From tuhs at tuhs.org Mon Dec 8 15:03:00 2025 From: tuhs at tuhs.org (Warren Toomey via TUHS) Date: Mon, 8 Dec 2025 15:03:00 +1000 Subject: [TUHS] Test of TUHS list Message-ID: All, it's been quiet. I suspect that the list broke somehow. This is a test to see what happens when I send in an e-mail. Cheers, Warren From tuhs at tuhs.org Mon Dec 8 15:13:01 2025 From: tuhs at tuhs.org (George Michaelson via TUHS) Date: Mon, 8 Dec 2025 15:13:01 +1000 Subject: [TUHS] Test of TUHS list In-Reply-To: References: Message-ID: You sent it while all the american residents are asleep. Lets sneak into the computer history museum and steal all the good bits. -G On Mon, Dec 8, 2025 at 3:11 PM Warren Toomey via TUHS wrote: > All, it's been quiet. I suspect that the list broke somehow. > This is a test to see what happens when I send in an e-mail. > > Cheers, Warren > From tuhs at tuhs.org Mon Dec 8 15:18:01 2025 From: tuhs at tuhs.org (Jason Bowen via TUHS) Date: Sun, 7 Dec 2025 23:18:01 -0600 (CST) Subject: [TUHS] Test of TUHS list In-Reply-To: <424d9522-454b-4e5a-be89-40c8a4b2a3e5@infinitecactus.com> References: <424d9522-454b-4e5a-be89-40c8a4b2a3e5@infinitecactus.com> Message-ID: Dec 7, 2025 23:15:35 Jason Bowen : > Dec 7, 2025 23:13:33 George Michaelson via TUHS : > >> You sent it while all the american residents are asleep. >> >> Lets sneak into the computer history museum and steal all the good bits. >> >> -G >> >> On Mon, Dec 8, 2025 at 3:11 PM Warren Toomey via TUHS wrote: >> >>> All, it's been quiet. I suspect that the list broke somehow. >>> This is a test to see what happens when I send in an e-mail. >>> >>> Cheers, Warren >>> (left off the list) I would guess a decent number of us are still awake :) Which bit is best? From tuhs at tuhs.org Sun Dec 7 06:16:24 2025 From: tuhs at tuhs.org (ron minnich via TUHS) Date: Sat, 6 Dec 2025 12:16:24 -0800 Subject: [TUHS] V4 in tv news Message-ID: https://youtu.be/IR-f07LN0-Y?si=vU3P5MLwIsMC84jV From tuhs at tuhs.org Mon Dec 8 15:21:10 2025 From: tuhs at tuhs.org (Warren Toomey via TUHS) Date: Mon, 8 Dec 2025 15:21:10 +1000 Subject: [TUHS] Test of TUHS list In-Reply-To: References: Message-ID: On Mon, Dec 08, 2025 at 03:03:00PM +1000, Warren Toomey via TUHS wrote: > All, it's been quiet. I suspect that the list broke somehow. > This is a test to see what happens when I send in an e-mail. Looks like the mailman3 LMTP service failed to restart when I rebooted the server last week. It's running again now! Cheers & thanks for all the replies :-) Warren From tuhs at tuhs.org Mon Dec 8 15:30:25 2025 From: tuhs at tuhs.org (Thalia Archibald via TUHS) Date: Mon, 08 Dec 2025 05:30:25 +0000 Subject: [TUHS] V4 in tv news In-Reply-To: References: Message-ID: Hi Ron, I’m a PhD student at the University of Utah and have been researching the history of our V4 tape ever since the news broke. Incidentally, I made an extensive list of 36 news stories on the tape just today[0]. That same repo has all the sources I’ve been collecting, except for some I don’t have scans for yet. I found that it was delivered the Martin Newell[1], famously known for the Utah Teapot model, and a letter from Ken to Martin[2] was later found that places our delivery around June 1974. I’m still looking for our license agreement. If any of you know of good early UNIX sources, particularly around 1973–1974, please let me know! Thalia [0]: https://github.com/thaliaarchi/unix-history/blob/main/users/utah/v4.md#News [1]: https://archive.org/details/unix_news_july-30-1975/page/n9/mode/1up [2]: https://github.com/thaliaarchi/unix-history/blob/8f887ce324bc7dc7fc6de5eca09d4d028fe8f554/users/utah/1974-05-31_ken_letter.jpg > On Dec 6, 2025, at 13:16, ron minnich via TUHS wrote: > > https://youtu.be/IR-f07LN0-Y?si=vU3P5MLwIsMC84jV From tuhs at tuhs.org Mon Dec 8 15:45:07 2025 From: tuhs at tuhs.org (Thalia Archibald via TUHS) Date: Mon, 08 Dec 2025 05:45:07 +0000 Subject: [TUHS] Test of TUHS list In-Reply-To: References: Message-ID: <9C1E0FFF-9955-4151-BFEE-90A4EA5D18A2@archibald.dev> On Dec 7, 2025, at 22:21, Warren Toomey via TUHS wrote: > Looks like the mailman3 LMTP service failed to restart when > I rebooted the server last week. It's running again now! It looks like Ron’s recent message was stuck unsent for a day. It was sent 2025-12-06 20:16:24 UTC, according to its Date header, but I received it 2025-12-08 05:20:03 UTC. Thalia From tuhs at tuhs.org Tue Dec 9 06:08:04 2025 From: tuhs at tuhs.org (Warren Toomey via TUHS) Date: Tue, 9 Dec 2025 06:08:04 +1000 Subject: [TUHS] Unix Magic poster annotation project Message-ID: Hi all, I just found this e-mail in my inbox. It's really cute :-) Cheers, Warren ----- Forwarded message from David Rio Deiros ----- Hi Warren, I've been working on a project to document all the hidden references in Gary Overacre's Unix Magic poster. It's a simple interactive site where you can click on parts of the poster and read what each reference means. code: https://github.com/drio/unixmagic static site: https://unixmagic.net We've got about 40 annotations so far. Thought the TUHS community might enjoy it and maybe you have some insights about some of the references. Anyway, thank you for reading and have a wonderful day! -drd ----- End forwarded message ----- From tuhs at tuhs.org Tue Dec 9 12:41:41 2025 From: tuhs at tuhs.org (Douglas McIlroy via TUHS) Date: Mon, 8 Dec 2025 21:41:41 -0500 Subject: [TUHS] Unix Magic poster annotation project Message-ID: David Rio Deiros has conjured up a fun game. Regardless of what #39 signifies, the object isn't a skull. Skulls don't have ears. I doubt that it symbolizes /dev/null, for it is the source of a steady flow INTO null. On close examination, the face is cat-like, but there's already a cat elsewhere. In any event, the flowing substance must represent data. The wizard is surrounded with jars of spices that adjust the flavor of the data that is being vigorously processed in the shell. Some data comes (typed in?) from sources in the fingers of both hands, and some comes from the file system (#38). The man (#2) cuts hay with a scythe. As a complement, the unlabeled pot next to the tar mortar (#12) may hold paste. Was Gary Overacre clairvoyant? What besides SE Linux could be signified by the hook at the end of the long-handled utensil that rests in the seething cauldron? Whose initials are hidden in the ribbon's twist (#5)? It looks like another DMR may follow the second KT, It's harder to guess who comes after BWK--possibly R(H)M. Doug From tuhs at tuhs.org Wed Dec 10 01:16:02 2025 From: tuhs at tuhs.org (=?utf-8?q?Cameron_M=C3=AD=C4=8Be=C3=A1l_Tyre_via_TUHS?=) Date: Tue, 09 Dec 2025 15:16:02 +0000 Subject: [TUHS] Tahoe Message-ID: Hello folks, Illness and work, both at the same time, prevented me from posting this a couple of months ago. I noted, when Apple released macOS 26 three months ago that it was called macOS Tahoe and I remembered that there was a processor specific release of 4.3 BSD back in 1988 that was officially referred to as "4.3BSD tahoe". I mention this only because, statistically, the chances of two Unix releases being given the same name must be slim, granted they were for different reasons, one referring to the platform it was to run on and the other because I guess Apple thought it was a nice name? I guess it could also be argued they didn't have the same name as one was "tahoe" and the other is "Tahoe" or that they did but that one being capitalized and one not is how the statistics were defeated. Have a safe rest of your week, everyone! Cameron From tuhs at tuhs.org Wed Dec 10 01:49:21 2025 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Tue, 9 Dec 2025 10:49:21 -0500 Subject: [TUHS] Tahoe In-Reply-To: References: Message-ID: below On Tue, Dec 9, 2025 at 10:16 AM Cameron Míċeál Tyre via TUHS wrote: > Hello folks, > > Illness and work, both at the same time, prevented me from posting this a > couple of months ago. > > I noted, when Apple released macOS 26 three months ago that it was called > macOS Tahoe and I remembered that there was a processor specific release of > 4.3 BSD back in 1988 that was officially referred to as "4.3BSD tahoe". > The target was Computer Consoles Inc.'s (CCI) Power 6/32 processor, which was codenamed Tahoe (after Lake Tahoe). > > I mention this only because, statistically, the chances of two Unix > releases being given the same name must be slim, granted they were for > different reasons, one referring to the platform it was to run on and the > other because I guess Apple thought it was a nice name? Different people, although both are named after the same place. In most (if not all) of the companies I have worked at, project code names are controlled by the Marketing group. The first nine releases of Mac OS X were codenamed after "Big Cats" [Cheetah 10.0 through Mountain Lion 10.8 ]. Starting with 10.9 [Maverick — the famous big-wave surfing spot near Half Moon Bay], Apple began using physical landmarks in California. I doubt that the people at Apple picking codenames knew that Lake Tahoe had already been used in the computer community, either by the CCI or by CRSG's specific release for the CCI's Power 6/32 (*a.k.a.* Tahoe Processor). > > > I guess it could also be argued they didn't have the same name as one was > "tahoe" and the other is "Tahoe" or that they did but that one being > capitalized and one not is how the statistics were defeated. > Not really, since both are referring to Lake Tahoe. I suspect ignorance and indifference to the previous product name being associated with another processor firm's product codename. From tuhs at tuhs.org Wed Dec 10 01:57:37 2025 From: tuhs at tuhs.org (Larry McVoy via TUHS) Date: Tue, 9 Dec 2025 07:57:37 -0800 Subject: [TUHS] Tahoe In-Reply-To: References: Message-ID: <20251209155737.GJ27513@mcvoy.com> On Tue, Dec 09, 2025 at 10:49:21AM -0500, Clem Cole via TUHS wrote: > ]. Starting with 10.9 [Maverick ??? the famous big-wave surfing spot near > Half Moon Bay], Apple began using physical landmarks in California. Not that it's important, but I fish out of Half Moon Bay regularly and go around Mavericks every time. It's a big reef that forces the waves to crest up, you are well served to go around it. In spite of that, I see idiots cross the reef all the time. Can't fix stupid I guess. -- --- Larry McVoy Retired to fishing http://www.mcvoy.com/lm/boat From tuhs at tuhs.org Wed Dec 10 02:38:41 2025 From: tuhs at tuhs.org (Rob Pritchard via TUHS) Date: Tue, 9 Dec 2025 16:38:41 +0000 Subject: [TUHS] Tahoe In-Reply-To: References: Message-ID: Well, if the next MacOS release is called ‘Reno’ then we know their marketing teams are running out of ideas! > On 9 Dec 2025, at 15:16, Cameron Míċeál Tyre via TUHS wrote: > > Hello folks, > > Illness and work, both at the same time, prevented me from posting this a couple of months ago. > > I noted, when Apple released macOS 26 three months ago that it was called macOS Tahoe and I remembered that there was a processor specific release of 4.3 BSD back in 1988 that was officially referred to as "4.3BSD tahoe". > > I mention this only because, statistically, the chances of two Unix releases being given the same name must be slim, granted they were for different reasons, one referring to the platform it was to run on and the other because I guess Apple thought it was a nice name? > > I guess it could also be argued they didn't have the same name as one was "tahoe" and the other is "Tahoe" or that they did but that one being capitalized and one not is how the statistics were defeated. > > Have a safe rest of your week, everyone! > > Cameron > > From tuhs at tuhs.org Wed Dec 10 03:15:40 2025 From: tuhs at tuhs.org (Al Kossow via TUHS) Date: Tue, 9 Dec 2025 09:15:40 -0800 Subject: [TUHS] Tahoe In-Reply-To: References: Message-ID: On 12/9/25 7:16 AM, Cameron Míċeál Tyre via TUHS wrote: > Hello folks, > > Illness and work, both at the same time, prevented me from posting this a couple of months ago. > > I noted, when Apple released macOS 26 three months ago that it was called macOS Tahoe and I remembered that there was a processor specific release of 4.3 BSD back in 1988 that was officially referred to as "4.3BSD tahoe". > Codenames get reused https://www.tuhs.org/cgi-bin/utree.pl?file=4.3BSD-Tahoe I worked on an internal Apple RISC project called Jaguar which has nothing to do with MacOS 10.2 From tuhs at tuhs.org Wed Dec 10 06:46:01 2025 From: tuhs at tuhs.org (christopher fujino via TUHS) Date: Tue, 9 Dec 2025 12:46:01 -0800 Subject: [TUHS] Tahoe In-Reply-To: References: Message-ID: I'm bet both Berkeley researchers and Apple product managers in Cupertino go skiing in Tahoe. On Tue, Dec 9, 2025 at 9:15 AM Al Kossow via TUHS wrote: > On 12/9/25 7:16 AM, Cameron Míċeál Tyre via TUHS wrote: > > Hello folks, > > > > Illness and work, both at the same time, prevented me from posting this > a couple of months ago. > > > > I noted, when Apple released macOS 26 three months ago that it was > called macOS Tahoe and I remembered that there was a processor specific > release of 4.3 BSD back in 1988 that was officially referred to as "4.3BSD > tahoe". > > > > Codenames get reused > > https://www.tuhs.org/cgi-bin/utree.pl?file=4.3BSD-Tahoe > > I worked on an internal Apple RISC project called Jaguar which has nothing > to do with MacOS 10.2 > > > From tuhs at tuhs.org Wed Dec 10 08:57:10 2025 From: tuhs at tuhs.org (Alan Coopersmith via TUHS) Date: Tue, 9 Dec 2025 14:57:10 -0800 Subject: [TUHS] XPG4, X/OPEN Curses, and SVID Issue 4 Physical Publication? In-Reply-To: References: Message-ID: On 10/24/25 10:57, segaloco via TUHS wrote: > Good day everyone, I've recently added an X/OPEN Portability Guide > section and a photo of the first SVID issue to the standards section of > the wiki: https://wiki.tuhs.org/doku.php?id=publications:standards > > I was wondering, does anyone here know if there were physically > published versions of XPG4, any X/OPEN Curses physical literature, > and/or published SVID Issue 4 copies? The latter I suspect is not very > likely as SVID Issue 4 concerns spun-off-USL SVR 4.2 I believe, which > didn't make the same splash earlier SVIDs did. I've only seen PDFs of > this Issue 4. I finally remembered to check our standards bookshelf when I was in the office today, and found printed copies of: - Docs with XPG4 branding: - CAE Specification: System Interfaces and Headers, Issue 4 - CAE Specification: Commands and Utilities, Issue 4 - CAE Specification: System Interface Definitions, Issue 4 - XPG3-XPG4 Base Migration Guide - X/Open Systems and Branded Products: XPG4 - Docs which didn't explictly have "XPG4" on the cover: - CAE Specification: Networking Services, Issue 4 - CAE Specification: X/Open Curses, Issue 4 While there were copies of earlier SVID generations, I didn't see Issue 4 there. I took quick pictures with my cell phone of the covers of each of these if you want them - they're not archival quality scans or anything, but it doesn't look like that's what you're going for on that wiki page. -- -Alan Coopersmith- alan.coopersmith at oracle.com Oracle Solaris Engineering - https://blogs.oracle.com/solaris From tuhs at tuhs.org Wed Dec 10 11:46:43 2025 From: tuhs at tuhs.org (=?utf-8?q?Cameron_M=C3=AD=C4=8Be=C3=A1l_Tyre_via_TUHS?=) Date: Wed, 10 Dec 2025 01:46:43 +0000 Subject: [TUHS] Tahoe Message-ID: Hello, folks, Thank you all for your comments on Tahoe and project names and I guess Lake Tahoe is on my bucket list because it looks beautiful. Rob's comment about Reno made me chuckle although I guess it's the wrong side of the state line for Apple to use although there is that famous saying, "Never say never!" Best regards, Cameron From tuhs at tuhs.org Wed Dec 10 12:53:52 2025 From: tuhs at tuhs.org (Erik E. Fair via TUHS) Date: Tue, 09 Dec 2025 18:53:52 -0800 Subject: [TUHS] Tahoe In-Reply-To: Message-ID: <21977.1765335232@cesium.clock.org> >From: =?utf-8?q?Cameron_Míċeál_Tyre_via_TUHS?= > >Hello, folks, > >Thank you all for your comments on Tahoe and project names and I guess Lake Tahoe is on my bucket list because it looks beautiful. > >Rob's comment about Reno made me chuckle although I guess it's the wrong side of the state line for Apple to use although there is that famous saying, "Never say never!" > >Best regards, > >Cameron Funny you should say "wrong side of the state line" - the wallpaper and other photos of Lake Tahoe that Apple likes to use (there's been one in MacOS for a decade or more, taken during wildfire smoke inundation) are taken from the east (Nevada) shoreline at Sand Harbor (Lake Tahoe Nevada state park) and Memorial point, along NV-28. I know because I've lived in Incline Village, NV (NE corner of Lake Tahoe) since 2002 and have been coming up to North Lake Tahoe (winter & summer) since I was a child. I'm quite familiar with the local landscape. So long as Apple's theme is "pretty/famous places in California", I don't expect they'll run out in the near term: they could use Point Reyes, Mount Shasta, Napa, Mount Lassen, Joshua Tree, Mammoth, Devil's Postpile, Bodie, Downieville, Portola, ... Erik Fair From tuhs at tuhs.org Wed Dec 10 13:39:02 2025 From: tuhs at tuhs.org (Larry McVoy via TUHS) Date: Tue, 9 Dec 2025 19:39:02 -0800 Subject: [TUHS] Tahoe In-Reply-To: References: Message-ID: <20251210033902.GK27513@mcvoy.com> If you, or anyone, gets close enough to go to Lake Tahoe, I can take you out to the Pacific. I've kind of left tech behind, I put in my 30 years, and while I like tracking tech, I love fishing on the ocean. It's my retirement, it's complicated enough that it engages my engineer brain a bit but it is way easier than the problems I solved as an engineer. I can put you on fish, for sure. Halibut, rockies, lings, crabs, salmon when they let us fish for those (which is not often). I'm still chasing tuna, hauled them in on my friends boat, not gotten them on my boat. On Wed, Dec 10, 2025 at 01:46:43AM +0000, Cameron M????e??l Tyre via TUHS wrote: > Hello, folks, > > Thank you all for your comments on Tahoe and project names and I guess Lake Tahoe is on my bucket list because it looks beautiful. > > Rob's comment about Reno made me chuckle although I guess it's the wrong side of the state line for Apple to use although there is that famous saying, "Never say never!" > > Best regards, > > Cameron -- --- Larry McVoy Retired to fishing http://www.mcvoy.com/lm/boat From tuhs at tuhs.org Wed Dec 10 17:06:14 2025 From: tuhs at tuhs.org (Arno Griffioen via TUHS) Date: Wed, 10 Dec 2025 08:06:14 +0100 Subject: [TUHS] Tahoe In-Reply-To: References: Message-ID: On Tue, Dec 09, 2025 at 09:15:40AM -0800, Al Kossow via TUHS wrote: > Codenames get reused > > https://www.tuhs.org/cgi-bin/utree.pl?file=4.3BSD-Tahoe > > I worked on an internal Apple RISC project called Jaguar which has nothing > to do with MacOS 10.2 And of course 'Jaguar' was also the name of the ill-fated Atari game console from the early 90's, so that nicely proves the point ;) Bye, Arno. From tuhs at tuhs.org Tue Dec 16 07:55:54 2025 From: tuhs at tuhs.org (Douglas McIlroy via TUHS) Date: Mon, 15 Dec 2025 16:55:54 -0500 Subject: [TUHS] Posix ed weirdness Message-ID: The ed j command joins a sequence of lines into one line and sets dot to the result. Thus 2,4j combines three lines and sets dot to 2. Naturally one expects 2,2j to make no visible change to the file, but to set dot to 2. Indeed, that's what v7 did. But Posix decrees that j "does nothing" in this case and leaves dot at the value it had before the command. Does anyone know why the Posix committee chose to break both the original behavior and Kernighan's law: " 'Do nothing' gracefully"? Doug From tuhs at tuhs.org Tue Dec 16 08:19:26 2025 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Mon, 15 Dec 2025 17:19:26 -0500 Subject: [TUHS] Posix ed weirdness In-Reply-To: References: Message-ID: Doug, Do you know what later AT&T Summit versions did? And/or SVID Clem On Mon, Dec 15, 2025 at 4:56 PM Douglas McIlroy via TUHS wrote: > The ed j command joins a sequence of lines into one line and sets dot > to the result. Thus > 2,4j > combines three lines and sets dot to 2. Naturally one expects > 2,2j > to make no visible change to the file, but to set dot to 2. > > Indeed, that's what v7 did. But Posix decrees that j "does nothing" in > this case and leaves dot at the value it had before the command. > > Does anyone know why the Posix committee chose to break both the > original behavior and Kernighan's law: " 'Do nothing' gracefully"? > > Doug > From tuhs at tuhs.org Tue Dec 16 11:18:29 2025 From: tuhs at tuhs.org (Douglas McIlroy via TUHS) Date: Mon, 15 Dec 2025 20:18:29 -0500 Subject: [TUHS] Posix ed weirdness In-Reply-To: <20251216001101.z77ydwf2woduxgdo@illithid> References: <20251216001101.z77ydwf2woduxgdo@illithid> Message-ID: > My guess is because that's what the "consensus" of implementations, or > at least those willing to pay for POSIX certification, was. Yes, that's roughly how malloc(0) got messed up. But the malloc(0) anomaly at least had a rationale: 0-length data structures were impermissible in bare C. I suspect there was more politics afoot in the committee. I doubt that many implementers would hit on the bizarre idea that deleting n-1 newlines should be special-cased for n=1. Curiously, Posix didn't foist the same special-case behavior onto ex. Doug On Mon, Dec 15, 2025 at 7:11 PM G. Branden Robinson wrote: > > Hi Doug, > > At 2025-12-15T16:55:54-0500, Douglas McIlroy via TUHS wrote: > > The ed j command joins a sequence of lines into one line and sets dot > > to the result. Thus > > 2,4j > > combines three lines and sets dot to 2. Naturally one expects > > 2,2j > > to make no visible change to the file, but to set dot to 2. > > > > Indeed, that's what v7 did. But Posix decrees that j "does nothing" in > > this case and leaves dot at the value it had before the command. > > > > Does anyone know why the Posix committee chose to break both the > > original behavior and Kernighan's law: " 'Do nothing' gracefully"? > > My guess is because that's what the "consensus" of implementations, or > at least those willing to pay for POSIX certification, was. > > $ uname -a > SunOS gcc-solaris10 5.10 Generic_Virtual sun4u sparc SUNW,SPARC-Enterprise > $ cat ward > one > two > three > four > $ ed ward > 19 > p > four > p > four > 2,2j > p > four > > Regards, > Branden From tuhs at tuhs.org Wed Dec 17 01:26:54 2025 From: tuhs at tuhs.org (=?utf-8?q?Cameron_M=C3=AD=C4=8Be=C3=A1l_Tyre_via_TUHS?