: >In article <964298714.421191@shelley.paradise.net.nz>, > don@news.daedalus.co.nz (Don Stokes) wrote: >>Well, time_t is a *signed* 32 bit int. For file systems, barring >>idiocy, there aren't any file dates < 1970, > Since I started contributing to social security >in 1962, I would certainly hope that there is software that >can read file dates before 1970. *giggle* Yikes, BAH, you must be *old* if you're expecting your social security contributions to make a difference :) I ask my students each semester how many expect to see anything out of social security. A couple of years ago it was still around 5%. Not only is it dropping, but now they *laugh*. >> ... so they can get an extra 68 >>years without a new file system by the simple expedient of declaring the >>time stamps to be unsigned. Cobine it with a 64 bit internal time_t (or >>declare time_t to be unsigned as well, which I'm sure would be bug food) >>and watch your file system problem get deferred until 2106. >It's thinking like this that caused all the Y2K expenditures. However, someone did an analysis on what it would have cost to *avoid* the Y2K changes, and found that that it dwarfed the cleanup costs . . . those extra digits (and entry time) added up . . . hawk, who doesn't agree that contributing to social security is equivalent to flushing the money down the toilet--at least that way he gets to watch it spin . . . -- Richard E. Hawkins dochawk@psu.edu [regardless of where the message says it comes from] These opinions will not be those of Penn State until they pay my retainer. ###### From: Charles Richmond Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Tue, 25 Jul 2000 13:46:37 -0700 Organization: Cannine Computer Center Lines: 29 Message-ID: <397DFCAD.FA278C52@ev1.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <8lh601$orq$1@flywheel.innovative.iinet.net.au> <8lh895$iui$3@bob.news.rcn.net> <8lk1l5$uah$1@flywheel.innovative.iinet.net.au> Reply-To: richmond@ev1.net X-Complaints-To: newsabuse@supernews.com X-Mailer: Mozilla 4.7C-CCK-MCD {C-UDP; EBM-APPLE} (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.cwix.com!sjc-peer.news.verio.net!news.verio.net!sn-xit-01!supernews.com!sn-inject-01!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60624 Bernd Felsche wrote: > > jmfbahciv@aol.com writes: > > [snip...] [snip...] [snip...] > > >You were very, very lucky, then. We had problems finding sources > >and build procedures for stuff only 3 years old that had lain > >fallow. We almost didn't have new RSX-20F releases because of > >that. > > Who said anything about lucky? The "documentation" was impenetrable > and/or lost after 6 months. Some source code was "lost" (version > "compression" had lost the version used to build the application > initially), fortunately, a later version of the same code worked. > It doesn't get much more difficult. > > The next level of un-debugability is when competent programmers and > analysts of the language have retired or died. We're just lucky that > C is still in fashion after about 30 years. > I heard that the sources for the WordPerfect word processor were lost... and when it was sold to Corel, the program had to be reconstructed. Does anyone have any info on this??? -- +-------------------------------------------------------------+ | Charles and Francis Richmond | +-------------------------------------------------------------+ ###### From: Alexandre Pechtchanski Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Organization: Rockefeller University Hospital (GCRC), New York Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <8lh601$orq$1@flywheel.innovative.iinet.net.au> <8lh895$iui$3@bob.news.rcn.net> <8lk1l5$uah$1@flywheel.innovative.iinet.net.au> <397DFCAD.FA278C52@ev1.net> X-Newsreader: Forte Agent 1.7/32.534 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 13 Date: Tue, 25 Jul 2000 17:54:00 -0400 NNTP-Posting-Host: 129.85.24.56 X-Trace: rockyd.rockefeller.edu 964562246 129.85.24.56 (Tue, 25 Jul 2000 17:57:26 EDT) NNTP-Posting-Date: Tue, 25 Jul 2000 17:57:26 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-koe1.dfn.de!news-was.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.nyu.edu!rockyd.rockefeller.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60620 On Tue, 25 Jul 2000 13:46:37 -0700, Charles Richmond wrote: [ snip ] >I heard that the sources for the WordPerfect word processor were lost... >and when it was sold to Corel, the program had to be reconstructed. >Does anyone have any info on this??? I've heard that this happened to the Windows Write - that's why we now have this abomination (WordPad) instead. -- [ When replying, remove *'s from address ] Alexandre Pechtchanski, Systems Manager, RUH, NY ###### From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Tue, 25 Jul 2000 19:42:05 -0600 Organization: Systematic Software Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca Message-ID: <82mnnsk34vsng2u857c46h5lb5br0q5k1i@4ax.com> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <964099762.86210@shelley.paradise.net.nz> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <964148894.748064@shelley.paradise.net.nz> <48A6A565160289F7.B6AB1B2AC2E18C44.8D36189F69B3CB75@lp.airnews.net> <441EF078B6B19400.7B2F5BFE6C01266F.68C4AF216247D0AA@lp.airnews.net> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.148.147.161 X-Original-NNTP-Posting-Host: 207.148.147.161 X-Trace: 25 Jul 2000 19:42:08 -0700, 207.148.147.161 Lines: 58 X-Original-NNTP-Posting-Host: 204.50.1.43 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-koe1.dfn.de!news-was.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.flash.net!nntp.cadvision.com!news.cadvision.com!207.148.147.161 Xref: chonsp.franklin.ch alt.folklore.computers:60617 On 22 Jul 2000 02:16:21 GMT, jmaynard@thebrain.conmicro.cx (Jay Maynard) wrote: >On Sat, 22 Jul 2000 01:54:35 GMT, Ric Werme wrote: >>Well, I can't speak for VMS (or Tru64 for that matter), but there's a >>barely known tweak to Tru64. If you enable the MICRO_TIME option in >>the /sys/conf/NAME file (and I thing NTP_TIME or somesuch) and rebuild >>your kernel, gettimeofday() will adjust the time based on the CPU's >>cycle counter to give you near microsecond resolution. If you need >>anything finer, you can always read the cycle counter with the RPCC >>instruction or library routine. (Too lazy to check from home.) > >Hmmm. Wonder if that's easy to add to Linux. ...and if there's a similar >thing that can be added to Intel platforms. I'm running into problems with >Hercules' emulation of the 370 time-of-day clock, which is defined as a >64-bit value where bit 51 (0=MSB) is incremented every microsecond...and the IIRC, the 370 TOD clock (and similarly the VAX equivalent) is incremented in units of 1/4096us at a rate comparable to the processor instruction cycle time (which gives unique TOD clock values for each instruction) which is equivalent to incrementing bit 51 every us (and bit 31 ~~1s) as you stated above. The epoch of the 370 TOD clock is 1-Jan-1900 00:00:00 UTC. IBM 370 timing could be done using the interval timer device (0x011? -- 1/300s?), the CPU timer, or the clock comparator, which interrupts when the TOD clock value equals the comparator value, and I believe that was the facility used by modern 370-390 OSes. I remember there were VM options to update the virtual CPU timer either with virtual processor time or real clock time each time the virtual machine was dispatched. ISTR the VAX clock incremented in similar if not identical units to the 370 TOC clock, but the epoch was the modified Julian day (JD - 2,400,000.5) 17-Nov-1858 00:00:00 UTC. I had to do some interconversion between the two clock formats a long while ago, but I think the code may in Rexx on an "escape tape" (3420-8 6250GCR -- no access to that hardware for years) in the basement somewhere. >standard Linux timing facilities only provide 1/100 second resolution. If a >program does things like try to time the CPU by reading the TOD clock, >running a 50,000 instruction loop, reading the TOD clock again, and >dividing, on a fast enough machine, that'll get a divide-by-zero program >check instead of a result. (As Hercules gets faster, the bar for "fast >enough machine" gets lower, and lower...) > >My current plan is to use the Linux /dev/rtc to provide a faster tick (by >default, 1024/second, but settable to 8192/second). If there's a reasonably >portable way to get faster, I'll use it. Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### From: Jim Thomas Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 25 Jul 2000 15:43:31 -1000 Organization: Canada France Hawai`i Telescope Lines: 20 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> <397D7A22.A2108963@netinsight.se> NNTP-Posting-Host: atlas.cfht.hawaii.edu X-Trace: news.hawaii.edu 964575813 12949 128.171.80.135 (26 Jul 2000 01:43:33 GMT) X-Complaints-To: usenet@hawaii.edu NNTP-Posting-Date: 26 Jul 2000 01:43:33 GMT X-Newsreader: Gnus v5.7/Emacs 20.6 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!fu-berlin.de!arclight.uoregon.edu!hammer.uoregon.edu!news.hawaii.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60614 >>>>> "Johnny" == Johnny Billquist writes: Johnny> Jim Thomas wrote: >> >> >>>>> "/BAH" == jmfbahciv writes: >> >> /BAH> Why spend time when a NVMS$;-3DIRSX$EX$$ will do? >> >> Ummm, I think you meant EX$$ :-) Johnny> That won't work either. The EX$$ will not be executed, since Johnny> the macro is aborted once the search fails. I don't remember Johnny> the modifier to return a code instead of aborting on fail, Johnny> but I'm sure anyone here can correct the code, given a few Johnny> minutes and a manual. :-) ? The ";" causes the search failure to terminate the loop. It does not cause the discarding of the rest of the command. The EX$$ happens. Nothead ###### From: Jim Thomas Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 25 Jul 2000 15:44:39 -1000 Organization: Canada France Hawai`i Telescope Lines: 20 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> <8ljvpf$o2c$5@bob.news.rcn.net> NNTP-Posting-Host: atlas.cfht.hawaii.edu X-Trace: news.hawaii.edu 964575879 12949 128.171.80.135 (26 Jul 2000 01:44:39 GMT) X-Complaints-To: usenet@hawaii.edu NNTP-Posting-Date: 26 Jul 2000 01:44:39 GMT X-Newsreader: Gnus v5.7/Emacs 20.6 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!uni-erlangen.de!newsfeed1.telenordia.se!news.algonet.se!algonet!newsfeedZ.netscum.dQ!netscum.int!newsfeed.stanford.edu!arclight.uoregon.edu!hammer.uoregon.edu!news.hawaii.