![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
thewayneThis is a remarkable historical find! (at least for computer people)
A storage room at the University of Utah was being cleaned out and they found a 9-track reel tape, labeled "UNIX Original From Bell Labs V4 (See Manual for format)". Univ V4 is a milestone version from 1973 in that it is the first version completely written in the C programming language, which became the standard for many years. Somehow the source code was lost, and this might be a recovery point!
The big question is: is the tape readable... And there's absolutely no way to know that until the tape is literally studied to see what shape it physically is in and then hopefully mounted on a tape drive and read.
A 9-track tape is the classic seen in old movies where you see people popping 14" tapes into drives that stand taller than a person, and the tape drops into a loop lower into the drive so there's slack, causing no direct tension on the tape itself as it spools back and forth. I spent some time in data centers in the '80s doing some apprenticeships and also working for a certain moving van rental company mounting them, which I actually found to be a lot of fun.
The problem is... FIVE DECADES? There's no information as to what sort of storage room the tape was found in. Was this a proper university library archive, with temperature and climate control? Was the tape stored flat, or upright? If it was stored flat on its back, then 50 years of gravity may have distorted an edge of the tape. Even upright, in less than an ideal environment, may have caused it to degrade and stick to itself.
There's absolutely no telling if the tape is readable. I don't remember if 9-track tapes stored much in the way of recovery data if part of it is unreadable, so if there's a bad patch, can information still be recovered? I have no idea. But there is hope: the tape is being sent to the Computer History Museum, where I believe they not only have a tape drive that can read it, they probably have old boffins who are familiar with the encoding format and have the expertise that might be able to recover more information from it if there is problems.
We shall see. Interesting times!
The information on it is purely of historical interest, there's no program code on it that will revolutionize current programming theory. At that time, Unix shipped as source code - the actual C programs - and you had to compile it on your specific computer to make it work. This made the operating system maintainable as you could fix any bugs that came up, then you could talk to the guys at AT&T and tell them what happened and they could theoretically incorporate a better fix in the master for the next release. But all subsequent generations of Unix built on V4 had better code implementations, so as I said, it's probably purely of historical interest. If it's recovered, people will have fun looking at the code, but they'd learn more of computer science studying current Linux source code.
Apparently they are going to drive the tape nearly 800 miles (about 12 hours) to the Computer History Museum rather than risking shipping it, I wish them safe travels! And the Museum already has plans on how to read the tape - though I hope they plan on doing a physical examination first, unless, of course, it was stored in ideal conditions the whole 50 years.
Yeah, I think I'd drive it, too, rather than ship it. And the way flying is screwed up right now with the government shutdown? Probably faster to drive.
https://www.theregister.com/2025/11/07/unix_fourth_edition_tape_rediscovered/
https://tech.slashdot.org/story/25/11/09/0528258/lost-unix-v4-possibly-recovered-on-a-forgotten-bell-labs-tape-from-1973
A storage room at the University of Utah was being cleaned out and they found a 9-track reel tape, labeled "UNIX Original From Bell Labs V4 (See Manual for format)". Univ V4 is a milestone version from 1973 in that it is the first version completely written in the C programming language, which became the standard for many years. Somehow the source code was lost, and this might be a recovery point!
The big question is: is the tape readable... And there's absolutely no way to know that until the tape is literally studied to see what shape it physically is in and then hopefully mounted on a tape drive and read.
A 9-track tape is the classic seen in old movies where you see people popping 14" tapes into drives that stand taller than a person, and the tape drops into a loop lower into the drive so there's slack, causing no direct tension on the tape itself as it spools back and forth. I spent some time in data centers in the '80s doing some apprenticeships and also working for a certain moving van rental company mounting them, which I actually found to be a lot of fun.
The problem is... FIVE DECADES? There's no information as to what sort of storage room the tape was found in. Was this a proper university library archive, with temperature and climate control? Was the tape stored flat, or upright? If it was stored flat on its back, then 50 years of gravity may have distorted an edge of the tape. Even upright, in less than an ideal environment, may have caused it to degrade and stick to itself.
There's absolutely no telling if the tape is readable. I don't remember if 9-track tapes stored much in the way of recovery data if part of it is unreadable, so if there's a bad patch, can information still be recovered? I have no idea. But there is hope: the tape is being sent to the Computer History Museum, where I believe they not only have a tape drive that can read it, they probably have old boffins who are familiar with the encoding format and have the expertise that might be able to recover more information from it if there is problems.
We shall see. Interesting times!
The information on it is purely of historical interest, there's no program code on it that will revolutionize current programming theory. At that time, Unix shipped as source code - the actual C programs - and you had to compile it on your specific computer to make it work. This made the operating system maintainable as you could fix any bugs that came up, then you could talk to the guys at AT&T and tell them what happened and they could theoretically incorporate a better fix in the master for the next release. But all subsequent generations of Unix built on V4 had better code implementations, so as I said, it's probably purely of historical interest. If it's recovered, people will have fun looking at the code, but they'd learn more of computer science studying current Linux source code.
Apparently they are going to drive the tape nearly 800 miles (about 12 hours) to the Computer History Museum rather than risking shipping it, I wish them safe travels! And the Museum already has plans on how to read the tape - though I hope they plan on doing a physical examination first, unless, of course, it was stored in ideal conditions the whole 50 years.
Yeah, I think I'd drive it, too, rather than ship it. And the way flying is screwed up right now with the government shutdown? Probably faster to drive.
https://www.theregister.com/2025/11/07/unix_fourth_edition_tape_rediscovered/
https://tech.slashdot.org/story/25/11/09/0528258/lost-unix-v4-possibly-recovered-on-a-forgotten-bell-labs-tape-from-1973
no subject
Date: 2025-11-11 04:14 am (UTC)I barely avoided having to work with punch cards. The class before me at the community college was the last one: they still had the equipment for them to finish up, and I did learn to read them: I could look at one and interpret the punches. But that was a long time ago and I've since forgotten. I learned on both a Hitachi mainframe that ran IBM OS and pretended to be an IBM 370, and on a CP/M micro with like 8 CRTs strung off it. The mainframe was a lot more fun! I enjoyed Cobol and was good at it, don't remember a thing about Fortran. Took a couple of courses on Unix, back before Linux hit it big - don't really remember much about it. I use a smattering of it on my Mac and often wish I were more fluent.
no subject
Date: 2025-11-11 10:30 am (UTC)At this point I remember zero programming or coding. I supposed putting formulas into Excel, or the occasional html command into Dreamwidth is a rudimentary form of programming, but not what most people mean by that.
Sent from Proton Mail for iOS.
no subject
Date: 2025-11-11 03:12 pm (UTC)I don't get to do nearly as much programming as I'd like as it's not really a job requirement. Mainly I dabble in database queries now and again. I plan on getting more into Python and perhaps another couple of languages when I retire in a couple of years.
no subject
Date: 2025-11-11 11:10 pm (UTC)