=) Date: Tue, 16 Dec 2025 15:26:54 +0000 Subject: [TUHS] Posix ed weirdness In-Reply-To: References: Message-ID: Hi Doug, Thank you. I use ed all the time, on every machine/device I have, but I did not know it was possible to join more than two lines at a time and I never thought of trying. I guess the syntax would have been simpler if only two could be joined at one time, something like 2j to join line 2 and line 3. The very fact that the syntax to join line 2 and line 3 would be 2,3j cries out, "I can join more than two lines!", but I just never figured it out. Now I know! Best regards, Cameron -------- Original Message -------- On Monday, 12/15/25 at 21:56 Douglas McIlroy via TUHS wrote, in part: The ed j command joins a sequence of lines into one line and sets dot to the result. Thus 2,4j combines three lines and sets dot to 2. From tuhs at tuhs.org Wed Dec 17 10:49:20 2025 From: tuhs at tuhs.org (Rob Pike via TUHS) Date: Wed, 17 Dec 2025 11:49:20 +1100 Subject: [TUHS] Posix ed weirdness In-Reply-To: References: Message-ID: Sam does it without special casing. -rob On Wed, Dec 17, 2025 at 2:27 AM Cameron Míċeál Tyre via TUHS wrote: > Hi Doug, > > Thank you. I use ed all the time, on every machine/device I have, but I > did not know it was possible to join more than two lines at a time and I > never thought of trying. > > I guess the syntax would have been simpler if only two could be joined at > one time, something like 2j to join line 2 and line 3. The very fact that > the syntax to join line 2 and line 3 would be > 2,3j > cries out, "I can join more than two lines!", but I just never figured it > out. > > Now I know! > > Best regards, > > Cameron > > > > > > -------- Original Message -------- > On Monday, 12/15/25 at 21:56 Douglas McIlroy via TUHS > wrote, in part: > > The ed j command joins a sequence of lines into one line and sets dot > to the result. Thus > 2,4j > combines three lines and sets dot to 2. > From tuhs at tuhs.org Wed Dec 17 13:19:14 2025 From: tuhs at tuhs.org (=?utf-8?q?Cameron_M=C3=AD=C4=8Be=C3=A1l_Tyre_via_TUHS?=) Date: Wed, 17 Dec 2025 03:19:14 +0000 Subject: [TUHS] Posix ed weirdness In-Reply-To: References: Message-ID: Hello Rob, Thank you. I've just been reading up on sam. Seems like I'll need to give it a try. Maybe that'll be my Christmas present to myself. An uninterrupted day learning a new editor. Best regards, Cameron -------- Original Message -------- On Wednesday, 12/17/25 at 00:50 Rob Pike via TUHS wrote: Sam does it without special casing. -rob From tuhs at tuhs.org Thu Dec 18 01:21:46 2025 From: tuhs at tuhs.org (Jacob Ritorto via TUHS) Date: Wed, 17 Dec 2025 10:21:46 -0500 Subject: [TUHS] update pdp11 system five? Message-ID: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> pdp11s are getting so popular these days that now seems a great time to perform the stunt of running SYSV well on them. Seems pretty normal to the adepts on the list who were there when it originally happened, but it’s kind of a niche mind-blow to newcomers that a PDP could run SYSV. But currently I think it’s missing networking. So a couple questions: Is it possible to build newer SYSV releases to run on pdp11? Like specifically SVr4? If not, is this feasible to achieve and would anyone be interested in a project to make that happen? If that can happen, the stock SVr4 networking code build might be a breeze. Apologies in advance if this has already been broached and completed and/or if this is a fool’s errand due to size constraints. And yes of course the easy thing to do is just run v6 or BSD. But then we miss out on the mind-blowing facet :) thx jake From tuhs at tuhs.org Thu Dec 18 01:41:05 2025 From: tuhs at tuhs.org (Arnold Robbins via TUHS) Date: Wed, 17 Dec 2025 08:41:05 -0700 Subject: [TUHS] update pdp11 system five? In-Reply-To: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> References: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> Message-ID: <202512171541.5BHFf5eD022558@freefriends.org> IIRC the last official System V release that ran on the PDP-11 was S5R2, but I don't remember for sure. S5R3 had STREAMS networking. S5R4 had real demand paged virtual memory and also, I think, socket support. Lack of address space will definitely be an issue; you'll have to look at overlays or some such, which I think 2.11 BSD does. But I bet you'll have fun! :-) Arnold Jacob Ritorto via TUHS wrote: > pdp11s are getting so popular these days that now seems a great time to perform the stunt of running SYSV well on them. Seems pretty normal to the adepts on the list who were there when it originally happened, but it’s kind of a niche mind-blow to newcomers that a PDP could run SYSV. > > But currently I think it’s missing networking. > > So a couple questions: > Is it possible to build newer SYSV releases to run on pdp11? Like specifically SVr4? > If not, is this feasible to achieve and would anyone be interested in a project to make that happen? > If that can happen, the stock SVr4 networking code build might be a breeze. > > Apologies in advance if this has already been broached and completed and/or if this is a fool’s errand due to size constraints. And yes of course the easy thing to do is just run v6 or BSD. But then we miss out on the mind-blowing facet :) > > thx > jake From tuhs at tuhs.org Thu Dec 18 01:49:17 2025 From: tuhs at tuhs.org (Larry McVoy via TUHS) Date: Wed, 17 Dec 2025 07:49:17 -0800 Subject: [TUHS] update pdp11 system five? In-Reply-To: <202512171541.5BHFf5eD022558@freefriends.org> References: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> <202512171541.5BHFf5eD022558@freefriends.org> Message-ID: <20251217154917.GC26336@mcvoy.com> On Wed, Dec 17, 2025 at 08:41:05AM -0700, Arnold Robbins via TUHS wrote: > IIRC the last official System V release that ran on the PDP-11 was > S5R2, but I don't remember for sure. > > S5R3 had STREAMS networking. I don't believe that's correct. SCO's S5R3 did not, I had to port it. > S5R4 had real demand paged virtual memory and also, I think, socket support. > > Lack of address space will definitely be an issue; you'll have to look > at overlays or some such, which I think 2.11 BSD does. > > But I bet you'll have fun! :-) > > Arnold > > Jacob Ritorto via TUHS wrote: > > > pdp11s are getting so popular these days that now seems a great time to perform the stunt of running SYSV well on them. Seems pretty normal to the adepts on the list who were there when it originally happened, but it???s kind of a niche mind-blow to newcomers that a PDP could run SYSV. > > > > But currently I think it???s missing networking. > > > > So a couple questions: > > Is it possible to build newer SYSV releases to run on pdp11? Like specifically SVr4? > > If not, is this feasible to achieve and would anyone be interested in a project to make that happen? > > If that can happen, the stock SVr4 networking code build might be a breeze. > > > > Apologies in advance if this has already been broached and completed and/or if this is a fool???s errand due to size constraints. And yes of course the easy thing to do is just run v6 or BSD. But then we miss out on the mind-blowing facet :) > > > > thx > > jake -- --- Larry McVoy Retired to fishing http://www.mcvoy.com/lm/boat From tuhs at tuhs.org Thu Dec 18 03:40:53 2025 From: tuhs at tuhs.org (Charles H Sauer (he/him) via TUHS) Date: Wed, 17 Dec 2025 11:40:53 -0600 Subject: [TUHS] update pdp11 system five? In-Reply-To: <20251217154917.GC26336@mcvoy.com> References: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> <202512171541.5BHFf5eD022558@freefriends.org> <20251217154917.GC26336@mcvoy.com> Message-ID: <5bd8f7f0-3e32-46e6-bd06-6eae610a4785@technologists.com> On 12/17/2025 9:49 AM, Larry McVoy via TUHS wrote: > On Wed, Dec 17, 2025 at 08:41:05AM -0700, Arnold Robbins via TUHS wrote: >> IIRC the last official System V release that ran on the PDP-11 was >> S5R2, but I don't remember for sure. >> >> S5R3 had STREAMS networking. > > I don't believe that's correct. SCO's S5R3 did not, I had to port it. Strange -- I thought all SVR3 had STREAMS. I'm pretty sure Dell's SVR3 had STREAMS, as provided by ISC. Heinz probably could be definitive. -- voice: +1.512.784.7526 e-mail: sauer at technologists.com fax: +1.512.346.5240 Web: https://technologists.com/sauer/ Facebook/Google/LinkedIn/mas.to: CharlesHSauer From tuhs at tuhs.org Thu Dec 18 04:47:33 2025 From: tuhs at tuhs.org (Ron Natalie via TUHS) Date: Wed, 17 Dec 2025 18:47:33 +0000 Subject: [TUHS] update pdp11 system five? In-Reply-To: <20251217154917.GC26336@mcvoy.com> References: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> <202512171541.5BHFf5eD022558@freefriends.org> <20251217154917.GC26336@mcvoy.com> Message-ID: I know for the BSD-ish networking on the PDP-11, it took one of the 8 kernel data segments to map mbufs. The top one is already reserved for the user struct/stack. The problem was that we were running out of address space even before putting TCP/IP into the kernel. We had to overlay the text pages. This is workable on the split I/D (23, 24, 44,45,55,70) but it killed off the idea of using those without it (11/34, 40, 60) for running networked UNIX. This was about the time I started recycling the PDP-11s into internet routers running the BRL-GATEWAY software. Eventually, even the 11-70’s succumbed to this. It was overkill, but I had them lying around. I also recycled one into being a control processor for the Denelcor HEP I/O system. The BRL-GATEWAY was my own invention running on a little pidgeon operating system called LOS (the little operating system). I did this when the MIT C Gateway lacked the features we needed and Noel Chiappa was at the time exiled to the Bahamas or something. -Ron I did get the BRL UNIX system (thanks guys) from McKusick’s repository for my PiDP-11 sitting on my desk. I suspect the LOS operating system and the accompanying router code has been lost to antiquity. ------ Original Message ------ >From "Larry McVoy via TUHS" To arnold at skeeve.com Cc tuhs at tuhs.org Date 12/17/2025 10:49:17 AM Subject [TUHS] Re: update pdp11 system five? >On Wed, Dec 17, 2025 at 08:41:05AM -0700, Arnold Robbins via TUHS wrote: >> IIRC the last official System V release that ran on the PDP-11 was >> S5R2, but I don't remember for sure. >> >> S5R3 had STREAMS networking. > >I don't believe that's correct. SCO's S5R3 did not, I had to port it. > >> S5R4 had real demand paged virtual memory and also, I think, socket support. >> >> Lack of address space will definitely be an issue; you'll have to look >> at overlays or some such, which I think 2.11 BSD does. >> >> But I bet you'll have fun! :-) >> >> Arnold >> >> Jacob Ritorto via TUHS wrote: >> >> > pdp11s are getting so popular these days that now seems a great time to perform the stunt of running SYSV well on them. Seems pretty normal to the adepts on the list who were there when it originally happened, but it???s kind of a niche mind-blow to newcomers that a PDP could run SYSV. >> > >> > But currently I think it???s missing networking. >> > >> > So a couple questions: >> > Is it possible to build newer SYSV releases to run on pdp11? Like specifically SVr4? >> > If not, is this feasible to achieve and would anyone be interested in a project to make that happen? >> > If that can happen, the stock SVr4 networking code build might be a breeze. >> > >> > Apologies in advance if this has already been broached and completed and/or if this is a fool???s errand due to size constraints. And yes of course the easy thing to do is just run v6 or BSD. But then we miss out on the mind-blowing facet :) >> > >> > thx >> > jake > >-- >--- >Larry McVoy Retired to fishing http://www.mcvoy.com/lm/boat From tuhs at tuhs.org Thu Dec 18 05:09:28 2025 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Wed, 17 Dec 2025 14:09:28 -0500 Subject: [TUHS] update pdp11 system five? In-Reply-To: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> References: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> Message-ID: below On Wed, Dec 17, 2025 at 10:22 AM Jacob Ritorto via TUHS wrote: > pdp11s are getting so popular these days that now seems a great time to > perform the stunt of running SYSV well on them. That will be a great deal of work, more in a minute. > Seems pretty normal to the adepts on the list who were there when it > originally happened, but it’s kind of a niche mind-blow to newcomers that a > PDP could run SYSV. > Be careful, System V Release 1 (SVR1) ran on both DEC PDP-11 and VAX-11/780; the PDP-11 was the reference architecture, while a VAX-11/780 port was also offered as an different UNIX implementation from 4.1BSD [the new memory management code for the Vax, was considered much cleaner - I believe much that came from the early Reisner Vax porting effort]. SVR2 (second release): The official "porting base" / reference architecture was the DEC VAX-11/780. With SVR2, AT&T's USG was shifting focus away from the PDP-11 as the target platform, though legacy code likely remained. That said, I don't remember a PDP-11 release for SRV2 SVR3 (third release): The official "porting base" / reference architecture was the WE 32000-based (called the 3B2, IIRC). With SVR3. Stream-based networking using the Transport Layer Interface (TLI) was added, but an IP/TCP stack was not provided. A number of folks such as Lachman offered them SVR4 (fourth release): The official "porting base" / reference architecture was the Intel 386/486 family > > But currently I think it’s missing networking. > It depends on which flavor of System V you start. > > So a couple questions: > Is it possible to build newer SYSV releases to run on pdp11? The SVR1 boot tape (which you can find in the wild, although the provenance might be shady. I've never seen a PDP-11 boot tape for any of later release (SVR2/3/4) > Like specifically SVr4? > Not out of the box. Your first big hurdle is going to be a PDP-11 compiler, much less one that self-hosts [modern C compiler will not fit in the 64K (even with 64K data/64K text for the separate I/D processors). > If not, is this feasible to achieve and would anyone be interested in a > project to make that happen? > If that can happen, the stock SVr4 networking code build might be a breeze. > I doubt it will be a breeze. As others have pointed out, 2.11BSD uses the DEC overlay code from the DEC PDP-11/V7 release, and managed to cram a much larger kernel into memory. But note that their IP/TCP stack runs in a separate address space (supervisor mode). > > Apologies in advance if this has already been broached and completed > and/or if this is a fool’s errand due to size constraints. And yes of > course the easy thing to do is just run v6 or BSD. But then we miss out on > the mind-blowing facet :) > Well, of course, V5/V6/V7 will all "just work". If you are willing to give up a networking API, you can also run SVR1. But if you really want a more modern take on UNIX on a PDP-11, 2.11BSD runs exceptionally well and has a reasonably present group of people who are creating bug fixes. But I will warn you: Steven Schultz (who leads this effort) makes it clear that the goal is not to recreate 4.2/4.3/4.4 on the PDP-11. Instead, 2.11BSD is to allow a modicum of modern tools to run there still. You can look for it at: 2.11BSD Clem From tuhs at tuhs.org Thu Dec 18 05:25:15 2025 From: tuhs at tuhs.org (Erik E. Fair via TUHS) Date: Wed, 17 Dec 2025 11:25:15 -0800 Subject: [TUHS] update pdp11 system five? In-Reply-To: References: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> Message-ID: <9076.1765999515@cesium.clock.org> If you can't make TCP/IP & Ethernet work in an old PDP-11, there's always UUCP & BerkNet! yes, I'm serial, Erik From tuhs at tuhs.org Thu Dec 18 05:27:14 2025 From: tuhs at tuhs.org (Larry McVoy via TUHS) Date: Wed, 17 Dec 2025 11:27:14 -0800 Subject: [TUHS] update pdp11 system five? In-Reply-To: References: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> Message-ID: <20251217192714.GE26336@mcvoy.com> On Wed, Dec 17, 2025 at 02:09:28PM -0500, Clem Cole via TUHS wrote: > With SVR3. Stream-based networking using the Transport Layer Interface > (TLI) was added, but an IP/TCP stack was not provided. A number of folks > such as Lachman offered them So far as I know, Lachman didn't write their stack, they bought it from Convergent. The internet says it was a joint development but my memory is Lachman bought a working stack. The only other STREAMS based TCP/IP stack I'm aware of is the Mentat one that was done for Sun, after some Sun VP, Larry somebody, paid Lachman a ton of money for theirs only to discover that the performance absolutely sucked. I did a little digging, Mentat made their own implementation of STREAMS because the Sys V one really sucked. I'm a little vague on the details but I think STREAMS queued at each layer so all the overhead was taking it off one layer's queue and putting it on a different layer's queue. Is anyone aware of a STREAMS TCP/IP besides the Lachman/Convergent and the Mentat ones? And in defense of Dennis' streams that USG morphed into STREAMS, he intended it for tty interfaces, not for networking. Ttys were slow enough that the queuing was fine. --lm From tuhs at tuhs.org Thu Dec 18 05:28:05 2025 From: tuhs at tuhs.org (Al Kossow via TUHS) Date: Wed, 17 Dec 2025 11:28:05 -0800 Subject: [TUHS] update pdp11 system five? In-Reply-To: <9076.1765999515@cesium.clock.org> References: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> <9076.1765999515@cesium.clock.org> Message-ID: <905e41b9-f445-f76f-756f-f702e0991ca2@bitsavers.org> On 12/17/25 11:25 AM, Erik E. Fair via TUHS wrote: > If you can't make TCP/IP & Ethernet work in an old PDP-11, there's always UUCP & BerkNet! > >     yes, I'm serial, > >     Erik Or fix the code on an Excelan card to speak modern TCP From tuhs at tuhs.org Thu Dec 18 05:54:30 2025 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Wed, 17 Dec 2025 14:54:30 -0500 Subject: [TUHS] update pdp11 system five? In-Reply-To: <20251217192714.GE26336@mcvoy.com> References: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> <20251217192714.GE26336@mcvoy.com> Message-ID: below On Wed, Dec 17, 2025 at 2:27 PM Larry McVoy wrote: > > Is anyone aware of a STREAMS TCP/IP besides the Lachman/Convergent and > the Mentat ones? > Those were the two most popular/. IICR Mentat was also used by HP. Also, I was under the impression that Doug Comer's stack was once stuffed into the TLI, and that BBN may have also developed/ported their stack to streams. > > And in defense of Dennis' streams that USG morphed into STREAMS, he > intended it for tty interfaces, not for networking. Ttys were slow > enough that the queuing was fine. > Yep, it was a nice trick to replace the early idea of "line disciplines." As with the monokernel vs. microkernel debate, using streams as a networking API might have solved some interesting issues. I once had a nice discussion with Denis about it, which was a classic theory vs. practice style discussion. That said, in practice, because the System V streams code performed so poorly, especially when a BSD socket stack was the comparison. Since almost all of the user code that did networking expected the sockets API, nobody felt that taking the time to make a TLI-based TCP/IP stack that was as performant as the traditional sockets was of any value. Remember that Metcalfe's law never talks about >>how<< you get connected, only that you are connected. SVR4, of course, offered both APIs, and I never saw any commercial code base that used the TLI. From tuhs at tuhs.org Thu Dec 18 06:48:03 2025 From: tuhs at tuhs.org (Johan Helsingius via TUHS) Date: Wed, 17 Dec 2025 21:48:03 +0100 Subject: [TUHS] update pdp11 system five? In-Reply-To: <9076.1765999515@cesium.clock.org> References: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> <9076.1765999515@cesium.clock.org> Message-ID: <896ea40f-d547-46bc-9b98-96cd9b171fe0@Julf.com> Now you have made me want to set up UUCP between two Raspberry Pis. Just because. Julf On 17/12/2025 8:25 pm, Erik E. Fair via TUHS wrote: > If you can't make TCP/IP & Ethernet work in an old PDP-11, there's always UUCP & BerkNet! > > yes, I'm serial, > > Erik From tuhs at tuhs.org Thu Dec 18 07:12:35 2025 From: tuhs at tuhs.org (Warner Losh via TUHS) Date: Wed, 17 Dec 2025 14:12:35 -0700 Subject: [TUHS] update pdp11 system five? In-Reply-To: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> References: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> Message-ID: On Wed, Dec 17, 2025 at 8:22 AM Jacob Ritorto via TUHS wrote: > pdp11s are getting so popular these days that now seems a great time to > perform the stunt of running SYSV well on them. Seems pretty normal to the > adepts on the list who were there when it originally happened, but it’s > kind of a niche mind-blow to newcomers that a PDP could run SYSV. > > But currently I think it’s missing networking. > > So a couple questions: > Is it possible to build newer SYSV releases to run on pdp11? Like > specifically SVr4? > Nope. The only extant version of system V that runs on the PDP-11 is r1. > If not, is this feasible to achieve and would anyone be interested in a > project to make that happen? > No. It's too big and requires demand paging. There's some nuance in this answer, but nobody has ported the R1 support code forward and the time I looked at it suggested there'd be a lot of work. If that can happen, the stock SVr4 networking code build might be a breeze. > Not really. If you look at 2.11BSD you'll see that to put the BSD Networking stack into it had to use both separate I&D space, but also needed to use one of the 'extra' priv levels to effectively run most of the BSD stack in an extra privileged process. Warner > Apologies in advance if this has already been broached and completed > and/or if this is a fool’s errand due to size constraints. And yes of > course the easy thing to do is just run v6 or BSD. But then we miss out on > the mind-blowing facet :) > > thx > jake From tuhs at tuhs.org Thu Dec 18 15:16:02 2025 From: tuhs at tuhs.org (Heinz Lycklama via TUHS) Date: Wed, 17 Dec 2025 21:16:02 -0800 Subject: [TUHS] update pdp11 system five? In-Reply-To: <5bd8f7f0-3e32-46e6-bd06-6eae610a4785@technologists.com> References: <9340A8B4-5AAB-4F85-9500-EDE1B1BDA0B7@gmail.com> <202512171541.5BHFf5eD022558@freefriends.org> <20251217154917.GC26336@mcvoy.com> <5bd8f7f0-3e32-46e6-bd06-6eae610a4785@technologists.com> Message-ID: <695bfa03-8b99-405c-8d32-d3cda6942661@osta.com> ISC did provide Streams, but I really cannot recall with which release. On 12/17/2025 9:40 AM, Charles H Sauer (he/him) via TUHS wrote: > On 12/17/2025 9:49 AM, Larry McVoy via TUHS wrote: >> On Wed, Dec 17, 2025 at 08:41:05AM -0700, Arnold Robbins via TUHS wrote: >>> IIRC the last official System V release that ran on the PDP-11 was >>> S5R2, but I don't remember for sure. >>> >>> S5R3 had STREAMS networking. >> >> I don't believe that's correct.  