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60615 >>>>> "/BAH" == jmfbahciv writes: /BAH> In article , /BAH> Jim Thomas wrote: >>>>>>> "/BAH" == jmfbahciv writes: >> /BAH> Answering SPRs >> /BAH> were not a favorite pasttime for most people. Actually, I can't >> /BAH> think of a single person who liked to answer them. >> >> Stand up and be counted. I learned a lot doing so :-) /BAH> Oh, I didn't mean it that way. One always learned a lot from /BAH> working on an SPR. I literally meant _answering_ the SPR. /BAH> That is, writing an answer in English ASCII that would pass /BAH> the reviewer and not get us into any lawsuit corners. Maybe we didn't have as many lawyers yet when I was there :-) Nothead ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 26 Jul 2000 02:34:22 GMT Organization: Neosoft (using Airnews.net!) Lines: 35 Message-ID: <053331B09EF38BC6.6E818D2DB25CFA9A.7C7A6187F382747E@lp.airnews.net> X-Orig-Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <964099762.86210@shelley.paradise.net.nz> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <964148894.748064@shelley.paradise.net.nz> <48A6A565160289F7.B6AB1B2AC2E18C44.8D36189F69B3CB75@lp.airnews.net> <441EF078B6B19400.7B2F5BFE6C01266F.68C4AF216247D0AA@lp.airnews.net> <82mnnsk34vsng2u857c46h5lb5br0q5k1i@4ax.com> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Tue Jul 25 21:34:22 2000 NNTP-Posting-Host: !Xc[f1k-W"D?J]X (Encoded at Airnews!) User-Agent: slrn/0.9.5.4 (UNIX) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!howland.erols.net!news-out.nntp.airnews.net.MISMATCH!cabal10.airnews.net!news.airnews.net!cabal14.airnews.net!news.airnews.net!cabal2.airnews.net!news-f.iadfw.net!jmaynard Xref: chonsp.franklin.ch alt.folklore.computers:60572 On Tue, 25 Jul 2000 19:42:05 -0600, Brian Inglis wrote: >IIRC, the 370 TOD clock (and similarly the VAX equivalent) is >incremented in units of 1/4096us at a rate comparable to the >processor instruction cycle time (which gives unique TOD clock >values for each instruction) which is equivalent to incrementing >bit 51 every us (and bit 31 ~~1s) as you stated above. The epoch >of the 370 TOD clock is 1-Jan-1900 00:00:00 UTC. The 370 POO only says that the clock will be incremented at such a rate as to make bit 51 increment every microsecond, and that the rate will be comparable to the rate of execution of the CPU. It doesn't tie the bits any more tightly than that. I don't know how many, if any, 370/390s implement the full 244-picosecond resolution - after all, that's a 4 GHz clock. The recommended 370 epoch is 0000 1 Jan 1900, but that wasn't universal; OS/360 used 0000 1 Jan 1960 (and assumed, at IPL time, that the clock HAD to be wrong if it was after 1999). >IBM 370 timing could be done using the interval timer device >(0x011? -- 1/300s?), the CPU timer, or the clock comparator, >which interrupts when the TOD clock value equals the comparator >value, and I believe that was the facility used by modern 370-390 >OSes. The interval timer lives at real address X'50', as a 32-bit value decremented in bit 23 (0=MSB) every 1/300 second - and not expressly related to anything. It is not present in 370/XA or ESA. >I had to do some interconversion between the two clock formats a >long while ago, but I think the code may in Rexx on an "escape >tape" (3420-8 6250GCR -- no access to that hardware for years) in >the basement somewhere. If you'd like, I can read that tape and get the data to you... ###### From: Louis RAPHAEL Subject: Re: Longest running computer Newsgroups: alt.folklore.computers References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <8lh601$orq$1@flywheel.innovative.iinet.net.au> <8lh895$iui$3@bob.news.rcn.net> <8lk1l5$uah$1@flywheel.innovative.iinet.net.au> <397DFCAD.FA278C52@ev1.net> Organization: Societe pour la promotion du petoncle vert User-Agent: tin/pre-1.4-980818 ("Laura") (UNIX) (SunOS/5.6 (sun4m)) Lines: 12 Message-ID: Date: Wed, 26 Jul 2000 03:35:49 GMT NNTP-Posting-Host: 132.206.51.205 X-Complaints-To: abuse@mcgill.ca X-Trace: carnaval.risq.qc.ca 964582549 132.206.51.205 (Tue, 25 Jul 2000 23:35:49 EDT) NNTP-Posting-Date: Tue, 25 Jul 2000 23:35:49 EDT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!nyc-news-feed1.bbnplanet.com!news.gtei.net!colt.net!news.maxwell.syr.edu!sunqbc.risq.qc.ca!carnaval.risq.qc.ca.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60607 Charles Richmond wrote: : I heard that the sources for the WordPerfect word processor were lost... : and when it was sold to Corel, the program had to be reconstructed. : Does anyone have any info on this??? None, except to say that IT SHOWS! when you try to use the old WordPerfect Corp (and even the Novell) versions, as compared to the Corel ones. The old ones are much more bug-free, and still much nicer, IMHO. Louis ###### From: fixxit@bright.net Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l146o$14ca@edrn.newsguy.com> <8la48u$h1q@nnrp2.farm.idt.net> X-Newsreader: Forte Agent 1.8/16.548 X-No-Archive: yes MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 21 Date: Wed, 26 Jul 2000 00:08:21 -0400 NNTP-Posting-Host: 216.201.28.69 X-Complaints-To: abuse@bright.net X-Trace: cletus.bright.net 964584378 216.201.28.69 (Wed, 26 Jul 2000 00:06:18 EDT) NNTP-Posting-Date: Wed, 26 Jul 2000 00:06:18 EDT Organization: bright.net Ohio Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!newsfeed1.online.no!newsfeed.online.no!uninett.no!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!205.212.123.11!cletus.bright.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60626 kragen@dnaco.net (Kragen Sitaker) wrote: >In article , > wrote: >>>Charles Eicher wrote: >>>[...] >>>: At that time, the CPU would easily have been at least 25 years old. >>... >> >>That seems kind of silly except for as a museum. A $500 computer from >>best buy could easilly hold all the software, their database, and run >>an interpreter to run their software. > >It probably wouldn't run without breaking for 25 years. Besides, "Where's the fun in that"? When replying, send a copy via email--sometimes I don't keep up with the group! Note the URL for email corrections--Visit my web page to see what I do... www.bright.net/~fixit ###### From: jmaynard@thebrain.conmicro.cx (Jay Maynard) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 26 Jul 2000 14:09:35 GMT Organization: Neosoft (using Airnews.net!) Lines: 7 Message-ID: <191EF92D5E4E9A14.0B76D0AABBC43419.531001220794130E@lp.airnews.net> X-Orig-Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <964099762.86210@shelley.paradise.net.nz> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <964148894.748064@shelley.paradise.net.nz> <48A6A565160289F7.B6AB1B2AC2E18C44.8D36189F69B3CB75@lp.airnews.net> <441EF078B6B19400.7B2F5BFE6C01266F.68C4AF216247D0AA@lp.airnews.net> <82mnnsk34vsng2u857c46h5lb5br0q5k1i@4ax.com> <8lmm7i$l8b$6@bob.news.rcn.net> Reply-To: jmaynard@conmicro.cx Abuse-Reports-To: abuse at airmail.net to report improper postings NNTP-Proxy-Relay: library2.airnews.net NNTP-Posting-Time: Wed Jul 26 09:09:35 2000 NNTP-Posting-Host: !b$p21k-Ve"Z wrote: > Definition request. What's an "escape tape"? It's a tape you build of stuff you want to take with you when you change jobs, typically created in the short interval between finding out you've got the new job and telling your boss you're leaving. The term is common in IBM shops. (Perhaps should be added to the Jargon File?) ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Thu, 27 Jul 00 09:32:05 GMT Organization: UltraNet Communications, Inc. Lines: 30 Message-ID: <8lpa0t$jeb$1@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <8lksuj$qr7$1@news.iastate.edu> <8lmmov$l8b$7@bob.news.rcn.net> X-Trace: 5gKF5KPBRMLPsmqKFoNqB5t96rNC1XMsncaQYNxBlW0= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 27 Jul 2000 12:30:21 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!news.rhrz.uni-bonn.de!RRZ.Uni-Koeln.DE!news.netcologne.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-216-205 Xref: chonsp.franklin.ch alt.folklore.computers:60674 In article , Jim Thomas wrote: >>>>>> "/BAH" == jmfbahciv writes: > > /BAH> Entry time wouldn't have taken time. Damn..I almost remembered > /BAH> the column punch that started a dup field.. > >"0" - normally "/" for an alpha dup :-) I don't think I used alpha dup very often. It seems to me that there was a side effect of numeric dup that would stop the card if [something] occurred. I often designed my drum cards to also serve as data checks (not to be confused with verifiers...I wasn't that blest). I do remember jobs that had every n+1th card different. Those 029s were fantastic--push a key and you get a temporary alternate drum card. When I only had 026s to work with, I would key in the all of the 1st record cards. Then I would key in the 2nd, 3rd, ..., nth cards inserting the first card in the hopper by hand. All of this to avoid changing drum cards or using 2 026s at once. It was a fantastically fun game to design drum cards that would cut the data entry time by 20%, including verifying which was always done by eyeballing. /BAH Subtract a hundred and four for e-mail. ###### From: Jim Thomas Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 26 Jul 2000 13:57:42 -1000 Organization: Canada France Hawai`i Telescope Lines: 8 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> <8ljvpf$o2c$5@bob.news.rcn.net> <8lmm0h$l8b$5@bob.news.rcn.net> NNTP-Posting-Host: atlas.cfht.hawaii.edu X-Trace: news.hawaii.edu 964655862 2594 128.171.80.135 (26 Jul 2000 23:57:43 GMT) X-Complaints-To: usenet@hawaii.edu NNTP-Posting-Date: 26 Jul 2000 23:57:43 GMT X-Newsreader: Gnus v5.7/Emacs 20.6 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.direct.ca!look.ca!hammer.uoregon.edu!news.hawaii.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60704 >>>>> "/BAH" == jmfbahciv writes: /BAH> Did you have to fill out those project time cards? Yup, well, some kind of sheets with various projects coded. I don't remember if they were already there when I arrived. Nothead ###### From: Jim Thomas Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 26 Jul 2000 13:59:59 -1000 Organization: Canada France Hawai`i Telescope Lines: 6 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <8lksuj$qr7$1@news.iastate.edu> <8lmmov$l8b$7@bob.news.rcn.net> NNTP-Posting-Host: atlas.cfht.hawaii.edu X-Trace: news.hawaii.edu 964656000 2594 128.171.80.135 (27 Jul 2000 00:00:00 GMT) X-Complaints-To: usenet@hawaii.edu NNTP-Posting-Date: 27 Jul 2000 00:00:00 GMT X-Newsreader: Gnus v5.7/Emacs 20.6 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!newsfeed.direct.ca!look.ca!hammer.uoregon.edu!news.hawaii.edu!