SCO's S5R3 did not, I had to port it. > > Strange -- I thought all SVR3 had STREAMS. I'm pretty sure Dell's SVR3 > had STREAMS, as provided by ISC. Heinz probably could be definitive. > From tuhs at tuhs.org Fri Dec 19 00:11:40 2025 From: tuhs at tuhs.org (Douglas McIlroy via TUHS) Date: Thu, 18 Dec 2025 09:11:40 -0500 Subject: [TUHS] sequence of groff's preprocessors in man page vs bug tracker Message-ID: In the same semi-serious vein as the original observation, I suggest that the preprocessors be listed in reverse topological order of execution precedence. In general this will bring the more important preprocessors to the top of the list. Unfortunately, it is not an unambiguous recipe; it does not settle the relative order of preprocessors that never interact, e.g. chem and grap. And, although it gathers the precedence rules in one place, the reverse order may cause some cognitive dissonance. Doug From tuhs at tuhs.org Sat Dec 20 06:14:03 2025 From: tuhs at tuhs.org (Matt Day via TUHS) Date: Fri, 19 Dec 2025 13:14:03 -0700 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: Rob Ricci updated today on his Mastodon page: > The attempt to read the UNIX V4 tape is underway! > I'm told "there is data" but I honestly don't know what that means yet. -- https://discuss.systems/@ricci/115747843169814700 The process is being filmed by a TV news crew and some footage is already available. On Thu, Nov 6, 2025 at 3:42 PM Rob Pike via TUHS wrote: > https://phanpy.social/#/hachyderm.io/s/115504720323483804 > > From mastodon: > > > Rob Ricci > ricci at discuss.systems > > While cleaning a storage room, our staff found this tape containing #UNIX > v4 from Bell Labs, circa 1973 > > Apparently no other complete copies are known to exist: > gunkies.org/wiki/UNIX_Fourth_E > > > We have arranged to deliver it to the Computer History Museum > From tuhs at tuhs.org Sat Dec 20 11:53:30 2025 From: tuhs at tuhs.org (Angelo Papenhoff via TUHS) Date: Sat, 20 Dec 2025 02:53:30 +0100 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: I extracted it: http://squoze.net/UNIX/v4/ aap On 19/12/25, Matt Day via TUHS wrote: > Rob Ricci updated today on his Mastodon page: > > The attempt to read the UNIX V4 tape is underway! > > I'm told "there is data" but I honestly don't know what that means yet. > -- https://discuss.systems/@ricci/115747843169814700 > The process is being filmed by a TV news crew and some footage is already > available. > > > On Thu, Nov 6, 2025 at 3:42 PM Rob Pike via TUHS wrote: > > > https://phanpy.social/#/hachyderm.io/s/115504720323483804 > > > > From mastodon: > > > > > > Rob Ricci > > ricci at discuss.systems > > > > While cleaning a storage room, our staff found this tape containing #UNIX > > v4 from Bell Labs, circa 1973 > > > > Apparently no other complete copies are known to exist: > > gunkies.org/wiki/UNIX_Fourth_E > > > > > > We have arranged to deliver it to the Computer History Museum > > From tuhs at tuhs.org Sat Dec 20 12:04:31 2025 From: tuhs at tuhs.org (Thalia Archibald via TUHS) Date: Sat, 20 Dec 2025 02:04:31 +0000 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: <1B35FD60-29DC-4AF2-9F3B-7C4F8DC7944E@archibald.dev> Very exciting day at the CHM archives! As you already found, I uploaded it at https://archive.org/details/utah_unix_v4_raw. Looking forward to booting it on my PiDP-11! Thalia From tuhs at tuhs.org Sat Dec 20 12:41:42 2025 From: tuhs at tuhs.org (=?utf-8?q?Cameron_M=C3=AD=C4=8Be=C3=A1l_Tyre_via_TUHS?=) Date: Sat, 20 Dec 2025 02:41:42 +0000 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: Hello Angelo, Thank you and everyone else who has been involved so far. Christmas came early! A couple of screenshots of me rumbling around in the extracted files. I should be in bed so this is all I did. It is 2:40 am here. https://ibb.co/fYFX8NsR The ed binary, file created June 10 1974, 14:37 https://ibb.co/BKKwybZ3 The first 40 lines of source code of the C compiler These images will auto-delete in one month. Have a wonderful weekend, everyone! Cameron On Saturday, December 20th, 2025 at 1:53 AM, Angelo Papenhoff via TUHS wrote: > > > I extracted it: http://squoze.net/UNIX/v4/ > > aap From tuhs at tuhs.org Sat Dec 20 12:50:18 2025 From: tuhs at tuhs.org (Angelo Papenhoff via TUHS) Date: Sat, 20 Dec 2025 03:50:18 +0100 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: It's booting too! added info to the readme. =uboot k unix login: root # ls bin dev etc lib mnt tmp unix usr # aap On 20/12/25, Angelo Papenhoff via TUHS wrote: > I extracted it: http://squoze.net/UNIX/v4/ > > aap > > On 19/12/25, Matt Day via TUHS wrote: > > Rob Ricci updated today on his Mastodon page: > > > The attempt to read the UNIX V4 tape is underway! > > > I'm told "there is data" but I honestly don't know what that means yet. > > -- https://discuss.systems/@ricci/115747843169814700 > > The process is being filmed by a TV news crew and some footage is already > > available. > > > > > > On Thu, Nov 6, 2025 at 3:42 PM Rob Pike via TUHS wrote: > > > > > https://phanpy.social/#/hachyderm.io/s/115504720323483804 > > > > > > From mastodon: > > > > > > > > > Rob Ricci > > > ricci at discuss.systems > > > > > > While cleaning a storage room, our staff found this tape containing #UNIX > > > v4 from Bell Labs, circa 1973 > > > > > > Apparently no other complete copies are known to exist: > > > gunkies.org/wiki/UNIX_Fourth_E > > > > > > > > > We have arranged to deliver it to the Computer History Museum > > > From tuhs at tuhs.org Sat Dec 20 14:28:04 2025 From: tuhs at tuhs.org (Matt Day via TUHS) Date: Fri, 19 Dec 2025 21:28:04 -0700 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: Cool, thanks Angelo.. I got it running easily thanks to your efforts. At a glance it looks like Fifth Edition to me... all the files are timestamped June 10 - 12, 1974. Commands present include col, dd, diff, eqn, glob, lpr, msh, neqn, pwd, spell, and tee -- according to https://dspinellis.github.io/unix-history-man/man1.html those commands first appeared in Fifth Edition and were not present in Fourth. On Fri, Dec 19, 2025 at 6:53 PM Angelo Papenhoff via TUHS wrote: > I extracted it: http://squoze.net/UNIX/v4/ > > aap > > On 19/12/25, Matt Day via TUHS wrote: > > Rob Ricci updated today on his Mastodon page: > > > The attempt to read the UNIX V4 tape is underway! > > > I'm told "there is data" but I honestly don't know what that means yet. > > -- https://discuss.systems/@ricci/115747843169814700 > > The process is being filmed by a TV news crew and some footage is already > > available. > > > > > > On Thu, Nov 6, 2025 at 3:42 PM Rob Pike via TUHS wrote: > > > > > https://phanpy.social/#/hachyderm.io/s/115504720323483804 > > > > > > From mastodon: > > > > > > > > > Rob Ricci > > > ricci at discuss.systems > > > > > > While cleaning a storage room, our staff found this tape containing > #UNIX > > > v4 from Bell Labs, circa 1973 > > > > > > Apparently no other complete copies are known to exist: > > > gunkies.org/wiki/UNIX_Fourth_E > > > > > > > > > We have arranged to deliver it to the Computer History Museum > > > > From tuhs at tuhs.org Sat Dec 20 14:35:45 2025 From: tuhs at tuhs.org (Matt Day via TUHS) Date: Fri, 19 Dec 2025 21:35:45 -0700 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: Correction: the commands present in Angelo's copy of the v4 tape include: dd, diff, glob, lpr, msh, pwd, tee Not present in Angelo's copy of the v4 tape: col, eqn, neqn, spell Present in Fifth Edition: all of the above, according to https://dspinellis.github.io/unix-history-man/man1.html On Fri, Dec 19, 2025 at 9:28 PM Matt Day wrote: > Cool, thanks Angelo.. I got it running easily thanks to your efforts. > > At a glance it looks like Fifth Edition to me... all the files are > timestamped June 10 - 12, 1974. Commands present include col, dd, diff, > eqn, glob, lpr, msh, neqn, pwd, spell, and tee -- according to > https://dspinellis.github.io/unix-history-man/man1.html those commands > first appeared in Fifth Edition and were not present in Fourth. > > On Fri, Dec 19, 2025 at 6:53 PM Angelo Papenhoff via TUHS > wrote: > >> I extracted it: http://squoze.net/UNIX/v4/ >> >> aap >> >> On 19/12/25, Matt Day via TUHS wrote: >> > Rob Ricci updated today on his Mastodon page: >> > > The attempt to read the UNIX V4 tape is underway! >> > > I'm told "there is data" but I honestly don't know what that means >> yet. >> > -- https://discuss.systems/@ricci/115747843169814700 >> > The process is being filmed by a TV news crew and some footage is >> already >> > available. >> > >> > >> > On Thu, Nov 6, 2025 at 3:42 PM Rob Pike via TUHS wrote: >> > >> > > https://phanpy.social/#/hachyderm.io/s/115504720323483804 >> > > >> > > From mastodon: >> > > >> > > >> > > Rob Ricci >> > > ricci at discuss.systems >> > > >> > > While cleaning a storage room, our staff found this tape containing >> #UNIX >> > > v4 from Bell Labs, circa 1973 >> > > >> > > Apparently no other complete copies are known to exist: >> > > gunkies.org/wiki/UNIX_Fourth_E >> > > >> > > >> > > We have arranged to deliver it to the Computer History Museum >> > > >> > From tuhs at tuhs.org Sat Dec 20 14:59:18 2025 From: tuhs at tuhs.org (Thalia Archibald via TUHS) Date: Sat, 20 Dec 2025 04:59:18 +0000 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: <7EDB2F32-8677-4A8A-8A9E-A7EB4B5CCED9@archibald.dev> On Fri, Dec 19, 2025 at 9:28 PM Matt Day wrote: > At a glance it looks like Fifth Edition to me... all the files are > timestamped June 10 - 12, 1974. We received the tape in June 1974, due to delay in printing more documentation. The documentation we received must have been for V4, since the label says to “see manual for format”, written later by Jay Lepreau, the OS researcher in whose documents it was found. Furthermore, the blue embossed label on the side says UNIX V4 DIST, which I presume was put there by Bell Labs. https://archive.org/details/thompson_to_newell_1974-05-31 UNIX wasn’t versioned as we know it today. In the early days, when you wanted to cut a tape, you’d ask Ken if it was a good day—whether the system was relatively bug-free—and copy off the research machine. The manuals were versioned and you got whatever the last one was. The V5 manual is dated June 1974 (anyone know a better date?), same as our tape. I suspect the next manual was finished days or weeks after this was sent, so this may perhaps be the most extreme drift in an extant copy of UNIX between the manual features and the shipped features. The Dennis_v5 distribution on TUHS is dated 21 March 1975, just two months away from the May 1975 release of the V6 manual. (Which explains why I found it so close to V6 when I studied it.) The Utah “V4” is a much cleaner V5 than the V5 we have. I’ve had been saying It’s probably V5 minus a tiny bit, which turned out to be quite true. Thalia From tuhs at tuhs.org Sat Dec 20 17:15:59 2025 From: tuhs at tuhs.org (Matt Day via TUHS) Date: Sat, 20 Dec 2025 00:15:59 -0700 Subject: [TUHS] unix v4 tape found In-Reply-To: <7EDB2F32-8677-4A8A-8A9E-A7EB4B5CCED9@archibald.dev> References: <7EDB2F32-8677-4A8A-8A9E-A7EB4B5CCED9@archibald.dev> Message-ID: On Fri, Dec 19, 2025 at 9:59 PM Thalia Archibald wrote: > The V5 manual is dated June 1974 (anyone know a better date?), same as our tape. The V5 man page for sort(1) is dated 6/11/74: http://squoze.net/UNIX/v5man/man1/sort The Utah tape's usr/source/s2/sort.c is dated 6/10/1974 and lacks support for the -b and -t flags that are documented in the 6/11/74 man page. The Fifth Edition UNIX Programmer's Manual (Dennis_v5) includes these man pages, but these commands are missing from the Utah tape: - eqn(1) dated 2/22/74 - neqn(1) dated 4/30/74 - spell(1) dated 2/26/74 - col(6) dated 5/20/74 (this list is probably incomplete) Thank you, Thalia! On Fri, Dec 19, 2025 at 9:59 PM Thalia Archibald wrote: > On Fri, Dec 19, 2025 at 9:28 PM Matt Day wrote: > > At a glance it looks like Fifth Edition to me... all the files are > > timestamped June 10 - 12, 1974. > > We received the tape in June 1974, due to delay in printing more > documentation. > The documentation we received must have been for V4, since the label says > to > “see manual for format”, written later by Jay Lepreau, the OS researcher in > whose documents it was found. Furthermore, the blue embossed label on the > side > says UNIX V4 DIST, which I presume was put there by Bell Labs. > https://archive.org/details/thompson_to_newell_1974-05-31 > > UNIX wasn’t versioned as we know it today. In the early days, when you > wanted to > cut a tape, you’d ask Ken if it was a good day—whether the system was > relatively > bug-free—and copy off the research machine. The manuals were versioned and > you > got whatever the last one was. > > The V5 manual is dated June 1974 (anyone know a better date?), same as our > tape. > I suspect the next manual was finished days or weeks after this was sent, > so > this may perhaps be the most extreme drift in an extant copy of UNIX > between the > manual features and the shipped features. > > The Dennis_v5 distribution on TUHS is dated 21 March 1975, just two months > away > from the May 1975 release of the V6 manual. (Which explains why I found it > so > close to V6 when I studied it.) The Utah “V4” is a much cleaner V5 than > the V5 > we have. I’ve had been saying It’s probably V5 minus a tiny bit, which > turned > out to be quite true. > > Thalia > From tuhs at tuhs.org Sat Dec 20 19:09:16 2025 From: tuhs at tuhs.org (segaloco via TUHS) Date: Sat, 20 Dec 2025 09:09:16 +0000 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: On Friday, December 19th, 2025 at 17:53, Angelo Papenhoff via TUHS wrote: > I extracted it: http://squoze.net/UNIX/v4/ > > aap Some observations: - In this init(1), /etc/getty is called as the only member of a single member array "com". In the current V5, /etc/getty is instead provided as the only execution target for dfork. The V5 manual adds that a line in /etc/ttys indicates the specifiable program in this array. The V6 manual removes again this idea of /etc/getty being one of many possible targets in a table, making this a more V5-ish init(1) than the current V5 copy. Additionally, the V4 manual indicates that init(1) still pulls TTYs from an internal table. This init(1) is using /etc/ttys. Finally, utmp is still in /tmp in this tape, whereas V6 sees it moved to /etc, as is the case in the current V5. - The copy of wc(1) present matches the V5 manual page in that it includes a number of options. - Stty(1) here also is more V5-ish in that it does not implement the hup options nor erase, kill, or ek. Just a few, but some areas marking this as matching V5 well. Probably tail end of calling it "V4" at all. - Matt G. V4->V5 /usr/man diff - https://gitlab.com/segaloco/mandiff/-/compare/v4...v5?from_project_id=44843539 V5->V6 /usr/man diff - https://gitlab.com/segaloco/mandiff/-/compare/v5...v6?from_project_id=44843539 From tuhs at tuhs.org Sat Dec 20 22:14:51 2025 From: tuhs at tuhs.org (Diomidis Spinellis via TUHS) Date: Sat, 20 Dec 2025 14:14:51 +0200 Subject: [TUHS] V4 in tv news In-Reply-To: References: Message-ID: <8e513be1-2841-45ee-ad1f-c01e65d841c2@aueb.gr> On 08-Dec-25 07:30, Thalia Archibald via TUHS wrote: […] > If any of you know of good early UNIX sources, particularly > around 1973–1974, please let me know! Consider the late Michael Sean Mahoney's project to create a history of Unix based on interviews with the system's developers. Mahoney, who died in 2008, was a Professor of the History of Science at Princeton University. His site seems to be down, but I have archived the project at https://github.com/dspinellis/oral-history-of-unix/ and https://doi.org/10.5281/zenodo.2525530. Diomidis - https://www.spinellis.gr From tuhs at tuhs.org Sat Dec 20 22:41:17 2025 From: tuhs at tuhs.org (=?utf-8?q?Cameron_M=C3=AD=C4=8Be=C3=A1l_Tyre_via_TUHS?=) Date: Sat, 20 Dec 2025 12:41:17 +0000 Subject: [TUHS] unix v4 tape found Message-ID: The first model year for the Ford Mustang is commonly described as 1964½ even though the vehicles were given 1965 VIN codes. This was because they appeared in the early half of 1964. Perhaps V...½ would be a good way of referencing some physical UNIX distributions, e.g. V4½ to describe a distribution that is neither fully V4 nor V5. Using the ½ as opposed to a .5 would hint, just like the earliest Ford Mustangs, that "we" gave the moniker, not "them". Best regards, Cameron Sent from my Verizon, Samsung Galaxy smartphone -------- Original Message -------- On Saturday, 12/20/25 at 04:59 Thalia Archibald via TUHS wrote: On Fri, Dec 19, 2025 at 9:28 PM Matt Day wrote: > At a glance it looks like Fifth Edition to me... all the files are > timestamped June 10 - 12, 1974. We received the tape in June 1974, due to delay in printing more documentation. The documentation we received must have been for V4, since the label says to “see manual for format”, written later by Jay Lepreau, the OS researcher in whose documents it was found. Furthermore, the blue embossed label on the side says UNIX V4 DIST, which I presume was put there by Bell Labs. https://archive.org/details/thompson_to_newell_1974-05-31 UNIX wasn’t versioned as we know it today... Sent from my Verizon, Samsung Galaxy smartphone From tuhs at tuhs.org Sat Dec 20 22:46:24 2025 From: tuhs at tuhs.org (Angelo Papenhoff via TUHS) Date: Sat, 20 Dec 2025 13:46:24 +0100 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: nice observations. Just a quick glance at numbers revealed around 4.1k lines of diffs between matching ken/*.c and dmr/*.c files between nsys and v4, and 1.1k between v4 and v5. Most changes are about the register keyword however, and prefixing these variables with 'r'. De-registerifying the v4 kernel would allow for a fairer comparison to nsys. It would also be a good step towards adding pipe support into the nsys kernel, which it is unfortunately lacking. That would make it an actually usable complete UNIX kernel i believe. aap On 20/12/25, segaloco wrote: > On Friday, December 19th, 2025 at 17:53, Angelo Papenhoff via TUHS wrote: > > > I extracted it: http://squoze.net/UNIX/v4/ > > > > aap > > Some observations: > > - In this init(1), /etc/getty is called as the only member of a single > member array "com". In the current V5, /etc/getty is instead provided > as the only execution target for dfork. The V5 manual adds that a > line in /etc/ttys indicates the specifiable program in this array. > The V6 manual removes again this idea of /etc/getty being one of many > possible targets in a table, making this a more V5-ish init(1) than > the current V5 copy. Additionally, the V4 manual indicates that > init(1) still pulls TTYs from an internal table. This init(1) is > using /etc/ttys. Finally, utmp is still in /tmp in this tape, whereas > V6 sees it moved to /etc, as is the case in the current V5. > > - The copy of wc(1) present matches the V5 manual page in that it > includes a number of options. > > - Stty(1) here also is more V5-ish in that it does not implement the hup > options nor erase, kill, or ek. > > Just a few, but some areas marking this as matching V5 well. Probably > tail end of calling it "V4" at all. > > - Matt G. > > V4->V5 /usr/man diff - https://gitlab.com/segaloco/mandiff/-/compare/v4...v5?from_project_id=44843539 > V5->V6 /usr/man diff - https://gitlab.com/segaloco/mandiff/-/compare/v5...v6?from_project_id=44843539 From tuhs at tuhs.org Sun Dec 21 01:37:17 2025 From: tuhs at tuhs.org (Al Kossow via TUHS) Date: Sat, 20 Dec 2025 07:37:17 -0800 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: <2a10d342-7059-88e2-0100-6ae415764057@bitsavers.org> The two bad blocks on the image have been fixed, so I've pushed that .tap file to http://bitsavers.org/bits/ATT/U-Utah_V4_tape along with a contents.txt of what all the blocks were. From tuhs at tuhs.org Sun Dec 21 04:13:58 2025 From: tuhs at tuhs.org (Thalia Archibald via TUHS) Date: Sat, 20 Dec 2025 18:13:58 +0000 Subject: [TUHS] unix v4 tape found In-Reply-To: <2a10d342-7059-88e2-0100-6ae415764057@bitsavers.org> References: <2a10d342-7059-88e2-0100-6ae415764057@bitsavers.org> Message-ID: <2CEE6D4C-B21A-4025-A104-111ECF351019@archibald.dev> On Dec 20, 2025, at 07:37, Al Kossow via TUHS wrote: > The two bad blocks on the image have been fixed, so > I've pushed that .tap file to > http://bitsavers.org/bits/ATT/U-Utah_V4_tape > along with a contents.txt of what all the blocks > were. Thank you, Al! For this and the whole event yesterday! How did you fix the blocks? Did you use the analog waveform? Yufeng Gao got in touch with me and has also recovered those blocks. He did it first by eyeballing the ASCII, then by hacking readtape to dump the CRC and LRC of each block and fixing a mistake in one of them. His matches yours. Thalia From tuhs at tuhs.org Sun Dec 21 04:45:35 2025 From: tuhs at tuhs.org (Al Kossow via TUHS) Date: Sat, 20 Dec 2025 10:45:35 -0800 Subject: [TUHS] unix v4 tape found In-Reply-To: <2CEE6D4C-B21A-4025-A104-111ECF351019@archibald.dev> References: <2a10d342-7059-88e2-0100-6ae415764057@bitsavers.org> <2CEE6D4C-B21A-4025-A104-111ECF351019@archibald.dev> Message-ID: On 12/20/25 10:13 AM, Thalia Archibald via TUHS wrote: > How did you fix the blocks? Did you use the analog waveform? I didn't do the work, I just used what was posted on the cctlk discord this morning From tuhs at tuhs.org Sun Dec 21 10:17:35 2025 From: tuhs at tuhs.org (Adam Thornton via TUHS) Date: Sat, 20 Dec 2025 17:17:35 -0700 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: "cubic" exists both here and on the v7 I have. There don't seem to be sources, though, at least not in the obvious places. I went looking online, and didn't find much besides Patashnik's 1980 paper (the acknowledgements there are a lovely rogues' gallery of exactly who'd you'd expect). I think (based on the oral history interview) that cubic here is probably a Ken program. I was playing with it, but trying to keep the board state in my head rather than on paper (which was a mistake) and was intrigued by the fact that it found a forced loss for me and laid it out for me. My question is, do the sources to "cubic" exist anywhere? Adam From tuhs at tuhs.org Sun Dec 21 10:24:27 2025 From: tuhs at tuhs.org (Jim Capp via TUHS) Date: Sat, 20 Dec 2025 19:24:27 -0500 Subject: [TUHS] V4 in tv news In-Reply-To: <8e513be1-2841-45ee-ad1f-c01e65d841c2@aueb.gr> References: <8e513be1-2841-45ee-ad1f-c01e65d841c2@aueb.gr> Message-ID: <76DF071F-D3B5-4383-9C78-13A44BE644BF@anteil.com> > On Dec 20, 2025, at 7:15 AM, Diomidis Spinellis via TUHS wrote: > >  > > On 08-Dec-25 07:30, Thalia Archibald via TUHS wrote: > […] >> If any of you know of good early UNIX sources, particularly >> around 1973–1974, please let me know! > Consider the late Michael Sean Mahoney's project to create a history of Unix based on interviews with the system's developers. Mahoney, who died in 2008, was a Professor of the History of Science at Princeton University. His site seems to be down, but I have archived the project at https://github.com/dspinellis/oral-history-of-unix/ and https://doi.org/10.5281/zenodo.2525530. > > Diomidis - https://www.spinellis.gr This is awesome! From tuhs at tuhs.org Sun Dec 21 10:42:23 2025 From: tuhs at tuhs.org (Adam Thornton via TUHS) Date: Sat, 20 Dec 2025 17:42:23 -0700 Subject: [TUHS] V4 in tv news In-Reply-To: <8e513be1-2841-45ee-ad1f-c01e65d841c2@aueb.gr> References: <8e513be1-2841-45ee-ad1f-c01e65d841c2@aueb.gr> Message-ID: Thanks for doing that. I was one of Prof. Mahoney's grad students. On Sat, Dec 20, 2025 at 5:21 AM Diomidis Spinellis via TUHS wrote: > > > On 08-Dec-25 07:30, Thalia Archibald via TUHS wrote: > […] > > If any of you know of good early UNIX sources, particularly > > around 1973–1974, please let me know! > Consider the late Michael Sean Mahoney's project to create a history of > Unix based on interviews with the system's developers. Mahoney, who > died in 2008, was a Professor of the History of Science at Princeton > University. His site seems to be down, but I have archived the project > at https://github.com/dspinellis/oral-history-of-unix/ and > https://doi.org/10.5281/zenodo.2525530. > > Diomidis - https://www.spinellis.gr > From tuhs at tuhs.org Sun Dec 21 12:05:55 2025 From: tuhs at tuhs.org (Thalia Archibald via TUHS) Date: Sun, 21 Dec 2025 02:05:55 +0000 Subject: [TUHS] Send me your UNIX license agreements! Message-ID: <75C561BD-210A-4601-B8AC-350B7BB3FB9A@archibald.dev> Hi everyone, I am collecting old UNIX license agreements to be able to establish the dates of early installations. So far, I have licenses for Regents of the University of California (1973-12-01), KU Nijmegen (1974-12-01), University of Manitoba (1975-02-01), and KU Leuven (1977-09-01), as well as 1982 license specimens found by segaloco and others from the ‘80s. https://archive.org/details/@archibits/lists/1/unix-licenses?sort=date https://github.com/thaliaarchi/unix-history/blob/main/licenses/README.md They did not track license dates at Bell Labs, so this is the next-best way to determine that. Before we found documentation on the receipt of the Utah UNIX V4 tape, I narrowed down when we could have received it from tracking down software agreements for institutions around Utah in Ken’s list of licensees. It’s also interesting to see the evolution of licenses. https://github.com/thaliaarchi/unix-history/blob/main/lists/README.md If you have UNIX software agreements that are public or I could publish, please send them my way! I’m also looking for first-hand accounts of such dates. Thalia From tuhs at tuhs.org Sun Dec 21 15:20:22 2025 From: tuhs at tuhs.org (Yufeng Gao via TUHS) Date: Sun, 21 Dec 2025 05:20:22 +0000 Subject: [TUHS] unix v4 tape found In-Reply-To: References: <2a10d342-7059-88e2-0100-6ae415764057@bitsavers.org> <2CEE6D4C-B21A-4025-A104-111ECF351019@archibald.dev> Message-ID: > From: Thalia Archibald > Yufeng Gao got in touch with me and has also recovered those blocks. He did it > first by eyeballing the ASCII, then by hacking readtape to dump the CRC and LRC > of each block and fixing a mistake in one of them. His matches yours. Indeed. Forgot to CC the TUHS list in my email, my apologies. Details are as follows: CRC and LRC of the two bad blocks are: Block 4281: CRC=341, LRC=290 Block 4367: CRC=302, LRC=445 Verifying that the CRC and LRC of the fixed records indeed match: $ crc good_4281.bin CRC=341 LRC=290 $ crc good_4367.bin CRC=302 LRC=445 > From: Al Kossow > I didn't do the work, I just used what was posted on the cctlk discord this morning I was the one who posted it on the CCMP Discord (in #general). I attached the CRCs and LRCs as well as a program to verify them in my email to Thalia. Sincerely, Yufeng From tuhs at tuhs.org Tue Dec 23 00:54:04 2025 From: tuhs at tuhs.org (Diomidis Spinellis via TUHS) Date: Mon, 22 Dec 2025 16:54:04 +0200 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: <322782d7-3dc7-40a0-a3c7-f7b5a6d5c499@aueb.gr> The finding and successful extraction of the tape is an amazing development! I updated the Unix History Repository on GitHub [1] to incorporate the tape's contents [2]. As with other snapshots, the commit timestamps are derived from the file timestamps while the commit authors are derived from a manually-created map file [3]. I updated the author map file based on information I had gathered for preceding and following Unix Research editions. I explicitly put "ken,dmr" in all source code files where I lacked author information (this is also the default introduced via a .* regular expression) to mark missing details. If you can fill any gaps or submit corrections (via a list reply, private email, or GitHub PR), I'll incorporate your input in the next repo update. In particular I think it would important to credit the authors of the following. /usr/sno: SNOBOL III /usr/source/s3/fp*.s: Floating point simulator /usr/source/s3/qsort.s /usr/source/s4/{cos,gamma,hypot,sin}.s: Math library [1] https://github.com/dspinellis/unix-history-repo [2] https://github.com/dspinellis/unix-history-repo/tree/Research-V4-Snapshot-Development [3] https://github.com/dspinellis/unix-history-make/blob/master/src/author-path/Research-V4 Diomidis - https://www.spinellis.gr On 20-Dec-25 03:53, Angelo Papenhoff via TUHS wrote: > I extracted it: http://squoze.net/UNIX/v4/ > > aap [...] >>> From mastodon: >>> >>> >>> Rob Ricci >>> ricci at discuss.systems >>> >>> While cleaning a storage room, our staff found this tape containing #UNIX >>> v4 from Bell Labs, circa 1973 >>> >>> Apparently no other complete copies are known to exist: >>> gunkies.org/wiki/UNIX_Fourth_E >>> >>> >>> We have arranged to deliver it to the Computer History Museum >>> From tuhs at tuhs.org Tue Dec 23 06:58:41 2025 From: tuhs at tuhs.org (Matt Day via TUHS) Date: Mon, 22 Dec 2025 13:58:41 -0700 Subject: [TUHS] unix v4 tape found In-Reply-To: <322782d7-3dc7-40a0-a3c7-f7b5a6d5c499@aueb.gr> References: <322782d7-3dc7-40a0-a3c7-f7b5a6d5c499@aueb.gr> Message-ID: In 2004, dmr said this about V4 "sno": > I think writing it was just a quick entertainment for Ken. -- https://www.tuhs.org/pipermail/tuhs/2004-June/004210.html On Mon, Dec 22, 2025 at 7:54 AM Diomidis Spinellis via TUHS wrote: > The finding and successful extraction of the tape is an amazing > development! I updated the Unix History Repository on GitHub [1] to > incorporate the tape's contents [2]. As with other snapshots, the > commit timestamps are derived from the file timestamps while the commit > authors are derived from a manually-created map file [3]. I updated the > author map file based on information I had gathered for preceding and > following Unix Research editions. I explicitly put "ken,dmr" in all > source code files where I lacked author information (this is also the > default introduced via a .* regular expression) to mark missing details. > If you can fill any gaps or submit corrections (via a list reply, > private email, or GitHub PR), I'll incorporate your input in the next > repo update. In particular I think it would important to credit the > authors of the following. > > > /usr/sno: SNOBOL III > /usr/source/s3/fp*.s: Floating point simulator > /usr/source/s3/qsort.s > /usr/source/s4/{cos,gamma,hypot,sin}.s: Math library > > > [1] https://github.com/dspinellis/unix-history-repo > [2] > > https://github.com/dspinellis/unix-history-repo/tree/Research-V4-Snapshot-Development > [3] > > https://github.com/dspinellis/unix-history-make/blob/master/src/author-path/Research-V4 > > > Diomidis - https://www.spinellis.gr > > On 20-Dec-25 03:53, Angelo Papenhoff via TUHS wrote: > > I extracted it: http://squoze.net/UNIX/v4/ > > > > aap > > [...] > > >>> From mastodon: > >>> > >>> > >>> Rob Ricci > >>> ricci at discuss.systems > >>> > >>> While cleaning a storage room, our staff found this tape containing > #UNIX > >>> v4 from Bell Labs, circa 1973 > >>> > >>> Apparently no other complete copies are known to exist: > >>> gunkies.org/wiki/UNIX_Fourth_E > >>> > >>> > >>> We have arranged to deliver it to the Computer History Museum > >>> > > From tuhs at tuhs.org Tue Dec 23 07:47:08 2025 From: tuhs at tuhs.org (Diomidis Spinellis via TUHS) Date: Mon, 22 Dec 2025 23:47:08 +0200 Subject: [TUHS] unix v4 tape found In-Reply-To: References: <322782d7-3dc7-40a0-a3c7-f7b5a6d5c499@aueb.gr> Message-ID: <8f9491f5-5335-4402-8799-ffe974105e25@aueb.gr> Thank you! I also got a separate authoritative confirmation: "definitely ken -- noone else ever touched it". I just pushed the updated the map file: https://github.com/dspinellis/unix-history-make/blob/master/src/author-path/Research-V4#L150 The corresponding history repo commits should appear in the next 24 hours. On 22-Dec-25 22:58, Matt Day wrote: > In 2004, dmr said this about V4 "sno": > > I think writing it was just a quick entertainment for Ken. > -- https://www.tuhs.org/pipermail/tuhs/2004-June/004210.html www.tuhs.org/pipermail/tuhs/2004-June/004210.html> > > On Mon, Dec 22, 2025 at 7:54 AM Diomidis Spinellis via TUHS > > wrote: > > The finding and successful extraction of the tape is an amazing > development!  I updated the Unix History Repository on GitHub [1] to > incorporate the tape's contents [2].  As with other snapshots, the > commit timestamps are derived from the file timestamps while the commit > authors are derived from a manually-created map file [3].  I updated > the > author map file based on information I had gathered for preceding and > following Unix Research editions.  I explicitly put "ken,dmr" in all > source code files where I lacked author information (this is also the > default introduced via a .* regular expression) to mark missing > details. >   If you can fill any gaps or submit corrections (via a list reply, > private email, or GitHub PR), I'll incorporate your input in the next > repo update.  In particular I think it would important to credit the > authors of the following. > > > /usr/sno: SNOBOL III > /usr/source/s3/fp*.s: Floating point simulator > /usr/source/s3/qsort.s > /usr/source/s4/{cos,gamma,hypot,sin}.s: Math library > > > [1] https://github.com/dspinellis/unix-history-repo github.com/dspinellis/unix-history-repo> > [2] > https://github.com/dspinellis/unix-history-repo/tree/Research-V4- > Snapshot-Development repo/tree/Research-V4-Snapshot-Development> > [3] > https://github.com/dspinellis/unix-history-make/blob/master/src/ > author-path/Research-V4 make/blob/master/src/author-path/Research-V4> > > > Diomidis - https://www.spinellis.gr > > On 20-Dec-25 03:53, Angelo Papenhoff via TUHS wrote: > > I extracted it: http://squoze.net/UNIX/v4/ UNIX/v4/> > > > > aap > > [...] > > >>>  From mastodon: > >>> > >>> > > >>> Rob Ricci > >>> ricci at discuss.systems discuss.systems/@ricci>> > >>> > >>> While cleaning a storage room, our staff found this tape > containing #UNIX > >>> tags/UNIX>> v4 from Bell Labs, circa 1973 > >>> > >>> Apparently no other complete copies are known to exist: > >>> gunkies.org/wiki/UNIX_Fourth_E UNIX_Fourth_E> > >>> gunkies.org/wiki/UNIX_Fourth_Edition>> > >>> > >>> We have arranged to deliver it to the Computer History Museum > >>> > From tuhs at tuhs.org Tue Dec 23 09:00:46 2025 From: tuhs at tuhs.org (Douglas McIlroy via TUHS) Date: Mon, 22 Dec 2025 18:00:46 -0500 Subject: [TUHS] unix v4 tape found Message-ID: > I think it would important to credit the authors of > /usr/sno: SNOBOL III > /usr/source/s3/fp*.s: Floating point simulator > /usr/source/s3/qsort.s > /usr/source/s4/{cos,gamma,hypot,sin}.s: Math library Lee McMahon wrote qsort. Bob Morris wrote the math library. It was quite good, with one exception. When Peter McIlroy redid -lm for BSD, with freshly derived approximations, he found that Morris's gamma function was utter nonsense near the poles. Doug From tuhs at tuhs.org Tue Dec 23 09:07:01 2025 From: tuhs at tuhs.org (Douglas McIlroy via TUHS) Date: Mon, 22 Dec 2025 18:07:01 -0500 Subject: [TUHS] README Message-ID: READMEs are everywhere now, but I am not aware of any before Dennis began to put them in Unix source directories. Does anyone know whether he invented the idea or borrowed it? Doug From tuhs at tuhs.org Tue Dec 23 10:19:10 2025 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Mon, 22 Dec 2025 19:19:10 -0500 Subject: [TUHS] README In-Reply-To: References: Message-ID: I suspect it was a good idea that was independently created in multiple places. I remember seeing them in PDP-6/10 arpanet would which I was introduced before discovering Unix. Sent from a handheld expect more typos than usual On Mon, Dec 22, 2025 at 6:07 PM Douglas McIlroy via TUHS wrote: > READMEs are everywhere now, but I am not aware of any before Dennis > began to put them in Unix source directories. Does anyone know whether > he invented the idea or borrowed it? > > Doug > From tuhs at tuhs.org Tue Dec 23 10:45:24 2025 From: tuhs at tuhs.org (Larry McVoy via TUHS) Date: Mon, 22 Dec 2025 16:45:24 -0800 Subject: [TUHS] README In-Reply-To: References: Message-ID: <20251223004524.GW26336@mcvoy.com> I wasn't exposed to the Unix sources early but I did READMEs in stuff that I open sourced so someone must have showed me that. Or it's just an obvious thing to do. I did upper case README so it listed first in ls. On Mon, Dec 22, 2025 at 07:19:10PM -0500, Clem Cole via TUHS wrote: > I suspect it was a good idea that was independently created in multiple > places. I remember seeing them in PDP-6/10 arpanet would which I was > introduced before discovering Unix. > > Sent from a handheld expect more typos than usual > > > On Mon, Dec 22, 2025 at 6:07???PM Douglas McIlroy via TUHS > wrote: > > > READMEs are everywhere now, but I am not aware of any before Dennis > > began to put them in Unix source directories. Does anyone know whether > > he invented the idea or borrowed it? > > > > Doug > > -- --- Larry McVoy Retired to fishing http://www.mcvoy.com/lm/boat From tuhs at tuhs.org Tue Dec 23 12:14:59 2025 From: tuhs at tuhs.org (Marc Donner via TUHS) Date: Mon, 22 Dec 2025 21:14:59 -0500 Subject: [TUHS] README In-Reply-To: <20251223004524.GW26336@mcvoy.com> References: <20251223004524.GW26336@mcvoy.com> Message-ID: https://en.wikipedia.org/wiki/README#History ===== mindthegapdialogs.com north-fork.info On Mon, Dec 22, 2025 at 7:45 PM Larry McVoy via TUHS wrote: > I wasn't exposed to the Unix sources early but I did READMEs in stuff that > I > open sourced so someone must have showed me that. Or it's just an obvious > thing to do. I did upper case README so it listed first in ls. > > On Mon, Dec 22, 2025 at 07:19:10PM -0500, Clem Cole via TUHS wrote: > > I suspect it was a good idea that was independently created in multiple > > places. I remember seeing them in PDP-6/10 arpanet would which I was > > introduced before discovering Unix. > > > > Sent from a handheld expect more typos than usual > > > > > > On Mon, Dec 22, 2025 at 6:07???PM Douglas McIlroy via TUHS < > tuhs at tuhs.org> > > wrote: > > > > > READMEs are everywhere now, but I am not aware of any before Dennis > > > began to put them in Unix source directories. Does anyone know whether > > > he invented the idea or borrowed it? > > > > > > Doug > > > > > -- > --- > Larry McVoy Retired to fishing > http://www.mcvoy.com/lm/boat > From tuhs at tuhs.org Tue Dec 23 12:20:16 2025 From: tuhs at tuhs.org (Al Kossow via TUHS) Date: Mon, 22 Dec 2025 18:20:16 -0800 Subject: [TUHS] README In-Reply-To: References: <20251223004524.GW26336@mcvoy.com> Message-ID: <3c531103-6411-5b74-0c30-83b98d454716@bitsavers.org> On 12/22/25 6:14 PM, Marc Donner via TUHS wrote: > https://en.wikipedia.org/wiki/README#History isn't this a reference to Alice in Wonderland? From tuhs at tuhs.org Tue Dec 23 12:41:56 2025 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Mon, 22 Dec 2025 21:41:56 -0500 Subject: [TUHS] README In-Reply-To: <3c531103-6411-5b74-0c30-83b98d454716@bitsavers.org> References: <20251223004524.GW26336@mcvoy.com> <3c531103-6411-5b74-0c30-83b98d454716@bitsavers.org> Message-ID: Below Sent from a handheld expect more typos than usual On Mon, Dec 22, 2025 at 9:20 PM Al Kossow via TUHS wrote: > On 12/22/25 6:14 PM, Marc Donner via TUHS wrote: > > https://en.wikipedia.org/wiki/README#History > > isn't this a reference to Alice in Wonderland? Certainly that was my experience. But as I said, I think it was an idea that was created a few times. I also remember files called “READ THIS 1ST” and similar names. Another thing that was sometimes done was put a few zeros In name so the file was the sorted to be the first file you would see when a directory was listed. BTW way the reference to Berkeley SPICE for the PDP-10 is telling. We never had a PDP-10 in the UCB CAD group. We had CDC6600 originally and an IBM 360. Ellis Cohen was the original author. And yes certainly he had README files in the basic distribution (FORTRAN 80 column card images) from the Industrial Liaison Office (ILO) in the early 1970s explaining how to compile and build it [FWIW: the ILO a few years later would be how Bill Joy distributed the Berkeley (UNIX) Software Distribution. From tuhs at tuhs.org Tue Dec 23 16:36:07 2025 From: tuhs at tuhs.org (Lars Brinkhoff via TUHS) Date: Tue, 23 Dec 2025 06:36:07 +0000 Subject: [TUHS] README In-Reply-To: (Douglas McIlroy via TUHS's message of "Mon, 22 Dec 2025 18:07:01 -0500") References: Message-ID: <7w3451so14.fsf@junk.nocrew.org> Douglas McIlroy via TUHS writes: > READMEs are everywhere now, but I am not aware of any before Dennis > began to put them in Unix source directories. Does anyone know whether > he invented the idea or borrowed it? The question has come up in some other places. I offered some information from ITS history where the file is usually called -READ- -THIS-; however there's a SHRDLU; %READ ME which up in 1973. https://softwareengineering.stackexchange.com/questions/96966/origin-of-readme https://retrocomputing.stackexchange.com/questions/6059/when-did-readme-files-start-showing-up-in-software From tuhs at tuhs.org Tue Dec 23 23:12:48 2025 