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60706 >>>>> "/BAH" == jmfbahciv writes: /BAH> Entry time wouldn't have taken time. Damn..I almost remembered /BAH> the column punch that started a dup field.. "0" - normally "/" for an alpha dup :-) ###### Message-ID: <398016A9.87219146@netinsight.se> From: Johnny Billquist Organization: Netinsight AB X-Mailer: Mozilla 4.5 [en] (X11; I; Linux 2.2.14 i686) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> <964269064.98678@shelley.paradise.net.nz> <8lc76n$5co$3@bob.news.rcn.net> <397D7A22.A2108963@netinsight.se> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Lines: 39 Date: Thu, 27 Jul 2000 11:02:02 GMT NNTP-Posting-Host: 194.16.221.33 X-Complaints-To: abuse@telia.com X-Trace: newsc.telia.net 964695722 194.16.221.33 (Thu, 27 Jul 2000 13:02:02 MET DST) NNTP-Posting-Date: Thu, 27 Jul 2000 13:02:02 MET DST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!rz.uni-karlsruhe.de!schlund.de!newsfeed01.sul.t-online.de!t-online.de!newsfeed.online.be!news.algonet.se!algonet!news-peer-europe.sprintlink.net!news.sprintlink.net!newsfeed.bahnhof.se!news.defero.net!news-sto.telia.net!masternews.telia.net!newsc.telia.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60658 Jim Thomas wrote: > > >>>>> "Johnny" == Johnny Billquist writes: > > Johnny> Jim Thomas wrote: > >> > >> >>>>> "/BAH" == jmfbahciv writes: > >> > >> /BAH> Why spend time when a NVMS$;-3DIRSX$EX$$ will do? > >> > >> Ummm, I think you meant EX$$ :-) > > Johnny> That won't work either. The EX$$ will not be executed, since > Johnny> the macro is aborted once the search fails. I don't remember > Johnny> the modifier to return a code instead of aborting on fail, > Johnny> but I'm sure anyone here can correct the code, given a few > Johnny> minutes and a manual. :-) > > ? The ";" causes the search failure to terminate the loop. It does not > cause the discarding of the rest of the command. The EX$$ happens. I guess I should look up the manuals before replying, but here I go... Yes, the ; is used to exit the loop, if the numeric argument is zero (or was it not zero?). However, a search-and-replace which fails will not return a value, but will instead abort the macro. Thus, you have a modifier you can stick in front of 'N' to get it to instead return a result value. I'll look up the manual tonight, and try running it on my RSX system. Johnny -- Johnny Billquist | johnny.billquist@netinsight.net Net Insight AB | phone: +46 8 685 04 88 Västberga Allé 9 | fax: +46 8 685 04 20 Box 42093 | SE-126 30 STOCKHOLM, Sweden | http://www.netinsight.net ###### Message-ID: <398018C6.9354C3A7@netinsight.se> From: Johnny Billquist Organization: Netinsight AB X-Mailer: Mozilla 4.5 [en] (X11; I; Linux 2.2.14 i686) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <39784803.44BB656@trailing-edge.com> <397dd579$1@news.ucsc.edu> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Lines: 48 Date: Thu, 27 Jul 2000 11:11:03 GMT NNTP-Posting-Host: 194.16.221.33 X-Complaints-To: abuse@telia.com X-Trace: newsc.telia.net 964696263 194.16.221.33 (Thu, 27 Jul 2000 13:11:03 MET DST) NNTP-Posting-Date: Thu, 27 Jul 2000 13:11:03 MET DST Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!newsfeedZ.netscum.dQ!netscum.int!uninett.no!news.ost.eltele.no!newsfeed1.enitel.no!masternews.telia.net!newsc.telia.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60662 Eugene Miya wrote: > > In article <39784803.44BB656@trailing-edge.com>, > Tim Shoppa wrote: > >No, Compaq did not change this; the granularity always has been, > >and will probably remain, 100 nanoseconds inside VMS. See the famous > >Y10K SPR at > > > > http://www.trailing-edge.com/www/y10k.txt > > Are you telling me that 11/780s had 0.1 ms clocks? > If no, when did the change occur? You are confusing two things here, Eugene. *VMS* have a 100 ns granularity clock. The VAX-11/780 only used a 10ms interrupt interval, which meant that the VMS clock was incremented 100 for each interrupt. Thus, on the -11/780, you could specify an interval as short as 100 ns, but the OS never guarantees you that *exactly* that interval have passed, only that *atleast* that interval passed. And the hardware interrupts occuring at 10 ms intervals meant that you'd on average wait 5 ms, even if you specified 100 ns. But that's the life on any OS you are using, only that most don't have such a fine granularity of time in the OS. Of course, you should specify exactly the time you do want to wait, since you might come across a system sometime in the future which actually do interrupt at 100 ns intervals, at which time the VMS time value would be incremented by 1 on each interrupt. So, VMS have *always* had 100 ns granularity on *all* time values, and will likely always have. All SYS$calls taking time values think that they are expressed in 100 ns increments, and that's the way it has been since VMS 1.0. Now, do we understand each other? :-) Johnny -- Johnny Billquist | johnny.billquist@netinsight.net Net Insight AB | phone: +46 8 685 04 88 Västberga Allé 9 | fax: +46 8 685 04 20 Box 42093 | SE-126 30 STOCKHOLM, Sweden | http://www.netinsight.net ###### From: "Charlie Gibbs" Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 27 Jul 00 12:14:46 -0800 Organization: http://extra.newsguy.com Lines: 28 Message-ID: <1044.243T2329T7345222@sky.bus.com> References: <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <8lksuj$qr7$1@news.iastate.edu> <8lmmov$l8b$7@bob.news.rcn.net> <8lpa0t$jeb$1@bob.news.rcn.net> NNTP-Posting-Host: p-655.newsdawg.com X-Newsreader: THOR 2.5a (Amiga;TCP/IP) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!newsfeed.stanford.edu!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!news2 Xref: chonsp.franklin.ch alt.folklore.computers:60727 In article <8lpa0t$jeb$1@bob.news.rcn.net> jmfbahciv@aol.com (jmfbahciv) writes: >In article , >Jim Thomas wrote: > >> "/BAH" == jmfbahciv writes: >> >> /BAH> Entry time wouldn't have taken time. Damn..I almost remembered >> /BAH> the column punch that started a dup field.. >> >>"0" - normally "/" for an alpha dup :-) > >I don't think I used alpha dup very often. It seems to me >that there was a side effect of numeric dup that would stop >the card if [something] occurred. A blank column. The machine would hang until you either pressed the alpha shift key, or said to hell with it and cleared the machine out (which was only an option on the 029 because the 026 didn't have the clear switch). That's why I always coded my dups as alpha ('/') - but then, I was usually punching programs, not numeric data where every column was supposed to be filled. -- cgibbs@sky.bus.com (Charlie Gibbs) Remove the first period after the "at" sign to reply. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Fri, 28 Jul 00 10:40:22 GMT Organization: UltraNet Communications, Inc. Lines: 34 Message-ID: <8ls2d4$423$2@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> <397B7B0A.F680ED9B@ev1.net> <5DNe5.35033$I7.719214@news-west.usenetserver.com> X-Trace: 73nwOwCmyPcoNl5etnqxJ/HPvQd1Pex+w/h1Bcs+ohs= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 28 Jul 2000 13:38:44 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!newsfeeds.belnet.be!news.belnet.be!newsfeed.icl.net!netnews.com!feed1.news.rcn.net!rcn!209-122-231-250 Xref: chonsp.franklin.ch alt.folklore.computers:60731 In article , Brian Inglis wrote: >On Mon, 24 Jul 2000 02:13:21 GMT, swaim@nol.net wrote: > >>Charles Richmond wrote: >>> IMHO the Oracle database has an excellent time/date storage facility. It >>> handles dates back to 2000-something B.C., and also way, way into the future. >>> This time/date storage should be made the ANSI standard format, and then >>> back-filled into VMS/UNIX/whatever...and the time/date problems will be >>> put to bed. It would make all this discussion of whose date runs out when >>> quite irrelevent. > >~4713BC (Julian day 0) through ~4713CE valid, but the data format >can store from 9900BC through 15400CE, using a one byte century >value held as excess 100, and excluding the 0 and 255 values for >conservative design reasons, IIRC. I've been a date/time fiend >since I was a kid. I thought Charles was talking about the nth day of the year; you're talking about the Julian date...not day. Or have I screwed those two up again. All I know is that recording things as the nth day of the year was a humongous PITA. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Fri, 28 Jul 00 10:37:16 GMT Organization: UltraNet Communications, Inc. Lines: 96 Message-ID: <8ls27d$423$1@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lk0ac$o2c$7@bob.news.rcn.net> X-Trace: 73nwOwCmyPcUDN/1HgmmEubwummTCdXMmFueorE7w0g= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 28 Jul 2000 13:35:41 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!netnews.com!feed1.news.rcn.net!rcn!209-122-231-250 Xref: chonsp.franklin.ch alt.folklore.computers:60732 In article , djb0x1312fd7@scream.org (Dan) wrote: >jmfbahciv@aol.com wrote: > >> >> File contents. Files written by user software that contain 32 bit date >> >> fields. The same problem an with 2 digit year numbers. Y2K tidy up >> >> code audits all over again. >> > >> >Ah, but... As long as the start of the epoch is 1/1/1970, wouldn't a >> >system with a 64-bit time_t still read a 32-bit "legacy" date, count >> >the appropriate number of seconds, and render the correct date? The >> >numbers themselves are not going to abruptly stop being valid, right? >> >> How are you going to know that number you have came from a >> 32-bit field rather than a 64-bit field _without setting some >> kind of flag that means "new format"_. And how many "new >> formats" can one have? > >Got me. :) A 32-bit value will fit into a 64-bit field, that's about >the extent of my firm grasp of the obvious. That makes a conversion sorta easier. How does a program decide to limit the bit pick up to 32 rather than 64? We ran into similar problems when the format went from a half word to a full word. Acutally, IIRC, the field went from 24 bits to 36 bits. > >> >Presuming, of course, that the system was coded in such a way that it >> >knew to Do The Right Thing if it encountered a 32-bit date. >> >> And that's a huge presumption when the coder is only working >> and debugging on a newer system and has never seen an old >> system. > >What the coder is doing on the system should not impact how the >system the coder was using is coded. But it does if the target system of the coder is the system he's using. I mean people do run their own code these days...