From: tuhs at tuhs.org (Diomidis Spinellis via TUHS) Date: Tue, 23 Dec 2025 15:12:48 +0200 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: With the tape now available as a synthetic Git repo [1], I run git-blame on the V4 and V5 files [2]. V4's composition is as follows in terms of lines: v4 75676 v3 6590 v2 168 So, a lot of new material and about 10% coming from earlier editions. V5 is as follows: v5 11181 v4 52238 v3 3296 v2 168 So, quite close to V5, but not entirely so. [1] https://github.com/dspinellis/unix-history-repo [2] for ref in Research-V4-Snapshot-Development \ Research-V5-Snapshot-Development ; do echo $ref git ls-tree -r --name-only $ref | grep -Ev 'README|LICENSE|\.pdf|\.ref' | xargs -I '{}' git blame -M -M -C -C $ref -- '{}' | sort | uniq -c | awk '{("git show " $2 "| awk '\''/Synthesized-from:/{print $2}'\''") | getline ver; total[ver] += $1 } END {for (v in total) print v, total[v]}' done On 20-Dec-25 06:28, Matt Day via TUHS wrote: > Cool, thanks Angelo.. I got it running easily thanks to your efforts. > > At a glance it looks like Fifth Edition to me... all the files are > timestamped June 10 - 12, 1974. Commands present include col, dd, diff, > eqn, glob, lpr, msh, neqn, pwd, spell, and tee -- according to > https://dspinellis.github.io/unix-history-man/man1.html those commands > first appeared in Fifth Edition and were not present in Fourth. From tuhs at tuhs.org Tue Dec 23 23:31:26 2025 From: tuhs at tuhs.org (Angelo Papenhoff via TUHS) Date: Tue, 23 Dec 2025 14:31:26 +0100 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: Quick update: Since that mail i spent almost all of my time inside the UNIX kernel trying to get nsys to work correctly. Putting in pipes was the easy bit, but i've been banging my head against the wall for days now. Something is leaking IO blocks so eventually the system gets stuck waiting for bfreelist. My test case is 'cc -c *.c' in /usr/sys/ken. Replacing bawrite and bdwrite with just plain bwrite gets me somewhat further in that process but eventually i hit a 'panic: trap', although it seems to be a bit non-deterministic. I even gotten into an endless panic loop. The good thing is that in this case there seem to be no bfreelist leaks at all so far. So definitely asynch IO is not working correctly, and i suspect this is related to synchronization bugs in the nsys code in general. I also started de-registerifying the v4 kernel to be aesthetically more similar to nsys. Perhaps the two approaches will somehow meet in the middle. Lots of frustration unfortunately, but on the other hand i never understood the unix kernel as deeply as i do now. At some point you *are* expected to understand this :) cheers, aap On 20/12/25, Angelo Papenhoff via TUHS wrote: > nice observations. Just a quick glance at numbers revealed around 4.1k > lines of diffs between matching ken/*.c and dmr/*.c files between nsys and v4, > and 1.1k between v4 and v5. Most changes are about the register keyword > however, and prefixing these variables with 'r'. De-registerifying the v4 > kernel would allow for a fairer comparison to nsys. It would also be a > good step towards adding pipe support into the nsys kernel, which it is > unfortunately lacking. That would make it an actually usable complete UNIX > kernel i believe. > > aap > > On 20/12/25, segaloco wrote: > > On Friday, December 19th, 2025 at 17:53, Angelo Papenhoff via TUHS wrote: > > > > > I extracted it: http://squoze.net/UNIX/v4/ > > > > > > aap > > > > Some observations: > > > > - In this init(1), /etc/getty is called as the only member of a single > > member array "com". In the current V5, /etc/getty is instead provided > > as the only execution target for dfork. The V5 manual adds that a > > line in /etc/ttys indicates the specifiable program in this array. > > The V6 manual removes again this idea of /etc/getty being one of many > > possible targets in a table, making this a more V5-ish init(1) than > > the current V5 copy. Additionally, the V4 manual indicates that > > init(1) still pulls TTYs from an internal table. This init(1) is > > using /etc/ttys. Finally, utmp is still in /tmp in this tape, whereas > > V6 sees it moved to /etc, as is the case in the current V5. > > > > - The copy of wc(1) present matches the V5 manual page in that it > > includes a number of options. > > > > - Stty(1) here also is more V5-ish in that it does not implement the hup > > options nor erase, kill, or ek. > > > > Just a few, but some areas marking this as matching V5 well. Probably > > tail end of calling it "V4" at all. > > > > - Matt G. > > > > V4->V5 /usr/man diff - https://gitlab.com/segaloco/mandiff/-/compare/v4...v5?from_project_id=44843539 > > V5->V6 /usr/man diff - https://gitlab.com/segaloco/mandiff/-/compare/v5...v6?from_project_id=44843539 From tuhs at tuhs.org Wed Dec 24 02:34:05 2025 From: tuhs at tuhs.org (Paul Winalski via TUHS) Date: Tue, 23 Dec 2025 11:34:05 -0500 Subject: [TUHS] README In-Reply-To: <3c531103-6411-5b74-0c30-83b98d454716@bitsavers.org> References: <20251223004524.GW26336@mcvoy.com> <3c531103-6411-5b74-0c30-83b98d454716@bitsavers.org> Message-ID: On Mon, Dec 22, 2025 at 9:26 PM Al Kossow via TUHS wrote: > > isn't this a reference to Alice in Wonderland? > > I always thought so. A reference to the bottle Alice finds that is labeled "Drink Me". I suspect README has been invented independently several times. Back in 1974 when I was an undergrad a fellow hacker wrote a S/360 DOS logical transient (privileged overlay code in the operating system) that, when called returned control to the caller in supervisor (privileged) mode. We used it as the basis for many a hack. All transient routines were named starting with $$B. This one was called $$BTRYME. -Paul W. From tuhs at tuhs.org Wed Dec 24 07:58:45 2025 From: tuhs at tuhs.org (Rob Pike via TUHS) Date: Wed, 24 Dec 2025 08:58:45 +1100 Subject: [TUHS] README In-Reply-To: References: <20251223004524.GW26336@mcvoy.com> <3c531103-6411-5b74-0c30-83b98d454716@bitsavers.org> Message-ID: Now we need to find the first README about READMEs. -rob From tuhs at tuhs.org Wed Dec 24 11:22:53 2025 From: tuhs at tuhs.org (steve jenkin via TUHS) Date: Wed, 24 Dec 2025 12:22:53 +1100 Subject: [TUHS] unix v4 tape found In-Reply-To: References: Message-ID: <37A35DF4-FC08-4702-AE71-3E707E57707F@canb.auug.org.au> Liam Proven has written a nice piece in ’The Register’, includes history, why finding this piece is important & more. Links to videos and many CHM and other sources, including Oral Histories. LP wonders “Is this a Christmas Miracle?” It’s certainly the original CSRC spirit of “Cooperation and Collegiality” alive & well. Link & extract at end. > On 7 Nov 2025, at 09:44, Al Kossow via TUHS wrote: > > On 11/6/25 2:41 PM, Rob Pike via TUHS wrote: > >> We have arranged to deliver it to the Computer History Museum > > another day, another project. =============== Original message on-line =============== UNIX V4 tape successfully recovered: First ever version of UNIX written in C is running again Crucial early evolutionary step found, imaged, and ... amazingly … works Liam Proven Tue 23 Dec 2025 Computer History Museum software curator Al Kossow has successfully retrieved the contents of the over-half-a-century old tape found at the University of Utah last month. Last month, we wrote about the remarkable discovery of a forgotten tape with a lost early version of Unix, found by Professor Robert Ricci at the Kahlert School of Computing at the University of Utah. At the time, we quoted the redoubtable Kossow, who also runs Bitsavers, as saying that it "has a pretty good chance of being recoverable.” The data was recovered using the readtape program by the Computer History Museum's Len Shustek. So, in the recovered files on the Internet Archive, you can see there's a 1.6 gigabyte file created from a tape that only held 40 MB or so of data. You probably don't want to download that. Fortunately, Angelo Papenhoff offers a processed version, complete with a README telling you how to run it. =============== -- From tuhs at tuhs.org Wed Dec 24 17:53:18 2025 From: tuhs at tuhs.org (George Michaelson via TUHS) Date: Wed, 24 Dec 2025 17:53:18 +1000 Subject: [TUHS] README In-Reply-To: References: <20251223004524.GW26336@mcvoy.com> <3c531103-6411-5b74-0c30-83b98d454716@bitsavers.org> Message-ID: https://books.google.com/ngrams/graph?content=README.TXT&year_start=1950&year_end=1980&corpus=en&smoothing=3&case_insensitive=true Suggests 1956/7 I wonder if DECUS tape submission instructions recommended? From tuhs at tuhs.org Thu Dec 25 04:04:31 2025 From: tuhs at tuhs.org (Ron Natalie via TUHS) Date: Wed, 24 Dec 2025 13:04:31 -0500 Subject: [TUHS] README In-Reply-To: References: Message-ID: In my short stint working for Unipress I did field a call that someone stated that the README program wasn’t working. From tuhs at tuhs.org Fri Dec 26 13:44:20 2025 From: tuhs at tuhs.org (Jonathan Gray via TUHS) Date: Fri, 26 Dec 2025 14:44:20 +1100 Subject: [TUHS] cubic (was: unix v4 tape found) In-Reply-To: References: Message-ID: On Sat, Dec 20, 2025 at 05:17:35PM -0700, Adam Thornton via TUHS wrote: > "cubic" exists both here and on the v7 I have. There don't seem to be > sources, though, at least not in the obvious places. > > I went looking online, and didn't find much besides Patashnik's 1980 paper > (the acknowledgements there are a lovely rogues' gallery of exactly who'd > you'd expect). > > I think (based on the oral history interview) that cubic here is probably a > Ken program. "Some of the things Ken wrote on the PDP-7 by 1969 include ... A graphical-interface 3D tic-tac-toe program (4x4x4) run with the light pen." https://www.nokia.com/bell-labs/about/dennis-m-ritchie/ken-games.html "His early work has been on MULTICS, file system design and simulation, computer chess, cubic (3D) tic-tac-toe, and language implementation." K Thompson, The Unix Command Language, 1976 https://archive.org/details/theunixcommandlanguage > > I was playing with it, but trying to keep the board state in my head rather > than on paper (which was a mistake) and was intrigued by the fact that it > found a forced loss for me and laid it out for me. > > My question is, do the sources to "cubic" exist anywhere? > > Adam https://github.com/DoctorWkt/pdp7-unix/blob/master/src/cmd/ttt1.s jms messg; i>;040;w>;i>;n>;0 https://github.com/DoctorWkt/pdp7-unix/blob/master/src/cmd/ttt2.s tuhs Applications/Dennis_Tapes/dmr_tapes.tgz games/ttt-play/ games/ttt-solve/ contain .s files with '3d tic-tac-toe' comments 'I have a force win' is in games/ttt-play/ot1.s From tuhs at tuhs.org Fri Dec 26 13:54:15 2025 From: tuhs at tuhs.org (Adam Thornton via TUHS) Date: Thu, 25 Dec 2025 20:54:15 -0700 Subject: [TUHS] cubic (was: unix v4 tape found) In-Reply-To: References: Message-ID: Thank you for the excellent Christmas gift! On Thu, Dec 25, 2025 at 8:44 PM Jonathan Gray wrote: > On Sat, Dec 20, 2025 at 05:17:35PM -0700, Adam Thornton via TUHS wrote: > > "cubic" exists both here and on the v7 I have. There don't seem to be > > sources, though, at least not in the obvious places. > > > > I went looking online, and didn't find much besides Patashnik's 1980 > paper > > (the acknowledgements there are a lovely rogues' gallery of exactly who'd > > you'd expect). > > > > I think (based on the oral history interview) that cubic here is > probably a > > Ken program. > > "Some of the things Ken wrote on the PDP-7 by 1969 include > ... > A graphical-interface 3D tic-tac-toe program (4x4x4) run with the light > pen." > https://www.nokia.com/bell-labs/about/dennis-m-ritchie/ken-games.html > > "His early work has been on MULTICS, file system design and simulation, > computer chess, cubic (3D) tic-tac-toe, and language implementation." > K Thompson, The Unix Command Language, 1976 > https://archive.org/details/theunixcommandlanguage > > > > > I was playing with it, but trying to keep the board state in my head > rather > > than on paper (which was a mistake) and was intrigued by the fact that it > > found a forced loss for me and laid it out for me. > > > > My question is, do the sources to "cubic" exist anywhere? > > > > Adam > > https://github.com/DoctorWkt/pdp7-unix/blob/master/src/cmd/ttt1.s > jms messg; i>;040;w>;i>;n>;0 > https://github.com/DoctorWkt/pdp7-unix/blob/master/src/cmd/ttt2.s > > tuhs Applications/Dennis_Tapes/dmr_tapes.tgz > games/ttt-play/ > games/ttt-solve/ > contain .s files with '3d tic-tac-toe' comments > > 'I have a force win' is in > games/ttt-play/ot1.s > From tuhs at tuhs.org Fri Dec 26 17:44:44 2025 From: tuhs at tuhs.org (Ken Thompson via TUHS) Date: Thu, 25 Dec 2025 23:44:44 -0800 Subject: [TUHS] cubic (was: unix v4 tape found) In-Reply-To: References: Message-ID: patasnik said he solved cubit as a forced win for the first player. he submitted it to elwyn berlekamp for verification and publication. patasnik had solved it by hand up to a point where a computer could make forcing moves to a win. elwyn asked me to verify the solution. i did -- it was a win. a by-product was a program that would find forced wins. from there a program was easy. i haven't thought about this since then. thank you for bringing it up. On Thu, Dec 25, 2025 at 7:54 PM Adam Thornton via TUHS wrote: > Thank you for the excellent Christmas gift! > > On Thu, Dec 25, 2025 at 8:44 PM Jonathan Gray wrote: > > > On Sat, Dec 20, 2025 at 05:17:35PM -0700, Adam Thornton via TUHS wrote: > > > "cubic" exists both here and on the v7 I have. There don't seem to be > > > sources, though, at least not in the obvious places. > > > > > > I went looking online, and didn't find much besides Patashnik's 1980 > > paper > > > (the acknowledgements there are a lovely rogues' gallery of exactly > who'd > > > you'd expect). > > > > > > I think (based on the oral history interview) that cubic here is > > probably a > > > Ken program. > > > > "Some of the things Ken wrote on the PDP-7 by 1969 include > > ... > > A graphical-interface 3D tic-tac-toe program (4x4x4) run with the light > > pen." > > https://www.nokia.com/bell-labs/about/dennis-m-ritchie/ken-games.html > > > > "His early work has been on MULTICS, file system design and simulation, > > computer chess, cubic (3D) tic-tac-toe, and language implementation." > > K Thompson, The Unix Command Language, 1976 > > https://archive.org/details/theunixcommandlanguage > > > > > > > > I was playing with it, but trying to keep the board state in my head > > rather > > > than on paper (which was a mistake) and was intrigued by the fact that > it > > > found a forced loss for me and laid it out for me. > > > > > > My question is, do the sources to "cubic" exist anywhere? > > > > > > Adam > > > > https://github.com/DoctorWkt/pdp7-unix/blob/master/src/cmd/ttt1.s > > jms messg; i>;040;w>;i>;n>;0 > > https://github.com/DoctorWkt/pdp7-unix/blob/master/src/cmd/ttt2.s > > > > tuhs Applications/Dennis_Tapes/dmr_tapes.tgz > > games/ttt-play/ > > games/ttt-solve/ > > contain .s files with '3d tic-tac-toe' comments > > > > 'I have a force win' is in > > games/ttt-play/ot1.s > > > From tuhs at tuhs.org Sat Dec 27 03:15:53 2025 From: tuhs at tuhs.org (Phil Budne via TUHS) Date: Fri, 26 Dec 2025 12:15:53 -0500 Subject: [TUHS] ken's sno soma solver? In-Reply-To: References: <322782d7-3dc7-40a0-a3c7-f7b5a6d5c499@aueb.gr> Message-ID: <202512261715.5BQHFsqT027097@ultimate.com> Matt Day wrote: > In 2004, dmr said this about V4 "sno": > > I think writing it was just a quick entertainment for Ken. > -- https://www.tuhs.org/pipermail/tuhs/2004-June/004210.html In the archived message Dennis wrote: > The "application" that has survived is a > 1-page program that solves the Soma (or Instant Insanity) > puzzle. As a snobol(*) enthusiast, I can't help asking (21 years after the fact), whether the soma solver survives? (*) I did a port of the original Macro SNOBOL4 to C [http://www.regressive.org/snobol/csnobol4/], and since I learned on SNOBOL4, SNOBOL3 has always seemed like alien technology, and I'd love to see an actual program! From tuhs at tuhs.org Sat Dec 27 08:00:53 2025 From: tuhs at tuhs.org (Jonathan Gray via TUHS) Date: Sat, 27 Dec 2025 09:00:53 +1100 Subject: [TUHS] ken's sno soma solver? In-Reply-To: <202512261715.5BQHFsqT027097@ultimate.com> References: <322782d7-3dc7-40a0-a3c7-f7b5a6d5c499@aueb.gr> <202512261715.5BQHFsqT027097@ultimate.com> Message-ID: On Fri, Dec 26, 2025 at 12:15:53PM -0500, Phil Budne via TUHS wrote: > Matt Day wrote: > > In 2004, dmr said this about V4 "sno": > > > I think writing it was just a quick entertainment for Ken. > > -- https://www.tuhs.org/pipermail/tuhs/2004-June/004210.html > > In the archived message Dennis wrote: > > The "application" that has survived is a > > 1-page program that solves the Soma (or Instant Insanity) > > puzzle. > > As a snobol(*) enthusiast, I can't help asking (21 years after the > fact), whether the soma solver survives? > > (*) I did a port of the original Macro SNOBOL4 to C > [http://www.regressive.org/snobol/csnobol4/], and since I learned on > SNOBOL4, SNOBOL3 has always seemed like alien technology, and I'd > love to see an actual program! tuhs Applications/Dennis_Tapes/dmr_tapes.tgz ken-sky contains soma.sno and ch.sno From tuhs at tuhs.org Mon Dec 29 05:37:32 2025 From: tuhs at tuhs.org (Rich Salz via TUHS) Date: Sun, 28 Dec 2025 14:37:32 -0500 Subject: [TUHS] OpenBSD stories Message-ID: Miod Vallet, an OpenBSD developer for 15 years, is starting to write some stories. Some might find them interesting: http://miod.online.fr/software/openbsd/stories/index.html. He has three, so far, and more to come. From tuhs at tuhs.org Mon Dec 29 07:43:36 2025 From: tuhs at tuhs.org (Brian "Beej" Hall via TUHS) Date: Sun, 28 Dec 2025 13:43:36 -0800 Subject: [TUHS] Photo of old Unix manuals Message-ID: Hey all, I'm looking for an impressive photograph of a bookshelf containing the Unix man page books. (I remember when I had an AT&T Unix PC it had a bunch of them. Not sure which weighed more--the PC or the books. Sadly I no longer possess it.) In the first week of the OS class I'm teaching, I talk about the different sections of the Unix manual and I thought it would be fun to show the students what it was like when you had to read "books". :) My searching is turning up nothing. Does anyone have one they wouldn't mind me sharing? -Beej -- Brian "Beej Jorgensen" Hall beej at beej.us https://beej.us/ From tuhs at tuhs.org Mon Dec 29 08:12:36 2025 From: tuhs at tuhs.org (segaloco via TUHS) Date: Sun, 28 Dec 2025 22:12:36 +0000 Subject: [TUHS] Photo of old Unix manuals In-Reply-To: References: Message-ID: On Sunday, December 28th, 2025 at 13:44, Brian "Beej" Hall via TUHS wrote: > Hey all, > > I'm looking for an impressive photograph of a bookshelf containing the > Unix man page books. (I remember when I had an AT&T Unix PC it had a > bunch of them. Not sure which weighed more--the PC or the books. Sadly I > no longer possess it.) > > In the first week of the OS class I'm teaching, I talk about the > different sections of the Unix manual and I thought it would be fun to > show the students what it was like when you had to read "books". :) > > My searching is turning up nothing. Does anyone have one they wouldn't > mind me sharing? > > -Beej > > -- > Brian "Beej Jorgensen" Hall > beej at beej.us https://beej.us/ Not sure if this is what you're looking for but here are some of my UNIX shelves: https://imgur.com/a/0XawAnx First image is V6 (+BTL Lions Commentary), V7, 3.0, 4.1, 5.0 (BTL), and early standards (/usr/group, SVID 1, X/Open, POSIX draft) and SVID 2-based User's Manual. Second image is 5.0 (WECo), SVR1 (+WWB), 4.2BSD, 4.3BSD, and SVR4. If you are looking for a particular staging I can set up specific entries for a "group photo", I've also got 4.1 and 4.4BSD sets as well as a smattering of SVR2 stuff, other X/Open, POSIX, and SVID issues, etc. - Matt G. From tuhs at tuhs.org Mon Dec 29 08:38:37 2025 From: tuhs at tuhs.org (Thalia Archibald via TUHS) Date: Sun, 28 Dec 2025 22:38:37 +0000 Subject: [TUHS] Photo of old Unix manuals In-Reply-To: References: Message-ID: On Dec 28, 2025, at 14:12, segaloco via TUHS wrote: > First image is V6 (+BTL Lions Commentary) I thought that Lions’ Commentary wasn’t distributed after BTL took over distribution? Do you know whether there’s a scan of that version anywhere? The BTL takeover was announced in the March 1978 issue of ;login:. https://archive.org/details/login_march-1978/page/n1/mode/1up Thalia From tuhs at tuhs.org Mon Dec 29 08:57:05 2025 From: tuhs at tuhs.org (segaloco via TUHS) Date: Sun, 28 Dec 2025 22:57:05 +0000 Subject: [TUHS] Photo of old Unix manuals In-Reply-To: References: Message-ID: <-RY8UG8lpwIabK6tLNho_zNbsgXacsn9nplASrw_6-PX2qI3SVQ1IQpFWOzxpQ2vmX3CTv2NsYkQyrcKNEH_UBunY1dPJ0oAWDt8SghQsh4=@protonmail.com> On Sunday, December 28th, 2025 at 14:38, Thalia Archibald via TUHS wrote: > On Dec 28, 2025, at 14:12, segaloco via TUHS tuhs at tuhs.org wrote: > > > First image is V6 (+BTL Lions Commentary) > > > I thought that Lions’ Commentary wasn’t distributed after BTL took over distribution? > Do you know whether there’s a scan of that version anywhere? > > The BTL takeover was announced in the March 1978 issue of ;login:. > https://archive.org/details/login_march-1978/page/n1/mode/1up > > Thalia Covers here: https://imgur.com/a/z59pJjO Should be the same content, just different appearance. - Matt G. From tuhs at tuhs.org Mon Dec 29 09:08:26 2025 From: tuhs at tuhs.org (segaloco via TUHS) Date: Sun, 28 Dec 2025 23:08:26 +0000 Subject: [TUHS] Did System V Really Prevent 5BSD? Message-ID: So the prevailing narrative I've heard is that 4.1BSD was so-named at AT&T's demand to avoid confusion with System V. However, the timeline leads me to question this. 4.1BSD is given as having been issued first June 1981. Of the Bell lineage, my 4.1 manual also gives June 1981 as a publication date. 5.0 wouldn't release inside AT&T for another year, and System V manuals give 1983 (i.e. post-divestiture) publication dates (although they went to print before, many have Bell logos on the cover still.) In any case, if System V as a product didn't exist until sometime 1983, how did the fear of confusion with this name prevent the naming of 5BSD in 1981, two years prior? Was System V as a future trademark already well accepted as a given in 1981? - Matt G. From tuhs at tuhs.org Mon Dec 29 09:33:46 2025 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Sun, 28 Dec 2025 18:33:46 -0500 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: References: Message-ID: Right idea, wrong release. 