[horrified emoticon squeaks in a very tiny voice]...don't they? > >Uh... time_t is a construct at the system library level in UNIX, >isn't it? Hopefully, as we approach 2038 (but the sooner the >better, if you ask me), the libraries and programming languages >will be updated to grok large numbers in time_t and deal with >them appropriately. That will only work if an existing executable can do a monitor call that says, "give me a date/time, and put it in this format for me". > >I don't know whether simply modifying the C library would fix it - >I'm not *that* good of a techie. Only if every single program was recompiled with the library that had the fix in it. In this day and age of "get the code from that site over there", I don't see how a program can determine that fixes are in without a lot of inline self testing. In order to have programmed inline self testing of formats, the programmer has to have had great ESP. We were good, but not that good. > If the C lib knew about 32-bit >and 64-bit time and could figure out either one, maybe recompiling >Perl would make that aware too. Or maybe Perl would need to be >modified internally to understand it, so that it wouldn't think >17:14:07 on January 18, 2038 was followed by 10:15:52 on December >13, 1901. But the libraries could know a hundred time formats. The key is how is the program who uses those libraries know _which_ format is getting handed to it? How does a program tell the difference between a 32-bit date and a 64-bit date when the data is handed to it in a 100-bit field...or, in the case of old programs, the field width defined was 16-bit. I'm slightly exagerrating here. > >But if anything beyond a recompile are required at the application >level, I'd say that somebody has gone and reinvented the wheel >instead of using the built-in functinos, and gotten what they >deserved. ;) See above. Doesn't anybody understand the _process_ of development and how important the process is? I used to think of this process as a bit flow and my job was to create the magic so that the bit flows of each and every developer in my group came together in one place and flowed out to our customers. Sort of like a symphony of musical notes. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Fri, 28 Jul 00 10:44:12 GMT Organization: UltraNet Communications, Inc. Lines: 41 Message-ID: <8ls2ka$423$3@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lk0ac$o2c$7@bob.news.rcn.net> X-Trace: Fa4bcROCkamJR06wAWbvgFJWdv/woAsAF3y0j5KapWo= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 28 Jul 2000 13:42:34 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!rz.uni-karlsruhe.de!schlund.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!diablo.theplanet.net!newsfeed.icl.net!netnews.com!feed1.news.rcn.net!rcn!209-122-231-250 Xref: chonsp.franklin.ch alt.folklore.computers:60736 In article , Brian Inglis wrote: >On Tue, 25 Jul 00 09:16:07 GMT, jmfbahciv@aol.com wrote: > >>In article , >> djb0x1312fca@scream.org (Dan) wrote: >>>Neil Franklin wrote: >>> >>>> File contents. Files written by user software that contain 32 bit date >>>> fields. The same problem an with 2 digit year numbers. Y2K tidy up >>>> code audits all over again. >>> >>>Ah, but... As long as the start of the epoch is 1/1/1970, wouldn't a >>>system with a 64-bit time_t still read a 32-bit "legacy" date, count >>>the appropriate number of seconds, and render the correct date? The >>>numbers themselves are not going to abruptly stop being valid, right? >> >>How are you going to know that number you have came from a >>32-bit field rather than a 64-bit field _without setting some >>kind of flag that means "new format"_. And how many "new >>formats" can one have? >>> >>>Presuming, of course, that the system was coded in such a way that it >>>knew to Do The Right Thing if it encountered a 32-bit date. >> >>And that's a huge presumption when the coder is only working >>and debugging on a newer system and has never seen an old >>system. > >And what system is he going to use when developing the new file >system code? And why will he be eating his own dog food until he >gets the new code with backward compatibility features working? You are assuming that he knows there was a different format. You are also assuming that the code he writes is not going to be run on a system his specification didn't include. (I think I've got my negatives correct in that last sentence.) /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Fri, 28 Jul 00 10:49:18 GMT Organization: UltraNet Communications, Inc. Lines: 44 Message-ID: <8ls2ts$423$4@bob.news.rcn.net> References: <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <8lksuj$qr7$1@news.iastate.edu> <8lmmov$l8b$7@bob.news.rcn.net> <8lpa0t$jeb$1@bob.news.rcn.net> <1044.243T2329T7345222@sky.bus.com> X-Trace: JFuwP8NR9etxz/HlKKiR7Ca0jqr1wjTW9CNGn3rJTHo= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 28 Jul 2000 13:47:40 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!netnews.com!feed1.news.rcn.net!rcn!209-122-231-250 Xref: chonsp.franklin.ch alt.folklore.computers:60737 In article <1044.243T2329T7345222@sky.bus.com>, "Charlie Gibbs" wrote: >In article <8lpa0t$jeb$1@bob.news.rcn.net> jmfbahciv@aol.com (jmfbahciv) >writes: > >>In article , >>Jim Thomas wrote: >> >>> "/BAH" == jmfbahciv writes: >>> >>> /BAH> Entry time wouldn't have taken time. Damn..I almost remembered >>> /BAH> the column punch that started a dup field.. >>> >>>"0" - normally "/" for an alpha dup :-) >> >>I don't think I used alpha dup very often. It seems to me >>that there was a side effect of numeric dup that would stop >>the card if [something] occurred. > >A blank column. The machine would hang until you either pressed >the alpha shift key, or said to hell with it and cleared the machine >out (which was only an option on the 029 because the 026 didn't have >the clear switch). That's why I always coded my dups as alpha ('/') - >but then, I was usually punching programs, not numeric data where >every column was supposed to be filled. Thanks. The only time I keyed in programs was at finals time (I got to key in all the kiddies' programs...except mine). And that was either FORTRAN II or SPS. Every once in a great while, one of the programmers would have me key in his code. Only the ones who wanted their code "reviewed" asked me; the ones who couldn't take a review didn't . Most of the keypunching that I did was all those bloody questionaires that the pcych types had filled out for their dissertations or the sociology types for their future book. I also did a Sunday job for the football team :-). But that only took 10 minutes so I usually did every thing by hand rather than a superduper designed drum card to help me catch keypunch errors. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Fri, 28 Jul 00 09:21:22 GMT Organization: UltraNet Communications, Inc. Lines: 48 Message-ID: <8lrtp0$gd$2@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8lepbo$gfh$5@bob.news.rcn.net> <8lh7qj$iui$1@bob.news.rcn.net> X-Trace: 1E2/v0hg5HuTq40m/pPkwyYMf6C5WcXFTtPiaecxXzI= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 28 Jul 2000 12:19:44 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!newsfeed.icl.net!netnews.com!feed1.news.rcn.net!rcn!207-172-255-178 Xref: chonsp.franklin.ch alt.folklore.computers:60738 In article , kragen@dnaco.net (Kragen Sitaker) wrote: >In article <8lh7qj$iui$1@bob.news.rcn.net>, wrote: >>In article , >> kragen@dnaco.net (Kragen Sitaker) wrote: >>>Unix filesystems don't have files created, last modified, or last >>>accessed before 1970, because no Unix filesystems created before 1970 >>>still exist. >> >>Sigh! Are you saying that Unix will not let me access a file >>that was created under another OS before 1970? I don't believe >>that. > >Well, if you're interpreting time_t in a strict sense, in which >negative numbers are undefined, you won't be able to create files on a >Unix filesystem with those dates. In practice, Unix works fine with >negative times, and so as long as the file dates are 1902 or later, >you'd be OK. Can you imaging the threads in this newsgroup 50 years from now wondering why in the world that old and crufty OS had code to deal with negative dates? > >It's not a matter of not letting you access things, but of not having >any way to store their dates. It's an old habit that, everytime you write a date, you write it in the "new" format. That's how we used attrition to convert from one architecture to another. But I disagree with you; the real problem is a matter of being able to read (or convert) the old date format. Think about newbies who never encountered it in their spanking-new hard/software systems and then go to work for the Federal government. > >By the way, I'm happy to see you back on my news server :) Damn! This bug has me driving up the wall, around the ceiling and back down again. My posts in this newsgroup just started appearing from the AOL side, too. And they moved the path from one node to another with no indication of what the problem was. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Fri, 28 Jul 00 09:13:16 GMT Organization: UltraNet Communications, Inc. Lines: 80 Message-ID: <8lrt9t$gd$1@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <8lh601$orq$1@flywheel.innovative.iinet.net.au> <8lh895$iui$3@bob.news.rcn.net> X-Trace: 1DSDHbsyQYzg5J/cj5BO/34Lhm3MjeueZ0ycoyZ/vlA= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 28 Jul 2000 12:11:41 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-255-178 Xref: chonsp.franklin.ch alt.folklore.computers:60741 In article , Brian Inglis wrote: >On Mon, 24 Jul 00 08:13:46 GMT, jmfbahciv@aol.com wrote: > >>In article <8lh601$orq$1@flywheel.innovative.iinet.net.au>, >> bernie@innovative.iinet.net.au (Bernd Felsche) wrote: >>>kragen@dnaco.net (Kragen Sitaker) writes: >>> >>>>In article <8lepbo$gfh$5@bob.news.rcn.net>, wrote: >>>>>In article <964298714.421191@shelley.paradise.net.nz>, >>>>> don@news.daedalus.co.nz (Don Stokes) wrote: >>>>>>Well, time_t is a *signed* 32 bit int. For file systems, barring >>>>>>idiocy, there aren't any file dates < 1970, >>>>> >>>>> Since I started contributing to social security >>>>>in 1962, I would certainly hope that there is software that >>>>>can read file dates before 1970. >>> >>>>Unix filesystems don't have files created, last modified, or last >>>>accessed before 1970, because no Unix filesystems created before 1970 >>>>still exist. >>> >>>Try this on Linux/Unix: :-) >>> $ TZ=GMT-1 touch -t 195809102030 born >>> $ ls -l born >>> -rw------- 1 bernie users 0 Sep 11 1958 born >>> >>>Never assume that somebody won't exploit the signed nature of the >>>timestamp. >>> >>>Old filesystems should be OK to mount read-only at any point in the >>>future. The filesystem type identified by the superblock tells you >>>about the structure anyway so you can make allowances for antique >>>media. >> >>Since you're insisting that the old formatted media be mounted >>read-only, how in the world are you going to mark it with a new >>datum type (assuming that this datum gets defined as part of >>a new file system design). > >I think that the intent is to mark the new file systems with a >new type/version indicating 64 bit times, as happened when drive >and file sizes were increased from 32 to 64 bits, but have the fs >code recognize and adapt to the older version for compatibility. That's [64 bits] at least a "second" version; so now there are 3 file system architectures. In this case, I could imagine that TW would want a bit to be trinary(sp?). So now there has to be at least two bits to define which format the structure is writtin in. This is the old problem of how much and how long backwards compatibility is necessary. > >>Don't these files have last accessed date/time? I can't begin >>to imagine not having that very important piece of information >>for debugging analysis. > >What did TOPS do when you mounted a pack read only or with write >disabled if the drive had that switch? I suspect that it totally >avoided writing to the pack, like most file systems since. I'm trying to remember. There were cases when we did not want anything to be touched on a pack...especially when PCOPYing packs for SDC was done. It certainly is true that the last accessed date of the MFD RIB had to have been modified at mount time (she says without looking at the code to verify it ;-)). And I can't remember how BACKUP restored files. If the last accessed date was the date/time of restoration or the date/time that was saved to tape. However we did it, it was the right way to handle those date/time stamps. The logic of how the last accessed date/time was not straight forward. Actually, in my years of USAGE accounting and BACKUP maintenance, date/time stamps were never straight forward :-). /BAH Subtract a hundred and four for e-mail. ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8lepbo$gfh$5@bob.news.rcn.net> <8lh7qj$iui$1@bob.news.rcn.net> Organization: None X-Newsreader: trn 4.0-test72 (19 April 1999) From: kragen@dnaco.net (Kragen Sitaker) Lines: 26 Message-ID: X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Thu, 27 Jul 2000 22:26:15 EDT Date: Fri, 28 Jul 2000 02:26:15 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!rz.uni-karlsruhe.de!schlund.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!portc01.blue.aol.com!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!news-west.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60778 In article <8lh7qj$iui$1@bob.news.rcn.net>, wrote: >In article , > kragen@dnaco.net (Kragen Sitaker) wrote: >>Unix filesystems don't have files created, last modified, or last >>accessed before 1970, because no Unix filesystems created before 1970 >>still exist. > >Sigh! Are you saying that Unix will not let me access a file >that was created under another OS before 1970? I don't believe >that. Well, if you're interpreting time_t in a strict sense, in which negative numbers are undefined, you won't be able to create files on a Unix filesystem with those dates. In practice, Unix works fine with negative times, and so as long as the file dates are 1902 or later, you'd be OK. It's not a matter of not letting you access things, but of not having any way to store their dates. By the way, I'm happy to see you back on my news server :) -- Kragen Sitaker Perilous to all of us are the devices of an art deeper than we ourselves possess. -- Gandalf the Grey [J.R.R. Tolkien, "Lord of the Rings"] ###### From: djb0x1312fd7@scream.org (Dan) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 28 Jul 2000 06:09:13 GMT Lines: 54 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lk0ac$o2c$7@bob.news.rcn.net> Reply-To: djb0x1312fd7@scream.org NNTP-Posting-Host: v2000.scream.org X-Trace: mochi.lava.net 964764553 28194 198.151.213.3 (28 Jul 2000 06:09:13 GMT) X-Complaints-To: abuse@lava.net NNTP-Posting-Date: 28 Jul 2000 06:09:13 GMT User-Agent: slrn/0.9.6.2 (Linux) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!rz.uni-karlsruhe.de!schlund.de!newsfeed01.sul.t-online.de!newsfeed00.sul.t-online.de!t-online.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news.he.net!news.lava.net!djb0x1312fd7 Xref: chonsp.franklin.ch alt.folklore.computers:60771 jmfbahciv@aol.com wrote: > >> File contents. Files written by user software that contain 32 bit date > >> fields. The same problem an with 2 digit year numbers. Y2K tidy up > >> code audits all over again. > > > >Ah, but... As long as the start of the epoch is 1/1/1970, wouldn't a > >system with a 64-bit time_t still read a 32-bit "legacy" date, count > >the appropriate number of seconds, and render the correct date? The > >numbers themselves are not going to abruptly stop being valid, right? > > How are you going to know that number you have came from a > 32-bit field rather than a 64-bit field _without setting some > kind of flag that means "new format"_. And how many "new > formats" can one have? Got me. :) A 32-bit value will fit into a 64-bit field, that's about the extent of my firm grasp of the obvious. > >Presuming, of course, that the system was coded in such a way that it > >knew to Do The Right Thing if it encountered a 32-bit date. > > And that's a huge presumption when the coder is only working > and debugging on a newer system and has never seen an old > system. What the coder is doing on the system should not impact how the system the coder was using is coded. Uh... time_t is a construct at the system library level in UNIX, isn't it? Hopefully, as we approach 2038 (but the sooner the better, if you ask me), the libraries and programming languages will be updated to grok large numbers in time_t and deal with them appropriately. I don't know whether simply modifying the C library would fix it - I'm not *that* good of a techie. If the C lib knew about 32-bit and 64-bit time and could figure out either one, maybe recompiling Perl would make that aware too. Or maybe Perl would need to be modified internally to understand it, so that it wouldn't think 17:14:07 on January 18, 2038 was followed by 10:15:52 on December 13, 1901. But if anything beyond a recompile are required at the application level, I'd say that somebody has gone and reinvented the wheel instead of using the built-in functinos, and gotten what they deserved. ;) -Dan -- Dan Birchall - Palolo Valley, Honolulu HI - http://dan.scream.org Post your reviews; get paid: http://epinions.scream.org/join.html My address expires - take out the hex stamp if your reply bounces ###### From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Fri, 28 Jul 2000 00:12:43 -0600 Organization: Systematic Software Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <964099762.86210@shelley.paradise.net.nz> <397711EB.CC1C5723@netinsight.se> <397776d6$1@news.ucsc.edu> <964148894.748064@shelley.paradise.net.nz> <48A6A565160289F7.B6AB1B2AC2E18C44.8D36189F69B3CB75@lp.airnews.net> <441EF078B6B19400.7B2F5BFE6C01266F.68C4AF216247D0AA@lp.airnews.net> <82mnnsk34vsng2u857c46h5lb5br0q5k1i@4ax.com> <8lmm7i$l8b$6@bob.news.rcn.net> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.148.136.153 X-Original-NNTP-Posting-Host: 207.148.136.153 X-Trace: 28 Jul 2000 00:12:44 -0700, 207.148.136.153 Lines: 22 X-Original-NNTP-Posting-Host: 204.50.1.43 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-koe1.dfn.de!news-was.dfn.de!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.flash.net!nntp.cadvision.com!news.cadvision.com!207.148.136.153 Xref: chonsp.franklin.ch alt.folklore.computers:60780 On Wed, 26 Jul 00 09:42:11 GMT, jmfbahciv@aol.com wrote: >In article <82mnnsk34vsng2u857c46h5lb5br0q5k1i@4ax.com>, > Brian Inglis wrote: >>On 22 Jul 2000 02:16:21 GMT, jmaynard@thebrain.conmicro.cx (Jay >>Maynard) wrote: >> >>>On Sat, 22 Jul 2000 01:54:35 GMT, Ric Werme >wrote: > Definition request. What's an "escape tape"? IBM MF talk for a tape to which you copy your favourite OS patches, scripts, sources, etc. before escaping your current job. >/BAH > >Subtract a hundred and four for e-mail. Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Fri, 28 Jul 2000 00:12:45 -0600 Organization: Systematic Software Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <8lh601$orq$1@flywheel.innovative.iinet.net.au> <8lh895$iui$3@bob.news.rcn.net> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.148.136.153 X-Original-NNTP-Posting-Host: 207.148.136.153 X-Trace: 28 Jul 2000 00:12:46 -0700, 207.148.136.153 Lines: 58 X-Original-NNTP-Posting-Host: 204.50.1.43 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-mue1.dfn.de!news.augsburg.net!news.idt.net!nntp.cadvision.com!news.cadvision.com!207.148.136.153 Xref: chonsp.franklin.ch alt.folklore.computers:60779 On Mon, 24 Jul 00 08:13:46 GMT, jmfbahciv@aol.com wrote: >In article <8lh601$orq$1@flywheel.innovative.iinet.net.au>, > bernie@innovative.iinet.net.au (Bernd Felsche) wrote: >>kragen@dnaco.net (Kragen Sitaker) writes: >> >>>In article <8lepbo$gfh$5@bob.news.rcn.net>, wrote: >>>>In article <964298714.421191@shelley.paradise.net.nz>, >>>> don@news.daedalus.co.nz (Don Stokes) wrote: >>>>>Well, time_t is a *signed* 32 bit int. For file systems, barring >>>>>idiocy, there aren't any file dates < 1970, >>>> >>>> Since I started contributing to social security >>>>in 1962, I would certainly hope that there is software that >>>>can read file dates before 1970. >> >>>Unix filesystems don't have files created, last modified, or last >>>accessed before 1970, because no Unix filesystems created before 1970 >>>still exist. >> >>Try this on Linux/Unix: :-) >> $ TZ=GMT-1 touch -t 195809102030 born >> $ ls -l born >> -rw------- 1 bernie users 0 Sep 11 1958 born >> >>Never assume that somebody won't exploit the signed nature of the >>timestamp. >> >>Old filesystems should be OK to mount read-only at any point in the >>future. The filesystem type identified by the superblock tells you >>about the structure anyway so you can make allowances for antique >>media. > >Since you're insisting that the old formatted media be mounted >read-only, how in the world are you going to mark it with a new >datum type (assuming that this datum gets defined as part of >a new file system design). I think that the intent is to mark the new file systems with a new type/version indicating 64 bit times, as happened when drive and file sizes were increased from 32 to 64 bits, but have the fs code recognize and adapt to the older version for compatibility. >Don't these files have last accessed date/time? I can't begin >to imagine not having that very important piece of information >for debugging analysis. What did TOPS do when you mounted a pack read only or with write disabled if the drive had that switch? I suspect that it totally avoided writing to the pack, like most file systems since. [snip] >/BAH Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Fri, 28 Jul 2000 00:12:46 -0600 Organization: Systematic Software Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> <397B7B0A.F680ED9B@ev1.net> <5DNe5.35033$I7.719214@news-west.usenetserver.com> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.148.136.153 X-Original-NNTP-Posting-Host: 207.148.136.153 X-Trace: 28 Jul 2000 00:12:47 -0700, 207.148.136.153 Lines: 31 X-Original-NNTP-Posting-Host: 204.50.1.43 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news0.de.colt.net!colt.net!newsfeed.icl.net!newsfeed.axxsys.net!news.idt.net!nntp.cadvision.com!news.cadvision.com!207.148.136.153 Xref: chonsp.franklin.ch alt.folklore.computers:60773 On Mon, 24 Jul 2000 02:13:21 GMT, swaim@nol.net wrote: >Charles Richmond wrote: >> IMHO the Oracle database has an excellent time/date storage facility. It >> handles dates back to 2000-something B.C., and also way, way into the future. >> This time/date storage should be made the ANSI standard format, and then >> back-filled into VMS/UNIX/whatever...and the time/date problems will be >> put to bed. It would make all this discussion of whose date runs out when >> quite irrelevent. ~4713BC (Julian day 0) through ~4713CE valid, but the data format can store from 9900BC through 15400CE, using a one byte century value held as excess 100, and excluding the 0 and 255 values for conservative design reasons, IIRC. I've been a date/time fiend since I was a kid. > Oddly enough, all of the Y2K problems I encountered involved Oracle and >dates. By default, Oracle rejects years with 4 digits. (We tried going 4 >digits with one of our databases, but that broke the links to all of the >other databases. Sigh.) Kill the damn dumb DBA who setup the Oracle DB with the default date format! The default date format can also be changed in the user's environment, the application environment and at the database session level. It's as easy as changing TZ in a Unix session. Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Fri, 28 Jul 2000 00:12:47 -0600 Organization: Systematic Software Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lk0ac$o2c$7@bob.news.rcn.net> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.148.136.153 X-Original-NNTP-Posting-Host: 207.148.136.153 X-Trace: 28 Jul 2000 00:12:48 -0700, 207.148.136.153 Lines: 36 X-Original-NNTP-Posting-Host: 204.50.1.43 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!news.idt.net!nntp.cadvision.com!news.cadvision.com!207.148.136.153 Xref: chonsp.franklin.ch alt.folklore.computers:60774 On Tue, 25 Jul 00 09:16:07 GMT, jmfbahciv@aol.com wrote: >In article , > djb0x1312fca@scream.org (Dan) wrote: >>Neil Franklin wrote: >> >>> File contents. Files written by user software that contain 32 bit date >>> fields. The same problem an with 2 digit year numbers. Y2K tidy up >>> code audits all over again. >> >>Ah, but... As long as the start of the epoch is 1/1/1970, wouldn't a >>system with a 64-bit time_t still read a 32-bit "legacy" date, count >>the appropriate number of seconds, and render the correct date? The >>numbers themselves are not going to abruptly stop being valid, right? > >How are you going to know that number you have came from a >32-bit field rather than a 64-bit field _without setting some >kind of flag that means "new format"_. And how many "new >formats" can one have? >> >>Presuming, of course, that the system was coded in such a way that it >>knew to Do The Right Thing if it encountered a 32-bit date. > >And that's a huge presumption when the coder is only working >and debugging on a newer system and has never seen an old >system. And what system is he going to use when developing the new file system code? And why will he be eating his own dog food until he gets the new code with backward compatibility features working? >/BAH Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### From: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 28 Jul 2000 15:25:30 GMT Organization: The National Capital FreeNet Lines: 12 Message-ID: <8ls8la$889$1@freenet9.carleton.ca> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lk0ac$o2c$7@bob.news.rcn.net> <8ls27d$423$1@bob.news.rcn.net> Reply-To: ab528@FreeNet.Carleton.CA (Heinz W. Wiggeshoff) NNTP-Posting-Host: freenet10 X-Trace: freenet9.carleton.ca 964797930 8457 134.117.136.30 (28 Jul 2000 15:25:30 GMT) X-Complaints-To: complaints@ncf.ca NNTP-Posting-Date: 28 Jul 2000 15:25:30 GMT X-Given-Sender: ab528@freenet10.carleton.ca (Heinz W. Wiggeshoff) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news-stu1.dfn.de!news-mue1.dfn.de!news.augsburg.net!nntp.primenet.com!nntp.frontiernet.net!nntp.gblx.net!xcski.com!freenet-news!FreeNet.Carleton.CA!ab528 Xref: chonsp.franklin.ch alt.folklore.computers:60754 (jmfbahciv@aol.com) writes: > ... > See above. Doesn't anybody understand the _process_ of > development and how important the process is? I used to > think of this process as a bit flow and my job was to > create the magic so that the bit flows of each and > every developer in my group came together in one place > and flowed out to our customers. Sort of like a symphony > of musical notes. Down to a few stems and seeds, are we? B-) ###### From: rhawkins@iastate.edu (Rick Hawkins) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 28 Jul 2000 22:15:59 GMT Organization: House of Hawkins Lines: 68 Message-ID: <8lt0mv$q5q$1@news.iastate.edu> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8lepbo$gfh$5@bob.news.rcn.net> <8lksuj$qr7$1@news.iastate.edu> <8lmmov$l8b$7@bob.news.rcn.net> NNTP-Posting-Host: pv2087.vincent.iastate.edu Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!logbridge.uoregon.edu!newsrelay.iastate.edu!news.iastate.edu!rhawkins Xref: chonsp.franklin.ch alt.folklore.computers:60801 In article <8lmmov$l8b$7@bob.news.rcn.net>, wrote: >In article <8lksuj$qr7$1@news.iastate.edu>, > rhawkins@iastate.edu (Rick Hawkins) wrote: >>> Since I started contributing to social security >>>in 1962, I would certainly hope that there is software that >>>can read file dates before 1970. >>*giggle* >>Yikes, BAH, you must be *old* if you're expecting your social security >>contributions to make a difference :) >Not the contributions...the fact that I did them. >That's going to be key rather than amounts. That's what I mean about being old :) In my case, I expect it to make a difference between a monthly check of $0, and a monthly check of $0. Assuming that SS still even exists when I retire, I assume that it will be means-tested to punish those of us who save for our own retirement. >>>It's thinking like this that caused all the Y2K expenditures. >>However, someone did an analysis on what it would have cost to >>*avoid* the Y2K changes, and found that that it dwarfed the cleanup >>costs . . . those extra digits (and entry time) added up . . . >Entry time wouldn't have taken time. Damn..I almost remembered >the column punch that started a dup field.. I'm thinking in terms of the number of spots on a card, and the possibility of being forced onto a second card for keeping those extra to digits-- the exception, but I assume it would have occasionally happened. >But you are right. What I was talking about was the guy just >making a decision without knowing the consequences. Most of >the people I worked with, who made those 2-digit vs. 4-digit >decisions, actually spent time thinking about the pros and cons >of not using a complete data field. What? Think first? What a concept . . . :) >>hawk, who doesn't agree that contributing to social security is equivalent >>to flushing the money down the toilet--at least that way he gets >>to watch it spin . . . >But that's to prevent you from getting dizzy, Hawk. The >government is there to protect you. Awe, isn't that sweet of them. And on that point, I finally heard the algore say something I agreed with--he called the other guy's proposal to allow part (2% of payroll) of the SS contribution to be invested. Quite reckless; it still leaves the gub'mnt 12% :) hawk, who for some reason doubts that that's what he *really* meant by the comment . . . >I guess I'd better put a grinning emoticon here before those >nuts decide I'm making a serious statement. ~ -- Richard E. Hawkins dochawk@psu.edu [regardless of where the message says it comes from] These opinions will not be those of Penn State until they pay my retainer. ###### From: Eric Fischer Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 29 Jul 2000 20:10:49 GMT Organization: EnterAct Corp. Lines: 9 Message-ID: <8lvdo9$l5$1@news.enteract.com> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> <6ubszohc1s.fsf@chonsp.franklin.ch> NNTP-Posting-Host: shell-3.enteract.com X-Trace: news.enteract.com 964901449 677 207.229.143.42 (29 Jul 2000 20:10:49 GMT) X-Complaints-To: abuse@enteract.com NNTP-Posting-Date: 29 Jul 2000 20:10:49 GMT X-Newsreader: trn 4.0-test72 (19 April 1999) Originator: enf@enteract.com (Eric Fischer) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!netnews.com!newsfeed.enteract.com!news.enteract.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60833 Neil Franklin wrote: > time_t is _signed_ int. Says who? I believe ISO/ANSI C says it can be any arithmetic type. POSIX specifically mentions "implementations in which time_t is a 32-bit unsigned integer." (IEEE 1003.1-1990, p. 249) eric ###### From: "Tracy Nelson" Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Sat, 29 Jul 2000 18:28:52 -0400 Organization: Posted via Supernews, http://www.supernews.com Lines: 14 Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8kv4bv$ju4$1@news.kersur.net> <3973B271.763C405@prescienttech.com> <3974935b$0$4933@reader3> <8l26rp$1fpk$1@ausnews.austin.ibm.com> X-Complaints-To: newsabuse@supernews.com X-Newsreader: Microsoft Outlook Express 4.72.3155.0 X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3155.0 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!feeder.qis.net!sn-xit-01!supernews.com!sn-inject-01!