4.1BSD was the “fastvax” update from 4.0BSD and really was an increment from the previous release. 4.1 was triggered by the Stanford push to use VMS for performance reasons as the official ARPA supported OS for the VAX. It was CSRG’s naming of 4.2BSD were things got strange. When you look at functionality the changes from 4.1 to 4.2 were huge [Henry Spencer’s famous line: “4.2 is just like Unix, only different.”]. As for why CSRG chose to call it 4.2 over 5.0BSD is a bit lost to time. I don’t think ATT was in any position to “demand” anything. Being at UCB and my memory of the time was that the folks at CSRG chose to stay away from the number 5 because they wanted to be sure their work was clearly distinguished from ATTs — and while the short lived 3.0BSD release was for the Vax, most people by then associated the numbering 2xBSD to be PDP-11 and 4xBSD to be Vax. Only later did CSRG start to add a suffix (Reno/Tahoe) to distinguish other processors as they branched out. But when 4.2BSD was released, CRSG had a DARPA contract to support UNIX on the VAX [BTW, they did not have the contract for IP/TCP, BBN had that contract and the stories of the issues this caused had been documented/discussed here in the past]. Sent from a handheld expect more typos than usual On Sun, Dec 28, 2025 at 6:08 PM segaloco via TUHS wrote: > So the prevailing narrative I've heard is that 4.1BSD was so-named at > AT&T's demand to avoid confusion with System V. However, the timeline > leads me to question this. 4.1BSD is given as having been issued first > June 1981. Of the Bell lineage, my 4.1 manual also gives June 1981 as a > publication date. 5.0 wouldn't release inside AT&T for another year, and > System V manuals give 1983 (i.e. post-divestiture) publication dates > (although they went to print before, many have Bell logos on the cover > still.) > > In any case, if System V as a product didn't exist until sometime 1983, > how did the fear of confusion with this name prevent the naming of 5BSD in > 1981, two years prior? Was System V as a future trademark already well > accepted as a given in 1981? > > - Matt G. > From tuhs at tuhs.org Mon Dec 29 09:47:57 2025 From: tuhs at tuhs.org (Warner Losh via TUHS) Date: Sun, 28 Dec 2025 16:47:57 -0700 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: References: Message-ID: On Sun, Dec 28, 2025, 4:34 PM Clem Cole via TUHS wrote: > Right idea, wrong release. 4.1BSD was the “fastvax” update from 4.0BSD and > really was an increment from the previous release. 4.1 was triggered by the > Stanford push to use VMS for performance reasons as the official ARPA > supported OS for the VAX. It was CSRG’s naming of 4.2BSD were things got > strange. When you look at functionality the changes from 4.1 to 4.2 were > huge [Henry Spencer’s famous line: “4.2 is just like Unix, only > different.”]. > > As for why CSRG chose to call it 4.2 over 5.0BSD is a bit lost to time. I > don’t think ATT was in any position to “demand” anything. I'm pretty sure that 4.1 was going to be 5BSD. I've heard this story from Kirk several times. He has a listing he labeled 4.5BSD. It was between 4BSD and 5BSD when he wrote 4.5 on the spine. Whatcwe know today as 4.1BSD was going to be 5BSD. But when it came time to try to do the paperwork, AT&T Balked and insisted no 5BSD since it would be confused with the forth coming System V. I don't know the nuts and bolts of the back and forth to know if this was a really firm request, or if threats were involved. I've heard this tale from both Kirk and Eric... Warner Being at UCB and > my memory of the time was that the folks at CSRG chose to stay away from > the number 5 because they wanted to be sure their work was clearly > distinguished from ATTs — and while the short lived 3.0BSD release was for > the Vax, most people by then associated the numbering 2xBSD to be PDP-11 > and 4xBSD to be Vax. Only later did CSRG start to add a suffix > (Reno/Tahoe) to distinguish other processors as they branched out. But > when 4.2BSD was released, CRSG had a DARPA contract to support UNIX on the > VAX [BTW, they did not have the contract for IP/TCP, BBN had that contract > and the stories of the issues this caused had been documented/discussed > here in the past]. > > Sent from a handheld expect more typos than usual > > > On Sun, Dec 28, 2025 at 6:08 PM segaloco via TUHS wrote: > > > So the prevailing narrative I've heard is that 4.1BSD was so-named at > > AT&T's demand to avoid confusion with System V. However, the timeline > > leads me to question this. 4.1BSD is given as having been issued first > > June 1981. Of the Bell lineage, my 4.1 manual also gives June 1981 as a > > publication date. 5.0 wouldn't release inside AT&T for another year, and > > System V manuals give 1983 (i.e. post-divestiture) publication dates > > (although they went to print before, many have Bell logos on the cover > > still.) > > > > In any case, if System V as a product didn't exist until sometime 1983, > > how did the fear of confusion with this name prevent the naming of 5BSD > in > > 1981, two years prior? Was System V as a future trademark already well > > accepted as a given in 1981? > > > > - Matt G. > > > From tuhs at tuhs.org Mon Dec 29 11:12:19 2025 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Sun, 28 Dec 2025 20:12:19 -0500 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: References: Message-ID: Interesting. I need to talk to them as that can’t be right. I’ll believe they might have been a discussion WRT to calling it 5BSD [which would have made sense from a stand point of being the 5th Berkeley (UNIX) Software Distribution tape from the Berkeley Industrial Liaisons Office since the ILO tried to keep track of the different tapes that folks sent out [originally CAD tools like SPICE, SPLICE, MOTIS long before and of the UNIX distributions]. ATT would not have had anything to do with it. System III was released Nov 1981 and System V late 1984 - plus Jan 1, 1984 is when Judge Green breaks allows ATT to be in the computer business so ATT could not have had anything to say before then. Calling it 4.1 made sense because it really was a small incremental difference from 4.0. Where numbering got weird was years later with the 4.2 release from CSRG. The point is that ATTs Summit’s next release at that time frame (pre Judge Green) was Unix 3.0 (which would be renamed System III) while Berkeley was already at the 4.x level. And getting back to BSD the key differences between 4.0 and 4.1 are pretty small and the time between them was short (Oct 1980 and June 1981). The primary differences are the #ifdef FASTVAX stuff that Joy did over the winter after the dust up that the Stanford folks started in the fall 1980 - Joy had to demonstrate that Unix was just as fast as VMS (which had been written in assembler). He instrumented a bunch of the kernel and if a couple places dropped into assembly and got Unix to perform within a very small epsilon on everything that DARPA cared about. So the issue became that ATT nor DEC was supporting Unix. CSRG does not yet exist. Bob Fabry goes to ARPA with a proposal (and the results from Bill’s work they winter) that UCB create a team to fill that void (CSRG) - DARPA would create an advisory team for them and Berkeley would add the features that DARPA required. CSRG was not created until the Fall/Winter 81/82. [For a historical prospective, Stanford had counter proposed using DEC/VMS and Australian Wollongong’s Unix for VMS and the Tek/CMU IP/TCP stack for VMS - two commercial products and the later FOSS. As we know Fabry’s proposal was accepted]. On Sun, Dec 28, 2025 at 6:48 PM Warner Losh wrote: > > > On Sun, Dec 28, 2025, 4:34 PM Clem Cole via TUHS wrote: > >> Right idea, wrong release. 4.1BSD was the “fastvax” update from 4.0BSD >> and >> really was an increment from the previous release. 4.1 was triggered by >> the >> Stanford push to use VMS for performance reasons as the official ARPA >> supported OS for the VAX. It was CSRG’s naming of 4.2BSD were things got >> strange. When you look at functionality the changes from 4.1 to 4.2 were >> huge [Henry Spencer’s famous line: “4.2 is just like Unix, only >> different.”]. >> >> As for why CSRG chose to call it 4.2 over 5.0BSD is a bit lost to time. I >> don’t think ATT was in any position to “demand” anything. > > > I'm pretty sure that 4.1 was going to be 5BSD. I've heard this story from > Kirk several times. He has a listing he labeled 4.5BSD. It was between 4BSD > and 5BSD when he wrote 4.5 on the spine. Whatcwe know today as 4.1BSD was > going to be 5BSD. But when it came time to try to do the paperwork, AT&T > Balked and insisted no 5BSD since it would be confused with the forth > coming System V. I don't know the nuts and bolts of the back and forth to > know if this was a really firm request, or if threats were involved. > > I've heard this tale from both Kirk and Eric... > > Warner > > Being at UCB and >> my memory of the time was that the folks at CSRG chose to stay away from >> the number 5 because they wanted to be sure their work was clearly >> distinguished from ATTs — and while the short lived 3.0BSD release was for >> the Vax, most people by then associated the numbering 2xBSD to be PDP-11 >> and 4xBSD to be Vax. Only later did CSRG start to add a suffix >> (Reno/Tahoe) to distinguish other processors as they branched out. But >> when 4.2BSD was released, CRSG had a DARPA contract to support UNIX on the >> VAX [BTW, they did not have the contract for IP/TCP, BBN had that contract >> and the stories of the issues this caused had been documented/discussed >> here in the past]. >> >> Sent from a handheld expect more typos than usual >> >> >> On Sun, Dec 28, 2025 at 6:08 PM segaloco via TUHS wrote: >> >> > So the prevailing narrative I've heard is that 4.1BSD was so-named at >> > AT&T's demand to avoid confusion with System V. However, the timeline >> > leads me to question this. 4.1BSD is given as having been issued first >> > June 1981. Of the Bell lineage, my 4.1 manual also gives June 1981 as a >> > publication date. 5.0 wouldn't release inside AT&T for another year, >> and >> > System V manuals give 1983 (i.e. post-divestiture) publication dates >> > (although they went to print before, many have Bell logos on the cover >> > still.) >> > >> > In any case, if System V as a product didn't exist until sometime 1983, >> > how did the fear of confusion with this name prevent the naming of 5BSD >> in >> > 1981, two years prior? Was System V as a future trademark already well >> > accepted as a given in 1981? >> > >> > - Matt G. >> > >> > From tuhs at tuhs.org Mon Dec 29 11:32:09 2025 From: tuhs at tuhs.org (Luther Johnson via TUHS) Date: Sun, 28 Dec 2025 18:32:09 -0700 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: References: Message-ID: <51baf73c-b28b-c64b-1119-2aa2ad9f663d@makerlisp.com> If you look at the dates on the files in the TUHS Unix tree, pdp11v (System V r1 for PDP-11) files are dated 10/11/1983, so I've always thought that System V release 1 must have been then. For example: https://www.tuhs.org/cgi-bin/utree.pl?file=pdp11v/usr/src/cmd On 12/28/2025 06:12 PM, Clem Cole via TUHS wrote: > Interesting. I need to talk to them as that can’t be right. I’ll believe > they might have been a discussion WRT to calling it 5BSD [which would have > made sense from a stand point of being the 5th Berkeley (UNIX) Software > Distribution tape from the Berkeley Industrial Liaisons Office since the > ILO tried to keep track of the different tapes that folks sent out > [originally CAD tools like SPICE, SPLICE, MOTIS long before and of the UNIX > distributions]. > > ATT would not have had anything to do with it. System III was released Nov > 1981 and System V late 1984 - plus Jan 1, 1984 is when Judge Green breaks > allows ATT to be in the computer business so ATT could not have had > anything to say before then. > > Calling it 4.1 made sense because it really was a small incremental > difference from 4.0. Where numbering got weird was years later with the > 4.2 release from CSRG. > > The point is that ATTs Summit’s next release at that time frame (pre Judge > Green) was Unix 3.0 (which would be renamed System III) while Berkeley was > already at the 4.x level. > > And getting back to BSD the key differences between 4.0 and 4.1 are pretty > small and the time between them was short (Oct 1980 and June 1981). The > primary differences are the #ifdef FASTVAX stuff that Joy did over the > winter after the dust up that the Stanford folks started in the fall 1980 > - Joy had to demonstrate that Unix was just as fast as VMS (which had been > written in assembler). He instrumented a bunch of the kernel and if a > couple places dropped into assembly and got Unix to perform within a very > small epsilon on everything that DARPA cared about. So the issue became > that ATT nor DEC was supporting Unix. CSRG does not yet exist. > > Bob Fabry goes to ARPA with a proposal (and the results from Bill’s work > they winter) that UCB create a team to fill that void (CSRG) - DARPA would > create an advisory team for them and Berkeley would add the features that > DARPA required. CSRG was not created until the Fall/Winter 81/82. [For a > historical prospective, Stanford had counter proposed using DEC/VMS and > Australian Wollongong’s Unix for VMS and the Tek/CMU IP/TCP stack for VMS - > two commercial products and the later FOSS. As we know Fabry’s proposal > was accepted]. > > > On Sun, Dec 28, 2025 at 6:48 PM Warner Losh wrote: > >> >> On Sun, Dec 28, 2025, 4:34 PM Clem Cole via TUHS wrote: >> >>> Right idea, wrong release. 4.1BSD was the “fastvax” update from 4.0BSD >>> and >>> really was an increment from the previous release. 4.1 was triggered by >>> the >>> Stanford push to use VMS for performance reasons as the official ARPA >>> supported OS for the VAX. It was CSRG’s naming of 4.2BSD were things got >>> strange. When you look at functionality the changes from 4.1 to 4.2 were >>> huge [Henry Spencer’s famous line: “4.2 is just like Unix, only >>> different.”]. >>> >>> As for why CSRG chose to call it 4.2 over 5.0BSD is a bit lost to time. I >>> don’t think ATT was in any position to “demand” anything. >> >> I'm pretty sure that 4.1 was going to be 5BSD. I've heard this story from >> Kirk several times. He has a listing he labeled 4.5BSD. It was between 4BSD >> and 5BSD when he wrote 4.5 on the spine. Whatcwe know today as 4.1BSD was >> going to be 5BSD. But when it came time to try to do the paperwork, AT&T >> Balked and insisted no 5BSD since it would be confused with the forth >> coming System V. I don't know the nuts and bolts of the back and forth to >> know if this was a really firm request, or if threats were involved. >> >> I've heard this tale from both Kirk and Eric... >> >> Warner >> >> Being at UCB and >>> my memory of the time was that the folks at CSRG chose to stay away from >>> the number 5 because they wanted to be sure their work was clearly >>> distinguished from ATTs — and while the short lived 3.0BSD release was for >>> the Vax, most people by then associated the numbering 2xBSD to be PDP-11 >>> and 4xBSD to be Vax. Only later did CSRG start to add a suffix >>> (Reno/Tahoe) to distinguish other processors as they branched out. But >>> when 4.2BSD was released, CRSG had a DARPA contract to support UNIX on the >>> VAX [BTW, they did not have the contract for IP/TCP, BBN had that contract >>> and the stories of the issues this caused had been documented/discussed >>> here in the past]. >>> >>> Sent from a handheld expect more typos than usual >>> >>> >>> On Sun, Dec 28, 2025 at 6:08 PM segaloco via TUHS wrote: >>> >>>> So the prevailing narrative I've heard is that 4.1BSD was so-named at >>>> AT&T's demand to avoid confusion with System V. However, the timeline >>>> leads me to question this. 4.1BSD is given as having been issued first >>>> June 1981. Of the Bell lineage, my 4.1 manual also gives June 1981 as a >>>> publication date. 5.0 wouldn't release inside AT&T for another year, >>> and >>>> System V manuals give 1983 (i.e. post-divestiture) publication dates >>>> (although they went to print before, many have Bell logos on the cover >>>> still.) >>>> >>>> In any case, if System V as a product didn't exist until sometime 1983, >>>> how did the fear of confusion with this name prevent the naming of 5BSD >>> in >>>> 1981, two years prior? Was System V as a future trademark already well >>>> accepted as a given in 1981? >>>> >>>> - Matt G. >>>> From tuhs at tuhs.org Mon Dec 29 12:52:24 2025 From: tuhs at tuhs.org (Charles H Sauer (he/him) via TUHS) Date: Sun, 28 Dec 2025 20:52:24 -0600 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: <51baf73c-b28b-c64b-1119-2aa2ad9f663d@makerlisp.com> References: <51baf73c-b28b-c64b-1119-2aa2ad9f663d@makerlisp.com> Message-ID: <102fddde-d8f2-406f-bba5-3acb8d3d49b8@technologists.com> On 12/28/2025 7:32 PM, Luther Johnson via TUHS wrote: > If you look at the dates on the files in the TUHS Unix tree, pdp11v > (System V r1 for PDP-11) files are dated 10/11/1983, so I've always > thought that System V release 1 must have been then. For example: > > https://www.tuhs.org/cgi-bin/utree.pl?file=pdp11v/usr/src/cmd IIRC, ISC folks started working with System V with IBM AIX folks by mid-year 1983, presumably because they had access before final release. (Initial ISC/IBM efforts were System III.) -- voice: +1.512.784.7526 e-mail: sauer at technologists.com fax: +1.512.346.5240 Web: https://technologists.com/sauer/ Facebook/Google/LinkedIn/mas.to: CharlesHSauer From tuhs at tuhs.org Mon Dec 29 12:53:41 2025 From: tuhs at tuhs.org (Jon Forrest via TUHS) Date: Sun, 28 Dec 2025 18:53:41 -0800 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: References: Message-ID: Some very minor notes: On 12/28/25 5:12 PM, Clem Cole via TUHS wrote: > And getting back to BSD the key differences between 4.0 and 4.1 are pretty > small and the time between them was short (Oct 1980 and June 1981). The > primary differences are the #ifdef FASTVAX stuff that Joy did over the > winter after the dust up that the Stanford folks started in the fall 1980 All the talk about Stanford should make it clear that it was actually the Stanford Research Institute (SRI), not Stanford University. SRI was also one of the first 4 nodes on the Arpanet. > - Joy had to demonstrate that Unix was just as fast as VMS (which had been > written in assembler). He instrumented a bunch of the kernel and if a > couple places dropped into assembly and got Unix to perform within a very > small epsilon on everything that DARPA cared about. So the issue became > that ATT nor DEC was supporting Unix. CSRG does not yet exist. In the early 1980s David Kashtan at SRI wrote Eunics, which was a way to run Unix (I don't recall which version) on top of VMS. It was an emulation layer, not a virtual machine. (I'm not aware of any VMs that ran on top of VMS). I used an VMS executable version of 'vi' that worked just fine for most of my VMS career. Kashtan and Joy went back and forth for a while. I think both communities (VMS and Unix) benefited from their work. > [For a > historical prospective, Stanford had counter proposed using DEC/VMS and > Australian Wollongong’s Unix for VMS and the Tek/CMU IP/TCP stack for VMS - > two commercial products and the later FOSS. I don't recall how Kashtan's Eunice, done at SRI, became part of the Wollongong Group. I do know that Kashtan and Ken Adelman wrote a IP/TCP for VMS called Multinet that was quite popular for a while. I'm not aware of it ever becoming FOSS. (For a fascinating non-Unix anecdote, check out https://www.californiacoastline.org/streisand/lawsuit.html about how Adelman presumably used some of the money he made when he and Kashtan sold Multinet to Cisco to fund his successful lawsuit defense against Barbra Streisand.) Sorry to mention VMS so much. However, early Unix history was often buffeted by what was happening in VMS and DEC. Jon From tuhs at tuhs.org Mon Dec 29 13:55:47 2025 From: tuhs at tuhs.org (Heinz Lycklama via TUHS) Date: Sun, 28 Dec 2025 19:55:47 -0800 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: References: Message-ID: <5fb3d6a9-890c-4d24-88f9-7a98654370fd@osta.com> Regarding the reference to "Eunics" below - it was actually called "Eunice" - a product introduced by the Wollongong Group In 1981. We (ISC) also introduced a product that provided the UNIX environment running on top of the DEC VMS system in 1979. I took the lead on that project/product for ISC when I started with ISC in early 1978. It turned out to be a successful product by ISC for many years in the 1980's. Heinz On 12/28/2025 6:53 PM, Jon Forrest via TUHS wrote: > Some very minor notes: > > On 12/28/25 5:12 PM, Clem Cole via TUHS wrote: > >> And getting back to BSD the key differences between 4.0 and 4.1 are >> pretty >> small and the time between them was short (Oct 1980 and June 1981).  