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:60845 ISTR that one computer is constantly running the re-entry program, so that even if the three primaries go down, they can still get home. Perhaps that's the "different version of the flight control program" you remembered. >Another is running a different version of the flight control >program, and can replace the others in the event of a software >error. A fifth one is powered up, and has the flight control >program loaded, but not running (e.g., hot standby). The sixth one >is unpowered, but can be installed to replace a failing system once >the shuttle is on orbit. Of course, all of this is from fuzzy >memory, and may not be correct, or the details may have changed. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Mon, 31 Jul 00 09:05:58 GMT Organization: UltraNet Communications, Inc. Lines: 22 Message-ID: <8m3q11$21o$1@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8lepbo$gfh$5@bob.news.rcn.net> <8lksuj$qr7$1@news.iastate.edu> <8lmmov$l8b$7@bob.news.rcn.net> <8lt0mv$q5q$1@news.iastate.edu> X-Trace: 4bk9IAZgQQzx5C0hprz8tGJraYNhuvNcjsebGFWGhVM= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 31 Jul 2000 12:04:49 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!newsfeeds.belnet.be!news.belnet.be!xfer13.netnews.com!xfe11.netnews.com!netnews.com!feed1.news.rcn.net!rcn!207-172-102-222 Xref: chonsp.franklin.ch alt.folklore.computers:60875 In article <8lt0mv$q5q$1@news.iastate.edu>, rhawkins@iastate.edu (Rick Hawkins) wrote: >In article <8lmmov$l8b$7@bob.news.rcn.net>, wrote: >>In article <8lksuj$qr7$1@news.iastate.edu>, >> rhawkins@iastate.edu (Rick Hawkins) wrote: >>But you are right. What I was talking about was the guy just >>making a decision without knowing the consequences. Most of >>the people I worked with, who made those 2-digit vs. 4-digit >>decisions, actually spent time thinking about the pros and cons >>of not using a complete data field. > >What? Think first? What a concept . . . :) Yup. I have been told, often, that I worked with an elite bunch and that those guys were not the norm. /BAH Subtract a hundred and four for e-mail. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Mon, 31 Jul 00 10:56:12 GMT Organization: UltraNet Communications, Inc. Lines: 109 Message-ID: <8m40fo$osp$4@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <8lh601$orq$1@flywheel.innovative.iinet.net.au> <8lh895$iui$3@bob.news.rcn.net> <8lrt9t$gd$1@bob.news.rcn.net> X-Trace: pRsS0ognMZ4qFWcIOf8Ivg4OAT1yYq8jBl52X9l0I9c= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 31 Jul 2000 13:55:04 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!logbridge.uoregon.edu!feed1.news.rcn.net!rcn!216-164-248-147 Xref: chonsp.franklin.ch alt.folklore.computers:60883 In article , Brian Inglis wrote: >On Fri, 28 Jul 00 09:13:16 GMT, jmfbahciv@aol.com wrote: > >>In article , >> Brian Inglis wrote: >>>On Mon, 24 Jul 00 08:13:46 GMT, jmfbahciv@aol.com wrote: >>> >>>>In article <8lh601$orq$1@flywheel.innovative.iinet.net.au>, >>>> bernie@innovative.iinet.net.au (Bernd Felsche) wrote: >>>>>kragen@dnaco.net (Kragen Sitaker) writes: >>>>> >>>>>>In article <8lepbo$gfh$5@bob.news.rcn.net>, wrote: >>>>>>>In article <964298714.421191@shelley.paradise.net.nz>, >>>>>>> don@news.daedalus.co.nz (Don Stokes) wrote: >>>>>>>>Well, time_t is a *signed* 32 bit int. For file systems, barring >>>>>>>>idiocy, there aren't any file dates < 1970, >>>>>>> >>>>>>> Since I started contributing to social security >>>>>>>in 1962, I would certainly hope that there is software that >>>>>>>can read file dates before 1970. >>>>> >>>>>>Unix filesystems don't have files created, last modified, or last >>>>>>accessed before 1970, because no Unix filesystems created before 1970 >>>>>>still exist. >>>>> >>>>>Try this on Linux/Unix: :-) >>>>> $ TZ=GMT-1 touch -t 195809102030 born >>>>> $ ls -l born >>>>> -rw------- 1 bernie users 0 Sep 11 1958 born >>>>> >>>>>Never assume that somebody won't exploit the signed nature of the >>>>>timestamp. >>>>> >>>>>Old filesystems should be OK to mount read-only at any point in the >>>>>future. The filesystem type identified by the superblock tells you >>>>>about the structure anyway so you can make allowances for antique >>>>>media. >>>> >>>>Since you're insisting that the old formatted media be mounted >>>>read-only, how in the world are you going to mark it with a new >>>>datum type (assuming that this datum gets defined as part of >>>>a new file system design). >>> >>>I think that the intent is to mark the new file systems with a >>>new type/version indicating 64 bit times, as happened when drive >>>and file sizes were increased from 32 to 64 bits, but have the fs >>>code recognize and adapt to the older version for compatibility. >> >>That's [64 bits] at least a "second" version; so now there >>are 3 file system architectures. In this case, I could imagine >>that TW would want a bit to be trinary(sp?). So now there has >>to be at least two bits to define which format the structure is >>writtin in. This is the old problem of how much and how long >>backwards compatibility is necessary. > >Unix filesystems (always?) have a 32 bit word identifing the >filesystem type and version number -- no need to hunt for bits. Only if the people who are designing honor that standard. Look, TOPS-10 had standards and rules for increasing the numbers in the categories of major version, minor version, edit, customer version, etc. The last TOPS-10 release by Digital broke those rules. You seem to believe that those formats were true back in the olden days, and will remain true forever. Uh, uh. If we couldn't keep it right, I don't see how developers spread all over the planet and all over different implementations will keep it right without a virtual baseball bat. > >The type and version are normally four ascii characters -- a >three character filesystem type and a one character version. > >Versions do not change very often: 64 bit sizes probably bumped >versions up to 2 or 3 and 64 bit times may bump them up to 3 or >4. Many Unix systems still support common older filesystem types. > >I don't know if the Unix OS 1st edition supported 16 or 32 bit >sizes and dates, but the 6th edition in Lions (1976) supported 24 >bit sizes (byte not block counts) and 32 bit times (seconds) on >PDP-11s with RK05s (see Lions listing lines 5605-5675). > >IMHO systems should support as much backward compatibility as >possible as long as possible. If you hit enough roadblocks in >later versions, it may be time to drop support for the older >stuff. Physical device support is what tends to get dropped as >technology changes: who needs card or paper tape reader/punch >drivers nowadays, or non-IDE/SCSI disks/tapes, excluding >floppies? Anybody who has archives. Historians are not going to have any source material on any subject since we assumed paper was passe. People who like to use census data to study trends. And those are just the people I can think of...there certainly are going to be people who will have uses for all this old stuff that we can't begin to anticipate. > OTOH, it may be reasonable to retain the use of an old >(small) filesystem type on small devices like floppies. And there >is fairly widespread support for various versions of MS FAT FS, >OS/2 HPFS, Win NT NTFS. All this is PC thinking. I'm beginning to think it's now a disease. /BAH Subtract a hundred and four for e-mail. ###### From: Brian Inglis Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Sun, 30 Jul 2000 23:24:50 -0600 Organization: Systematic Software Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca Message-ID: References: <3972935c$0$8328@wodc7nh6.news.uu.net> <3979B287.3EA2B030@trailing-edge.com> <964298714.421191@shelley.paradise.net.nz> <8lepbo$gfh$5@bob.news.rcn.net> <8lh601$orq$1@flywheel.innovative.iinet.net.au> <8lh895$iui$3@bob.news.rcn.net> <8lrt9t$gd$1@bob.news.rcn.net> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.148.143.108 X-Original-NNTP-Posting-Host: 207.148.143.108 X-Trace: 30 Jul 2000 23:24:53 -0700, 207.148.143.108 Lines: 86 X-Original-NNTP-Posting-Host: 204.50.1.43 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!sunqbc.risq.qc.ca!nntp.cadvision.com!news.cadvision.com!207.148.143.108 Xref: chonsp.franklin.ch alt.folklore.computers:60905 On Fri, 28 Jul 00 09:13:16 GMT, jmfbahciv@aol.com wrote: >In article , > Brian Inglis wrote: >>On Mon, 24 Jul 00 08:13:46 GMT, jmfbahciv@aol.com wrote: >> >>>In article <8lh601$orq$1@flywheel.innovative.iinet.net.au>, >>> bernie@innovative.iinet.net.au (Bernd Felsche) wrote: >>>>kragen@dnaco.net (Kragen Sitaker) writes: >>>> >>>>>In article <8lepbo$gfh$5@bob.news.rcn.net>, wrote: >>>>>>In article <964298714.421191@shelley.paradise.net.nz>, >>>>>> don@news.daedalus.co.nz (Don Stokes) wrote: >>>>>>>Well, time_t is a *signed* 32 bit int. For file systems, barring >>>>>>>idiocy, there aren't any file dates < 1970, >>>>>> >>>>>> Since I started contributing to social security >>>>>>in 1962, I would certainly hope that there is software that >>>>>>can read file dates before 1970. >>>> >>>>>Unix filesystems don't have files created, last modified, or last >>>>>accessed before 1970, because no Unix filesystems created before 1970 >>>>>still exist. >>>> >>>>Try this on Linux/Unix: :-) >>>> $ TZ=GMT-1 touch -t 195809102030 born >>>> $ ls -l born >>>> -rw------- 1 bernie users 0 Sep 11 1958 born >>>> >>>>Never assume that somebody won't exploit the signed nature of the >>>>timestamp. >>>> >>>>Old filesystems should be OK to mount read-only at any point in the >>>>future. The filesystem type identified by the superblock tells you >>>>about the structure anyway so you can make allowances for antique >>>>media. >>> >>>Since you're insisting that the old formatted media be mounted >>>read-only, how in the world are you going to mark it with a new >>>datum type (assuming that this datum gets defined as part of >>>a new file system design). >> >>I think that the intent is to mark the new file systems with a >>new type/version indicating 64 bit times, as happened when drive >>and file sizes were increased from 32 to 64 bits, but have the fs >>code recognize and adapt to the older version for compatibility. > >That's [64 bits] at least a "second" version; so now there >are 3 file system architectures. In this case, I could imagine >that TW would want a bit to be trinary(sp?). So now there has >to be at least two bits to define which format the structure is >writtin in. This is the old problem of how much and how long >backwards compatibility is necessary. Unix filesystems (always?) have a 32 bit word identifing the filesystem type and version number -- no need to hunt for bits. The type and version are normally four ascii characters -- a three character filesystem type and a one character version. Versions do not change very often: 64 bit sizes probably bumped versions up to 2 or 3 and 64 bit times may bump them up to 3 or 4. Many Unix systems still support common older filesystem types. I don't know if the Unix OS 1st edition supported 16 or 32 bit sizes and dates, but the 6th edition in Lions (1976) supported 24 bit sizes (byte not block counts) and 32 bit times (seconds) on PDP-11s with RK05s (see Lions listing lines 5605-5675). IMHO systems should support as much backward compatibility as possible as long as possible. If you hit enough roadblocks in later versions, it may be time to drop support for the older stuff. Physical device support is what tends to get dropped as technology changes: who needs card or paper tape reader/punch drivers nowadays, or non-IDE/SCSI disks/tapes, excluding floppies? OTOH, it may be reasonable to retain the use of an old (small) filesystem type on small devices like floppies. And there is fairly widespread support for various versions of MS FAT FS, OS/2 HPFS, Win NT NTFS. [snip] >/BAH Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### Newsgroups: alt.folklore.computers Subject: Re: Longest running computer References: <3972935c$0$8328@wodc7nh6.news.uu.net> <39784803.44BB656@trailing-edge.com> <397dd579$1@news.ucsc.edu> <398018C6.9354C3A7@netinsight.se> Organization: UC Santa Cruz CIS/CE From: eugene@cse.ucsc.edu (Eugene Miya) NNTP-Posting-Host: sundance.cse.ucsc.edu Message-ID: <39861602$1@news.ucsc.edu> Date: 31 Jul 2000 17:12:50 -0800 X-Trace: 31 Jul 2000 17:12:50 -0800, sundance.cse.ucsc.edu Lines: 54 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!logbridge.uoregon.edu!newsfeed.berkeley.edu!ucberkeley!news.ucsc.edu!eugene Xref: chonsp.franklin.ch alt.folklore.computers:60955 In article <398018C6.9354C3A7@netinsight.se>, Johnny Billquist wrote: >Eugene Miya wrote: >> In article <39784803.44BB656@trailing-edge.com>, >> Tim Shoppa wrote: >> >No, Compaq did not change this; the granularity always has been, >> >and will probably remain, 100 nanoseconds inside VMS. See the famous >> >Y10K SPR at >> > >> > http://www.trailing-edge.com/www/y10k.txt >> >> Are you telling me that 11/780s had 0.1 ms clocks? >> If no, when did the change occur? > >You are confusing two things here, Eugene. *VMS* have a 100 ns >granularity clock. I know about the interrupt granularity (the hardware) from the VMS (software) formatting issues. >The VAX-11/780 only used a 10ms interrupt interval, which meant >that the VMS clock was incremented 100 for each interrupt. Roughly. >Thus, on the -11/780, you could specify an interval as short as 100 ns, >but the OS never guarantees you that *exactly* that interval have passed, >only that *atleast* that interval passed. And the hardware interrupts >occuring at 10 ms intervals meant that you'd on average wait 5 ms, >even if you specified 100 ns. > >But that's the life on any OS you are using, only that most don't >have such a fine granularity of time in the OS. This I have learned. UNICOS does not suffer from this problem. Nor ConvexOS. >Of course, you should specify exactly the time you do want to wait, >since you might come across a system sometime in the future which >actually do interrupt at 100 ns intervals, at which time the >VMS time value would be incremented by 1 on each interrupt. > >So, VMS have *always* had 100 ns granularity on *all* time values, >and will likely always have. All SYS$calls taking time values think >that they are expressed in 100 ns increments, and that's the way >it has been since VMS 1.0. > >Now, do we understand each other? :-) It's a headache which still doesn't solve my problems, and I can only hope that VMS passes away more quickly. I'll leave out the DEC reference paper on machine speed variability. ###### From: jmfbahciv@aol.com Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: Thu, 03 Aug 00 11:05:16 GMT Organization: UltraNet Communications, Inc. Lines: 26 Message-ID: <8mbu5h$5pp$1@bob.news.rcn.net> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <6ubszohc1s.fsf@chonsp.franklin.ch> <8lfj1i$404$1@teabag.cbhnet> <8lh8k3$iui$5@bob.news.rcn.net> <8ma8nt$arf$1@news.iastate.edu> X-Trace: d1juGZ1UbAb4SX2i+wZ97+k+qp4QZKlSkoCd8x4JBcU= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 3 Aug 2000 14:04:33 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!netnews.com!feed1.news.rcn.net!rcn!216-164-248-111 Xref: chonsp.franklin.ch alt.folklore.computers:61128 In article <8ma8nt$arf$1@news.iastate.edu>, rhawkins@iastate.edu (Rick Hawkins) wrote: >In article <8lh8k3$iui$5@bob.news.rcn.net>, wrote: >>In article <8lfj1i$404$1@teabag.cbhnet>, >> cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) wrote: >>>In article <6ubszohc1s.fsf@chonsp.franklin.ch>, >>> Neil Franklin writes: >>>> Just like Y2K was believed to be by many. :-) > >>>It wasn't only believed to be a red-herring but was proven to be >>>so in most cases. > >>I wouldn't get too smug since there seems to be people who are >>writing Y2K bugs today. > >Like you just did :) That one went right over my head. Explanation, please? :-0 >Now they're MMC bugs . . . Nope, they're going to have to get through MMI first. /BAH Subtract a hundred and four for e-mail. ###### From: rhawkins@iastate.edu (Rick Hawkins) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 2 Aug 2000 22:52:45 GMT Organization: Iowa State University, Ames, Iowa USA Lines: 24 Message-ID: <8ma8nt$arf$1@news.iastate.edu> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <6ubszohc1s.fsf@chonsp.franklin.ch> <8lfj1i$404$1@teabag.cbhnet> <8lh8k3$iui$5@bob.news.rcn.net> NNTP-Posting-Host: isua1.iastate.edu Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!HSNX.atgi.net!cyclone.bc.net!logbridge.uoregon.edu!newsrelay.iastate.edu!news.iastate.edu!rhawkins Xref: chonsp.franklin.ch alt.folklore.computers:61144 In article <8lh8k3$iui$5@bob.news.rcn.net>, wrote: >In article <8lfj1i$404$1@teabag.cbhnet>, > cbh@REMOVE_THIS.teabag.demon.co.uk (Chris Hedley) wrote: >>In article <6ubszohc1s.fsf@chonsp.franklin.ch>, >> Neil Franklin writes: >>> Just like Y2K was believed to be by many. :-) >>It wasn't only believed to be a red-herring but was proven to be >>so in most cases. >I wouldn't get too smug since there seems to be people who are >writing Y2K bugs today. Like you just did :) Now they're MMC bugs . . . hawk -- Richard E. Hawkins dochawk@psu.edu [regardless of where the message says it comes from] These opinions will not be those of Penn State until they pay my retainer. ###### From: rhawkins@iastate.edu (Rick Hawkins) Newsgroups: alt.folklore.computers Subject: Re: Longest running computer Date: 2 Aug 2000 23:17:05 GMT Organization: Iowa State University, Ames, Iowa USA Lines: 16 Message-ID: <8maa5h$b14$1@news.iastate.edu> References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8lehhe$g74$1@flywheel.innovative.iinet.net.au> <397B7B0A.F680ED9B@ev1.net> NNTP-Posting-Host: isua1.iastate.edu Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!logbridge.uoregon.edu!newsrelay.iastate.edu!news.iastate.edu!rhawkins Xref: chonsp.franklin.ch alt.folklore.computers:61121 In article <397B7B0A.F680ED9B@ev1.net>, Charles Richmond wrote: >IMHO the Oracle database has an excellent time/date storage facility. It >handles dates back to 2000-something B.C., and also way, way into the future. So I need to keep on looking for an OS that can handle my records from the Arc . . . :) -- Richard E. Hawkins dochawk@psu.edu [regardless of where the message says it comes from] These opinions will not be those of Penn State until they pay my retainer. ###### Sender: Ian Stirling From: Ian Stirling Subject: Re: Longest running computer Newsgroups: alt.folklore.computers References: <3972935c$0$8328@wodc7nh6.news.uu.net> <8l96el$kad$2@teabag.demon.co.uk> <397848A1.424F06D7@trailing-edge.com> <8la02l$116k$1@teabag.demon.co.uk> <3979b755$0$8317$3c090ad1@news.plethora.net> <6uhf9im2j7.fsf@chonsp.franklin.ch> <8lk0ac$o2c$7@bob.news.rcn.net> <8ls27d$423$1@bob.news.rcn.net> Organization: None.. User-Agent: tin/pre-1.4-980818 ("Laura") (UNIX) (Linux/2.2.14 (i586)) Lines: 38 Message-ID: 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 X-Complaints-To: support@usenetserver.com NNTP-Posting-Date: Thu, 03 Aug 2000 13:42:52 EDT Date: Thu, 03 Aug 2000 17:42:52 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!newsxfer.eecs.umich.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!portc03.blue.aol.com!cyclone2.usenetserver.com!news-out.usenetserver.com!cyclone1.usenetserver.com!news-east.usenetserver.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:61083 jmfbahciv@aol.com wrote: >In article , > djb0x1312fd7@scream.org (Dan) wrote: >>jmfbahciv@aol.com wrote: >> >>> >> File contents. Files written by user software that contain 32 bit >date >>> >> fields. The same problem an with 2 digit year numbers. Y2K tidy up >>> >> code audits all over again. >>> > >>> >Ah, but... As long as the start of the epoch is 1/1/1970, wouldn't a >>> >system with a 64-bit time_t still read a 32-bit "legacy" date, count >>> >the appropriate number of seconds, and render the correct date? The >>> >numbers themselves are not going to abruptly stop being valid, right? >>> >>> How are you going to know that number you have came from a >>> 32-bit field rather than a 64-bit field _without setting some >>> kind of flag that means "new format"_. And how many "new >>> formats" can one have? >> >>Got me. :) A 32-bit value will fit into a 64-bit field, that's about >>the extent of my firm grasp of the obvious. >That makes a conversion sorta easier. How does a program decide >to limit the bit pick up to 32 rather than 64? We ran into >similar problems when the format went from a half word to a full >word. Acutally, IIRC, the field went from 24 bits to 36 bits. An appropriately timed leap-second every few decades could remove the ambiguity. -- http://inquisitor.i.am/ | mailto:inquisitor@i.am | Ian Stirling. ---------------------------+-------------------------+-------------------------- "Melchett : Unhappily Blackadder, the Lord High Executioner is dead Blackadder : Oh woe! Murdered of course. Melchett : No, oddly enough no. They usually are but this one just got careless one night and signed his name on the wrong dotted line. They came for him while he slept." - Blackadder II