The >> primary differences are the #ifdef FASTVAX stuff that Joy did over the >> winter after the dust up that the Stanford folks started  in the fall >> 1980 > > All the talk about Stanford should make it clear that it was actually > the Stanford Research Institute (SRI), not Stanford University. > SRI was also one of the first 4 nodes on the Arpanet. > >> - Joy had to demonstrate that Unix was just as fast as VMS (which had >> been >> written in assembler). He instrumented a bunch of the kernel and if a >> couple places dropped into assembly and got Unix to perform within a >> very >> small epsilon on everything that DARPA cared about.  So the issue became >> that ATT nor DEC was supporting Unix. CSRG does not yet exist. > > In the early 1980s David Kashtan at SRI wrote Eunics, which was a way to > run Unix (I don't recall which version) on top of VMS. It was an > emulation layer, not a virtual machine. (I'm not aware of any VMs that > ran on top of VMS). I used an VMS executable version of 'vi' that worked > just fine for most of my VMS career. > > Kashtan and Joy went back and forth for a while. I think both > communities (VMS and Unix) benefited from their work. > >> [For a >> historical prospective, Stanford had counter proposed using DEC/VMS and >> Australian Wollongong’s Unix for VMS and the Tek/CMU IP/TCP stack for >> VMS - >> two commercial products and the later FOSS. > > I don't recall how Kashtan's Eunice, done at SRI, became part of the > Wollongong Group. I do know that Kashtan and Ken Adelman wrote a > IP/TCP for VMS called Multinet that was quite popular for a while. > I'm not aware of it ever becoming FOSS. > > (For a fascinating non-Unix anecdote, check out > https://www.californiacoastline.org/streisand/lawsuit.html about how > Adelman presumably used some of the money he made when he and Kashtan > sold Multinet to Cisco to fund his successful lawsuit defense against > Barbra Streisand.) > > Sorry to mention VMS so much. However, early Unix history was often > buffeted by what was happening in VMS and DEC. > > Jon > From tuhs at tuhs.org Mon Dec 29 14:09:27 2025 From: tuhs at tuhs.org (Jon Forrest via TUHS) Date: Sun, 28 Dec 2025 20:09:27 -0800 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: <5fb3d6a9-890c-4d24-88f9-7a98654370fd@osta.com> References: <5fb3d6a9-890c-4d24-88f9-7a98654370fd@osta.com> Message-ID: On 12/28/25 7:55 PM, Heinz Lycklama via TUHS wrote: > Regarding the reference to "Eunics" below - it was actually > called "Eunice" - Right. That was a stupid typo on my part. Jon From tuhs at tuhs.org Mon Dec 29 14:11:23 2025 From: tuhs at tuhs.org (Wesley Parish via TUHS) Date: Mon, 29 Dec 2025 17:11:23 +1300 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: <5fb3d6a9-890c-4d24-88f9-7a98654370fd@osta.com> References: <5fb3d6a9-890c-4d24-88f9-7a98654370fd@osta.com> Message-ID: <71af2301-7de7-4b4d-abdc-b5b119dce3ff@gmail.com> Regarding Eunice, was the source actually distributed in any way? And even if not, are there any source tapes still surviving? Wesley Parish On 29/12/2025 16:55, Heinz Lycklama via TUHS wrote: > Regarding the reference to "Eunics" below - it was actually > called "Eunice" - a product introduced by the Wollongong Group > In 1981. We (ISC) also introduced a product that provided the UNIX > environment running on top of the DEC VMS system in 1979. > I took the lead on that project/product for ISC when I started with > ISC in early 1978. It turned out to be a successful product by > ISC for many years in the 1980's. > > Heinz > > On 12/28/2025 6:53 PM, Jon Forrest via TUHS wrote: >> Some very minor notes: >> >> On 12/28/25 5:12 PM, Clem Cole via TUHS wrote: >> >>> And getting back to BSD the key differences between 4.0 and 4.1 are >>> pretty >>> small and the time between them was short (Oct 1980 and June 1981).  >>> The >>> primary differences are the #ifdef FASTVAX stuff that Joy did over the >>> winter after the dust up that the Stanford folks started  in the >>> fall 1980 >> >> All the talk about Stanford should make it clear that it was actually >> the Stanford Research Institute (SRI), not Stanford University. >> SRI was also one of the first 4 nodes on the Arpanet. >> >>> - Joy had to demonstrate that Unix was just as fast as VMS (which >>> had been >>> written in assembler). He instrumented a bunch of the kernel and if a >>> couple places dropped into assembly and got Unix to perform within a >>> very >>> small epsilon on everything that DARPA cared about.  So the issue >>> became >>> that ATT nor DEC was supporting Unix. CSRG does not yet exist. >> >> In the early 1980s David Kashtan at SRI wrote Eunics, which was a way to >> run Unix (I don't recall which version) on top of VMS. It was an >> emulation layer, not a virtual machine. (I'm not aware of any VMs that >> ran on top of VMS). I used an VMS executable version of 'vi' that worked >> just fine for most of my VMS career. >> >> Kashtan and Joy went back and forth for a while. I think both >> communities (VMS and Unix) benefited from their work. >> >>> [For a >>> historical prospective, Stanford had counter proposed using DEC/VMS and >>> Australian Wollongong’s Unix for VMS and the Tek/CMU IP/TCP stack >>> for VMS - >>> two commercial products and the later FOSS. >> >> I don't recall how Kashtan's Eunice, done at SRI, became part of the >> Wollongong Group. I do know that Kashtan and Ken Adelman wrote a >> IP/TCP for VMS called Multinet that was quite popular for a while. >> I'm not aware of it ever becoming FOSS. >> >> (For a fascinating non-Unix anecdote, check out >> https://www.californiacoastline.org/streisand/lawsuit.html about how >> Adelman presumably used some of the money he made when he and Kashtan >> sold Multinet to Cisco to fund his successful lawsuit defense against >> Barbra Streisand.) >> >> Sorry to mention VMS so much. However, early Unix history was often >> buffeted by what was happening in VMS and DEC. >> >> Jon >> > From tuhs at tuhs.org Mon Dec 29 15:26:22 2025 From: tuhs at tuhs.org (Larry McVoy via TUHS) Date: Sun, 28 Dec 2025 21:26:22 -0800 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: References: Message-ID: <20251229052622.GC12694@mcvoy.com> On Sun, Dec 28, 2025 at 06:53:41PM -0800, Jon Forrest via TUHS wrote: > In the early 1980s David Kashtan at SRI wrote Eunics, which was a way to > run Unix (I don't recall which version) on top of VMS. I ran on that. It was a very lonely place. When you ran it, you were the only person visible, you didn't see any of the VMS people logged in. I don't know if you saw other Eunice people, I never did. Contrast that with 4.2BSD on the Vax, You saw everyone, there were tools to see them and talk to them, finger, talk, etc. BSD was just so much more friendly and inclusive. I know that TUHS is all about v7 et al, and I get how amazing it was that Ken & crew made that work multi user on PDP-11s in a tiny amount of memory. Mad respect for that, it's a lost art (he says as his web browser has a 32GB foot print and don't get me started about slack). But as much as I respect the early Unix work, it's BSD that made me feel like Unix was home. Rob can complain about cat coming back wagging tails, I think that misses the point. And I almost never disagree with Rob, I'd like to be Rob, I'm just not. But the point it misses is BSD made Unix really friendly. Welcoming in a way the V6, V7, even Sys III and Sys V were not. Lots of fun stuff that felt like it was connecting you to the other people on the machine. I ported Lachman's networking stack to SCO and it was a huge step backwards in time, I don't remember what release they had but it was early. It was not a friendly environment in my opinion. Sorry for the rant, the Eunice thing just made me remember that it was nice but lonely. --lm From tuhs at tuhs.org Mon Dec 29 15:42:55 2025 From: tuhs at tuhs.org (Warner Losh via TUHS) Date: Sun, 28 Dec 2025 22:42:55 -0700 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: <71af2301-7de7-4b4d-abdc-b5b119dce3ff@gmail.com> References: <5fb3d6a9-890c-4d24-88f9-7a98654370fd@osta.com> <71af2301-7de7-4b4d-abdc-b5b119dce3ff@gmail.com> Message-ID: 9 On Sun, Dec 28, 2025, 9:11 PM Wesley Parish via TUHS wrote: > Regarding Eunice, was the source actually distributed in any way? And > even if not, are there any source tapes still surviving? > Eunice was marketed by Wollongong, but it was Kashtan's work. It never had a source distribution. Internally, source was tightly controlled. TWG's VMS TCP/IP product had part of Eunice inside. Maybe some of the engineers from the Falls Church office might have a copy, but I doubt it. TWG was sold to attachmate, and the IP scattered I was told. Maybe Frank is around and can say... Warner Wesley Parish > > On 29/12/2025 16:55, Heinz Lycklama via TUHS wrote: > > Regarding the reference to "Eunics" below - it was actually > > called "Eunice" - a product introduced by the Wollongong Group > > In 1981. We (ISC) also introduced a product that provided the UNIX > > environment running on top of the DEC VMS system in 1979. > > I took the lead on that project/product for ISC when I started with > > ISC in early 1978. It turned out to be a successful product by > > ISC for many years in the 1980's. > > > > Heinz > > > > On 12/28/2025 6:53 PM, Jon Forrest via TUHS wrote: > >> Some very minor notes: > >> > >> On 12/28/25 5:12 PM, Clem Cole via TUHS wrote: > >> > >>> And getting back to BSD the key differences between 4.0 and 4.1 are > >>> pretty > >>> small and the time between them was short (Oct 1980 and June 1981). > >>> The > >>> primary differences are the #ifdef FASTVAX stuff that Joy did over the > >>> winter after the dust up that the Stanford folks started in the > >>> fall 1980 > >> > >> All the talk about Stanford should make it clear that it was actually > >> the Stanford Research Institute (SRI), not Stanford University. > >> SRI was also one of the first 4 nodes on the Arpanet. > >> > >>> - Joy had to demonstrate that Unix was just as fast as VMS (which > >>> had been > >>> written in assembler). He instrumented a bunch of the kernel and if a > >>> couple places dropped into assembly and got Unix to perform within a > >>> very > >>> small epsilon on everything that DARPA cared about. So the issue > >>> became > >>> that ATT nor DEC was supporting Unix. CSRG does not yet exist. > >> > >> In the early 1980s David Kashtan at SRI wrote Eunics, which was a way to > >> run Unix (I don't recall which version) on top of VMS. It was an > >> emulation layer, not a virtual machine. (I'm not aware of any VMs that > >> ran on top of VMS). I used an VMS executable version of 'vi' that worked > >> just fine for most of my VMS career. > >> > >> Kashtan and Joy went back and forth for a while. I think both > >> communities (VMS and Unix) benefited from their work. > >> > >>> [For a > >>> historical prospective, Stanford had counter proposed using DEC/VMS and > >>> Australian Wollongong’s Unix for VMS and the Tek/CMU IP/TCP stack > >>> for VMS - > >>> two commercial products and the later FOSS. > >> > >> I don't recall how Kashtan's Eunice, done at SRI, became part of the > >> Wollongong Group. I do know that Kashtan and Ken Adelman wrote a > >> IP/TCP for VMS called Multinet that was quite popular for a while. > >> I'm not aware of it ever becoming FOSS. > >> > >> (For a fascinating non-Unix anecdote, check out > >> https://www.californiacoastline.org/streisand/lawsuit.html about how > >> Adelman presumably used some of the money he made when he and Kashtan > >> sold Multinet to Cisco to fund his successful lawsuit defense against > >> Barbra Streisand.) > >> > >> Sorry to mention VMS so much. However, early Unix history was often > >> buffeted by what was happening in VMS and DEC. > >> > >> Jon > >> > > > From tuhs at tuhs.org Tue Dec 30 00:43:10 2025 From: tuhs at tuhs.org (Jim Capp via TUHS) Date: Mon, 29 Dec 2025 09:43:10 -0500 (EST) Subject: [TUHS] Photo of old Unix manuals In-Reply-To: Message-ID: <19830742.3904.1767019390932.JavaMail.root@zimbraanteil> You mean like this: http://www2.anteil.com/My_UNIX_Bookshelf.JPG The two large volumes are from the Seventh Edition. Volumes 1-5 are from AT&T System V. The rest should be self explanatory. Jim From: "Brian \"Beej\" Hall via TUHS" To: "TUHS mailing list" Sent: Sunday, December 28, 2025 4:43:36 PM Subject: [TUHS] Photo of old Unix manuals Hey all, I'm looking for an impressive photograph of a bookshelf containing the Unix man page books. (I remember when I had an AT&T Unix PC it had a bunch of them. Not sure which weighed more--the PC or the books. Sadly I no longer possess it.) In the first week of the OS class I'm teaching, I talk about the different sections of the Unix manual and I thought it would be fun to show the students what it was like when you had to read "books". :) My searching is turning up nothing. Does anyone have one they wouldn't mind me sharing? -Beej -- Brian "Beej Jorgensen" Hall beej at beej.us https://beej.us/ From tuhs at tuhs.org Tue Dec 30 02:27:45 2025 From: tuhs at tuhs.org (Paul Winalski via TUHS) Date: Mon, 29 Dec 2025 11:27:45 -0500 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: References: Message-ID: On Sun, Dec 28, 2025 at 9:53 PM Jon Forrest via TUHS wrote: > In the early 1980s David Kashtan at SRI wrote Eunics, which was a way to > run Unix (I don't recall which version) on top of VMS. It was an > emulation layer, not a virtual machine. (I'm not aware of any VMs that > ran on top of VMS). > It is not possible to implement a virtual machine hypervisor on the original VAX architecture. The reason is the PROBER and PROBEW instructions. These allow one to test a memory address for read accessibility (PROBER) or write accessibility (PROBEW). The problem is that they are non-privileged instructions. To implement a virtual machine the hypervisor must be able to trap both of these instructions and to "lie" to the virtual machine about the accessibility of the address. -Paul W. From tuhs at tuhs.org Tue Dec 30 02:36:43 2025 From: tuhs at tuhs.org (Al Kossow via TUHS) Date: Mon, 29 Dec 2025 08:36:43 -0800 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: References: Message-ID: <3df19765-f201-14c4-712c-ef906719f373@bitsavers.org> On 12/29/25 8:27 AM, Paul Winalski via TUHS wrote: > On Sun, Dec 28, 2025 at 9:53 PM Jon Forrest via TUHS wrote: > >> In the early 1980s David Kashtan at SRI wrote Eunics, which was a way to >> run Unix (I don't recall which version) on top of VMS. from someone who knows the history not on TUHS "the reason for Eunice’s existence was that Kashtan (then at SRI) wanted to run Franz Lisp but had a VMS VAX, not a BSD Unix VAX. So he wrote a 4BSD system call emulator for VMS so it could run 4BSD user programs. Eventually he was able to run Franz Lisp. I’m not sure how much else of the 4BSD userland he had to bring over to get that to happen. " From tuhs at tuhs.org Tue Dec 30 02:55:00 2025 From: tuhs at tuhs.org (Paul Winalski via TUHS) Date: Mon, 29 Dec 2025 11:55:00 -0500 Subject: [TUHS] Pipes on Eunice In-Reply-To: <3df19765-f201-14c4-712c-ef906719f373@bitsavers.org> References: <3df19765-f201-14c4-712c-ef906719f373@bitsavers.org> Message-ID: I assume that Eunice implemented Unix pipes. If so, how? Did they have their own pipe device driver? -Paul W. From tuhs at tuhs.org Tue Dec 30 02:56:29 2025 From: tuhs at tuhs.org (Brian "Beej" Hall via TUHS) Date: Mon, 29 Dec 2025 08:56:29 -0800 Subject: [TUHS] Photo of old Unix manuals In-Reply-To: <19830742.3904.1767019390932.JavaMail.root@zimbraanteil> References: <19830742.3904.1767019390932.JavaMail.root@zimbraanteil> Message-ID: On Mon, Dec 29, 2025 at 09:43:10AM -0500, Jim Capp via TUHS wrote: > You mean like this: [...] These are all excellent. Thanks so much, everyone. My college students might have never seen a POTS phone or _The Matrix_, but I'm going to try to keep books alive. ;) -Beej -- Brian "Beej Jorgensen" Hall beej at beej.us https://beej.us/ From tuhs at tuhs.org Tue Dec 30 03:22:24 2025 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Mon, 29 Dec 2025 12:22:24 -0500 Subject: [TUHS] Release Dates, Systems History - (was Did System V Really Prevent 5BSD?) Message-ID: First, thanks to everyone for filling in many of the gaps I left out/glossed over. Luther is correct: SVR1 was sometime in mid to late 83 (pre-Judge Green), and SVR2 is a year or so later (post-Judge Green). It is essential when considering >>anything<< WRT to AT&T and UNIX before and after the 1984 decree. Until that time, the 1956 consent decree drove everything in their behavior with regard to the UNIX marketplace, since they could not officially participate. Sorry that I left the impression that Kashtan beef was from Stanford University, not SRI - in my defense, many of us at Berkeley just looked at anything across the bay as the same 😉 IIRC, the reason Kasktan built his system call emulator (which, as others have said, was named Eunice but sold and supported by Wollongong) was to run Franz Lisp and to move Fateman's "Vaxima" [moving maxima from the PDP-10 and Maclisp had been the driver for developing a VM system on the Vax at UCB]. But the primary issue was the "unsupported" status of UNIX at the time, either by AT&T or DEC. What was funny about that complaint was that, at the time, the ARPA community was switching from PDP-10s to Vaxen as the research platform that ARPA supplied to many of the folks it funded. The "pro-VMS" argument was that not only was the OS fully supported, but DEC had superior language tools. However, DEC's behavior had changed from the late 80s to the early 80s regarding where/how you could run them on non-DEC OS for the PDP-10 (or 11s for that matter). And Kashtan needed a non-DEC-supported Lisp (DEC's VAX LISP was a Common Lisp bootstrapped from CMU's) - in this case, Franz. So part of his argument for using VMS was just not reasonable. And for the record, a number of us had a few of his tools independently of the entire subsystem (in my own case, I had his vi and a couple of others on a 9-track tape - which I think I still have somewhere). In my case, DEC was funding the UCB CAD group and had given us a fully tricked-out 785 configuration (*a.k.a.* "ucbcad"). Which we originally ran 4.1 BSD UNIX (and was also one of the test machines for 4.1A/B/C and 4.2 - particularly for many DEC-specific peripherals [TU78 - https://gunkies.org/wiki/TU78_Magnetic_Tape_Transport and DEUNA - https://gunkies.org/wiki/Digital_Ethernet_UNIBUS_Network_Adapter], as it was the only site that had them). But while we ran UNIX on it, some of us also had login privileges on a 785 in MRO where Hanover's team (our sponsor) lives, as the CAD group had agreed to ensure all our tools ran on VMS. I never knew the story of how Kashtan's work ended up in Wollongong (others may know). On Sun, Dec 28, 2025 at 8:32 PM Luther Johnson via TUHS > wrote: > If you look at the dates on the files in the TUHS Unix tree, pdp11v > (System V r1 for PDP-11) files are dated 10/11/1983, so I've always > thought that System V release 1 must have been then. For example: > > https://www.tuhs.org/cgi-bin/utree.pl?file=pdp11v/usr/src/cmd > > On 12/28/2025 06:12 PM, Clem Cole via TUHS wrote: > > Interesting. I need to talk to them as that can’t be right. I’ll believe > > they might have been a discussion WRT to calling it 5BSD [which would > have > > made sense from a stand point of being the 5th Berkeley (UNIX) Software > > Distribution tape from the Berkeley Industrial Liaisons Office since the > > ILO tried to keep track of the different tapes that folks sent out > > [originally CAD tools like SPICE, SPLICE, MOTIS long before and of the > UNIX > > distributions]. > > > > ATT would not have had anything to do with it. System III was released > Nov > > 1981 and System V late 1984 - plus Jan 1, 1984 is when Judge Green breaks > > allows ATT to be in the computer business so ATT could not have had > > anything to say before then. > > > > Calling it 4.1 made sense because it really was a small incremental > > difference from 4.0. Where numbering got weird was years later with the > > 4.2 release from CSRG. > > > > The point is that ATTs Summit’s next release at that time frame (pre > Judge > > Green) was Unix 3.0 (which would be renamed System III) while Berkeley > was > > already at the 4.x level. > > > > And getting back to BSD the key differences between 4.0 and 4.1 are > pretty > > small and the time between them was short (Oct 1980 and June 1981). The > > primary differences are the #ifdef FASTVAX stuff that Joy did over the > > winter after the dust up that the Stanford folks started in the fall > 1980 > > - Joy had to demonstrate that Unix was just as fast as VMS (which had > been > > written in assembler). He instrumented a bunch of the kernel and if a > > couple places dropped into assembly and got Unix to perform within a very > > small epsilon on everything that DARPA cared about. So the issue became > > that ATT nor DEC was supporting Unix. CSRG does not yet exist. > > > > Bob Fabry goes to ARPA with a proposal (and the results from Bill’s work > > they winter) that UCB create a team to fill that void (CSRG) - DARPA > would > > create an advisory team for them and Berkeley would add the features that > > DARPA required. CSRG was not created until the Fall/Winter 81/82. > [For a > > historical prospective, Stanford had counter proposed using DEC/VMS and > > Australian Wollongong’s Unix for VMS and the Tek/CMU IP/TCP stack for > VMS - > > two commercial products and the later FOSS. As we know Fabry’s proposal > > was accepted]. > > > > > > On Sun, Dec 28, 2025 at 6:48 PM Warner Losh > > wrote: > > > >> > >> On Sun, Dec 28, 2025, 4:34 PM Clem Cole via TUHS > wrote: > >> > >>> Right idea, wrong release. 4.1BSD was the “fastvax” update from 4.0BSD > >>> and > >>> really was an increment from the previous release. 4.1 was triggered by > >>> the > >>> Stanford push to use VMS for performance reasons as the official ARPA > >>> supported OS for the VAX. It was CSRG’s naming of 4.2BSD were things > got > >>> strange. When you look at functionality the changes from 4.1 to 4.2 > were > >>> huge [Henry Spencer’s famous line: “4.2 is just like Unix, only > >>> different.”]. > >>> > >>> As for why CSRG chose to call it 4.2 over 5.0BSD is a bit lost to > time. I > >>> don’t think ATT was in any position to “demand” anything. > >> > >> I'm pretty sure that 4.1 was going to be 5BSD. I've heard this story > from > >> Kirk several times. He has a listing he labeled 4.5BSD. It was between > 4BSD > >> and 5BSD when he wrote 4.5 on the spine. Whatcwe know today as 4.1BSD > was > >> going to be 5BSD. But when it came time to try to do the paperwork, AT&T > >> Balked and insisted no 5BSD since it would be confused with the forth > >> coming System V. I don't know the nuts and bolts of the back and forth > to > >> know if this was a really firm request, or if threats were involved. > >> > >> I've heard this tale from both Kirk and Eric... > >> > >> Warner > >> > >> Being at UCB and > >>> my memory of the time was that the folks at CSRG chose to stay away > from > >>> the number 5 because they wanted to be sure their work was clearly > >>> distinguished from ATTs — and while the short lived 3.0BSD release was > for > >>> the Vax, most people by then associated the numbering 2xBSD to be > PDP-11 > >>> and 4xBSD to be Vax. Only later did CSRG start to add a suffix > >>> (Reno/Tahoe) to distinguish other processors as they branched out. But > >>> when 4.2BSD was released, CRSG had a DARPA contract to support UNIX on > the > >>> VAX [BTW, they did not have the contract for IP/TCP, BBN had that > contract > >>> and the stories of the issues this caused had been documented/discussed > >>> here in the past]. > >>> > >>> Sent from a handheld expect more typos than usual > >>> > >>> > >>> On Sun, Dec 28, 2025 at 6:08 PM segaloco via TUHS > wrote: > >>> > >>>> So the prevailing narrative I've heard is that 4.1BSD was so-named at > >>>> AT&T's demand to avoid confusion with System V. However, the timeline > >>>> leads me to question this. 4.1BSD is given as having been issued > first > >>>> June 1981. Of the Bell lineage, my 4.1 manual also gives June 1981 > as a > >>>> publication date. 5.0 wouldn't release inside AT&T for another year, > >>> and > >>>> System V manuals give 1983 (i.e. post-divestiture) publication dates > >>>> (although they went to print before, many have Bell logos on the cover > >>>> still.) > >>>> > >>>> In any case, if System V as a product didn't exist until sometime > 1983, > >>>> how did the fear of confusion with this name prevent the naming of > 5BSD > >>> in > >>>> 1981, two years prior? Was System V as a future trademark already > well > >>>> accepted as a given in 1981? > >>>> > >>>> - Matt G. > >>>> > > From tuhs at tuhs.org Tue Dec 30 03:24:20 2025 From: tuhs at tuhs.org (segaloco via TUHS) Date: Mon, 29 Dec 2025 17:24:20 +0000 Subject: [TUHS] Photo of old Unix manuals In-Reply-To: References: <19830742.3904.1767019390932.JavaMail.root@zimbraanteil> Message-ID: On Monday, December 29th, 2025 at 08:57, Brian "Beej" Hall via TUHS wrote: > On Mon, Dec 29, 2025 at 09:43:10AM -0500, Jim Capp via TUHS wrote: > > > You mean like this: [...] > > > These are all excellent. Thanks so much, everyone. > > My college students might have never seen a POTS phone or The Matrix, > but I'm going to try to keep books alive. ;) > > -Beej > > -- > Brian "Beej Jorgensen" Hall > beej at beej.us https://beej.us/ Well speaking of POTS, in my second shelf pic there the shelf under UNIX is a bunch of Bell System technical literature. They are to phones, cabelling, and hardware what the shelf above is to UNIX. I'm curious, are these photos going in a slide show, pamphlet, etc? - Matt G. From tuhs at tuhs.org Tue Dec 30 03:34:54 2025 From: tuhs at tuhs.org (Brian "Beej" Hall via TUHS) Date: Mon, 29 Dec 2025 09:34:54 -0800 Subject: [TUHS] Photo of old Unix manuals In-Reply-To: References: <19830742.3904.1767019390932.JavaMail.root@zimbraanteil> Message-ID: On Mon, Dec 29, 2025 at 05:24:20PM +0000, segaloco via TUHS wrote: > Well speaking of POTS, in my second shelf pic there the shelf under > UNIX is a bunch of Bell System technical literature. Nice! > I'm curious, are these photos going in a slide show, pamphlet, etc? Alas, nothing so prestigious as even a slide show--just in the Canvas LMS the students use online. -Beej -- Brian "Beej Jorgensen" Hall beej at beej.us https://beej.us/ From tuhs at tuhs.org Tue Dec 30 03:37:44 2025 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Mon, 29 Dec 2025 12:37:44 -0500 Subject: [TUHS] Pipes on Eunice In-Reply-To: References: <3df19765-f201-14c4-712c-ef906719f373@bitsavers.org> Message-ID: > > The docs still exist: https://vtda.org/docs/computing/TheWollongongGroup/ On Mon, Dec 29, 2025 at 11:55 AM Paul Winalski via TUHS wrote: > I assume that Eunice implemented Unix pipes. If so, how? Did they have > their own pipe device driver? > > -Paul W. > EUNICE BSD Reference Manual August 1988 [pg 32] 3.2.2 *Pipes* UNIX pipes are implemented using VMS mailboxes (a virtual 1/0 device). The default mailbox size used by EUNICE BSD is 512 bytes. Therefore, writes to a pipe will block once there are 512 bytes waiting to be read from the pipe (in UNIX the quantity is 4096 bytes). It is extremely rare for a UNIX program to use the fact that it may write 4096 bytes without blocking. In this case, the user has the option of going into the EUNICE BSD data structures at run time and changing the default mailbox size to 4096 bytes. † Since mailboxes provide a more general interprocess communication (IPC) mechanism than UNIX pipes, a non-UNIX IPC facility is also provided with EUNICE BSD. This facility allows for pipe-like objects on which the user can perfonn *read *and *write *operations identical to those on pipes. The *write* operation guarantees that the buffer written will be delivered through the IPC facility as a single atomic entity (as long as the buffer is no larger than the maximum message length specification for the mailbox -- which is under the control of the program creating the mailbox). Unlike UNIX pipes, these objects need not be inherited from a parent process. They are given unique names which programs may use with the *open *call to gain access to a given IPC object. Access pennission to these objects may be specified by the creating program in the same way that access pennissions to files are specified. The creating program may also give these objects specific names which are recognized locally (to a specific group of processes) or globally (to the entire system); this greatly simplifies the task of creating and communicating with specialized setver processes. Typically, a user might create a uniquely named IPC object, open a globally named IPC object for access to a server, and then pass the name of its entity in a message to the server for a reply channel. †A file of "C" definitions of all internal EUNICE BSD data structmes is included with EUNICE BSD so that users *who * require the ability to manipulate the internal state of EUNICE BSD or who wish to add/replace routines in EUNICE BSD ·· .. " may do so. From tuhs at tuhs.org Tue Dec 30 04:06:19 2025 From: tuhs at tuhs.org (Bakul Shah via TUHS) Date: Mon, 29 Dec 2025 10:06:19 -0800 Subject: [TUHS] Release Dates, Systems History - (was Did System V Really Prevent 5BSD?) In-Reply-To: References: Message-ID: On Dec 29, 2025, at 9:22 AM, Clem Cole via TUHS wrote: > > IIRC, the reason Kasktan built his system call emulator (which, as others > have said, was named Eunice but sold and supported by Wollongong) was to > run Franz Lisp and to move Fateman's "Vaxima" [moving maxima from the > PDP-10 and Maclisp had been the driver for developing a VM system on the > Vax at UCB]. But the primary issue was the "unsupported" status of UNIX at > the time, either by AT&T or DEC. What was funny about that complaint was > that, at the time, the ARPA community was switching from PDP-10s to Vaxen > as the research platform that ARPA supplied to many of the folks it > funded. The "pro-VMS" argument was that not only was the OS fully > supported, but DEC had superior language tools. However, DEC's behavior > had changed from the late 80s to the early 80s regarding where/how you > could run them on non-DEC OS for the PDP-10 (or 11s for that matter). And > Kashtan needed a non-DEC-supported Lisp (DEC's VAX LISP was a Common Lisp > bootstrapped from CMU's) - in this case, Franz. So part of his argument > for using VMS was just not reasonable. From The Maxima Book[1], pages 8-9: Around 1980, the idea of porting Macsyma began to be more interesting, and the Unix based vaxima distribution, which ran on a Lisp system built at the University of California at Berkeley for VAX UNIX demonstrated that it was both possible and practical to run the software on less expensive systems. (This system, Franz Lisp, was implemented primarily in Lisp with some parts written in C.) ... 1982 was a watershed year in many respects for Macsyma - it marks clearly the branching of Macsyma into two distinct products, and ultimately gave rise to the events which have made Maxima both possible and desirable. ... The new version was distributed via the National Energy Software Center, and called DOE Macsyma. It had been re-coded in a dialect of lisp written for the VAX at MIT called NIL. There was never a complete implementation. At about the same time a VAX/UNIX version "VAXIMA" was put into the same library by Berkeley. This ran on any of hundreds of machines running the Berkeley version of VAX Unix, and through a UNIX simulator on VMS, on any VAX system. [1] https://maxima.sourceforge.io/docs/maximabook/maximabook-19-Sept-2004.pdf Authors: de Souza, Fateman, Moses & Yapp [I was interested in Macsyma long before I touched Unix & still have a fanfold printed Macsyma manual from 1977 or so!] From tuhs at tuhs.org Tue Dec 30 04:19:28 2025 From: tuhs at tuhs.org (Warner Losh via TUHS) Date: Mon, 29 Dec 2025 11:19:28 -0700 Subject: [TUHS] Release Dates, Systems History - (was Did System V Really Prevent 5BSD?) In-Reply-To: References: Message-ID: On Mon, Dec 29, 2025 at 10:23 AM Clem Cole via TUHS wrote: > I never knew the story of how Kashtan's work ended up in Wollongong (others > may know). When I worked there it was complicated. Wollongong was marketing all the things related to Unix because it had a V6 license that it inherited from The Wollongong University when they licensed their V6 port to them for sale in the US. Originally, Eunice was 4BSD or 4.1BSD based (the differences being rather slight for what it did), but later 4.2BSD was ported and that was also marketed by TWG. Afterwards, TWG marketed a number of different Unix ports for obscure hardware related to P&E original. Then they branched out into network stacks (ports of 4.2BSD) to various System III and V related systems (boy there was a lot if 3B* hardware around the testing lab). They were one of the few places that could effectively market Eunice, I was told, due to their V6 license, so they got it originally. The business deal soon turned south when TWG marketed TCP/IP for VMS (which is why it had Eunice inside). TWG was allowed to sell Eunice, but Kashtan didn't like that TWG had created a new product and shipped it under the old agreement. I don't know if this was before or after he went off to found TGV. There was always a lot of tension at trade shows that we were at together. TGV was a rewrite of the glue around the BSD networking stack, with many things completely rewritten for VMS to give them a native VMS interface, rather than the clunky (to the VMS market) Unix interface. TGV also had better performance much of the time, but TWG was first to market and had market share when I was around. I know that TGV's Multinet also ported all the BSD protocols, whereas Eunice just did TCP/IP, and though I never got a look at the code, the people that did tell me that it was much cleaner than the Eunice code (which was a hack that kinda worked: Clem posted an example of why with the mailbox default settings changes... it only worked if you upped a lot of SYSGEN limits). I'm sure I have many of the details wrong, but the flavor is likely right. I heard from an old friend at TWG who doesn't have a Eunice tape or anything like that. But he told me the story of attachmate adding showers to the offices and their creation of a 'time capsule' behind one of the walls they put in for the showers an old Eunice tape (years after it was out of support). Google has the buildings now, and nobody knows what happened to it in the transition. There might be some people that have the source, but it's obscure enough that any of them are likely to surface and if they do, it's not clear if the tapes can be read. I've also heard back from Kirk who confirmed my recollection of the story. After BSD, then 2BSD, then 3BSD, then 4BSD, 5BSD was the next logical release. AT&T had already started to think ahead about marketing Unix as System V at the time, so they objected to the next BSD release being 5BSD, so 4.1BSD was born. The listing that I saw was in between 4BSD and what is now known as 4.1BSD. It was only after 4.2BSD that 2BSD became 2.8BSD and 2.9BSD as the 4.2BSD networking bits were back-ported to the pdp-11 Unix... 2.8BSD was because they'd shipped about 80 tapes in a 'rolling release' sort of source code control. Warner From tuhs at tuhs.org Tue Dec 30 05:03:15 2025 From: tuhs at tuhs.org (Paul Winalski via TUHS) Date: Mon, 29 Dec 2025 14:03:15 -0500 Subject: [TUHS] Pipes on Eunice In-Reply-To: References: <3df19765-f201-14c4-712c-ef906719f373@bitsavers.org> Message-ID: On Mon, Dec 29, 2025 at 12:38 PM Clem Cole wrote: > > 3.2.2 *Pipes* > > UNIX pipes are implemented using VMS mailboxes (a virtual 1/0 device). > The default mailbox size > > used by EUNICE BSD is 512 bytes. Therefore, writes to a pipe will block > once there are 512 bytes > > waiting to be read from the pipe (in UNIX the quantity is 4096 bytes). It > is extremely rare for a > > UNIX program to use the fact that it may write 4096 bytes without > blocking. In this case, the user has > > the option of going into the EUNICE BSD data structures at run time and > changing the default > > mailbox size to 4096 bytes. > Thanks, Clem. I suspected that they might use VMS mailboxes to implement pipes. It certainly is the VMS feature closest in behavior to pipes. But mailboxes differ from pipes in some key aspects. As mentioned in the EUNICE doc excerpt quoted above, there is a size associated with VMS mailboxes and when the mailbox is full the process issuing the write operation is stalled until some data are read from the mailbox. This happens on Unix, too, with pipes. But on VMS the writing process is stalled in a particularly nasty way. It is placed into a resource wait state called RWMBX (resource wait mailbox full). When that happens the stalled process can't be deleted until it leaves RWMBX state. Your process hangs until some data are read from the mailbox. This can be nasty. The pipe(2) call returns two file descriptors, one for the read end of the pipe and the other for the write end. The equivalent system call in VMS is $CREMBX, which creates a mailbox and assigns an I/O channel (VMS equivalent of a file descriptor) to it. To emulate pipe(2) one would have to assign a second channel to the mailbox. But the catch is that these channels are bidirectional--you can read or write to either of them. This has a bad consequence: You can't do broken pipe detection using mailboxes. With Unix pipes a read will report a "broken pipe" error if the file descriptor for the write end of the pipe is closed. Conversely, a write will report a "broken pipe" error if the file descriptor for the write end of the pipe is closed. I/O channels assigned to a mailbox aren't read-only or write-only. You can't tell if you're writing but there are no readers, or reading and there are no writers. In the mid-1980s DEC did a port of the Unix Bourne shell to VMS. This was marketed under the name DEC Shell. We (VMS Languages and Software Development Tools Group) were aware of these two behavioral differences as well as other gotchas when you try to use mailboxes as pipes. We ended up cloning the mailbox device driver code and tweaking it to implement true Unix pipe behavior. It did not place your process into an uninterruptible resource wait state if the pipe filled up--you got a conventional wait-for-I/O-completion. I/O channels assigned to the pipe were tagged as read-only or write-only based on the first I/O operation done to them. This allowed for "broken pipe" detection. This pipe device driver shipped along with DEC Shell, which otherwise was non-privileged. -Paul W. From tuhs at tuhs.org Tue Dec 30 05:15:34 2025 From: tuhs at tuhs.org (Rik Farrow via TUHS) Date: Mon, 29 Dec 2025 12:15:34 -0700 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: References: Message-ID: I have to disagree with Clem about the date of the SysV release. As an example, see this PDF from a set of manuals for System V: https://bitsavers.org/pdf/morrow/tricep/Uniplus+_System_V_Administrators_Manual_1983.pdf I was running SVR3 on a Motorola 68010 by 1984, and used it to write a book about administering to System V. On Sun, Dec 28, 2025 at 6:12 PM Clem Cole via TUHS wrote: > > ATT would not have had anything to do with it. System III was released Nov > 1981 and System V late 1984 - plus Jan 1, 1984 is when Judge Green breaks > allows ATT to be in the computer business so ATT could not have had > anything to say before then. > > Rik From tuhs at tuhs.org Tue Dec 30 05:17:20 2025 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Mon, 29 Dec 2025 14:17:20 -0500 Subject: [TUHS] Release Dates, Systems History - (was Did System V Really Prevent 5BSD?) In-Reply-To: References: Message-ID: Sigh - dyslexia strikes again ... nasty typo screwingf the meaning On Mon, Dec 29, 2025 at 12:22 PM Clem Cole wrote: > .. However, DEC's behavior had changed from the late 80s 60s to the > early 80s regarding where/how you could run them ona non-DEC OS ... > From tuhs at tuhs.org Tue Dec 30 05:20:09 2025 From: tuhs at tuhs.org (Rik Farrow via TUHS) Date: Mon, 29 Dec 2025 12:20:09 -0700 Subject: [TUHS] Photo of old Unix manuals In-Reply-To: References: <19830742.3904.1767019390932.JavaMail.root@zimbraanteil> Message-ID: I don't have a photo of the binders I used when working with V7, but they had about the same number of pages as the BSD 4.4 manuals seen in http://www2.anteil.com/My_UNIX_Bookshelf.JPG. It was actually quite possible to read all the manuals (I read most of them as part of book writing research). By comparison, I visited a friend who wound up working as an IBM systems programmer, and he had over 50 feet of shelf space with IBM docs. Rik On Mon, Dec 29, 2025 at 10:36 AM Brian "Beej" Hall via TUHS wrote: > On Mon, Dec 29, 2025 at 05:24:20PM +0000, segaloco via TUHS wrote: > > Well speaking of POTS, in my second shelf pic there the shelf under > > UNIX is a bunch of Bell System technical literature. > > Nice! > > > I'm curious, are these photos going in a slide show, pamphlet, etc? > > Alas, nothing so prestigious as even a slide show--just in the Canvas > LMS the students use online. > > -Beej > > -- > Brian "Beej Jorgensen" Hall > beej at beej.us https://beej.us/ > From tuhs at tuhs.org Tue Dec 30 05:49:49 2025 From: tuhs at tuhs.org (Warren Toomey via TUHS) Date: Tue, 30 Dec 2025 05:49:49 +1000 Subject: [TUHS] Pipes on Eunice Message-ID: ----- Forwarded message from "G. Branden Robinson" [CCing Warren so he can bounce my message to TUHS, with which GMail seems to have an issue, if necessary] At 2025-12-29T14:03:15-0500, Paul Winalski via TUHS wrote: > As mentioned in the EUNICE doc excerpt quoted above, there is a size > associated with VMS mailboxes and when the mailbox is full the process > issuing the write operation is stalled until some data are read from > the mailbox. This happens on Unix, too, with pipes. But on VMS the > writing process is stalled in a particularly nasty way. It is placed > into a resource wait state called RWMBX (resource wait mailbox full). > When that happens the stalled process can't be deleted until it leaves > RWMBX state. Your process hangs until some data are read from the > mailbox. This can be nasty. Sounds to me a bit like the Linux kernel's 'D' state for processes: "uninterruptible wait". Woe betide you should that happen to a process that has a file descriptor open on a file system you want to unmount. > In the mid-1980s DEC did a port of the Unix Bourne shell to VMS. This > was marketed under the name DEC Shell. We (VMS Languages and Software > Development Tools Group) were aware of these two behavioral > differences as well as other gotchas when you try to use mailboxes as > pipes. We ended up cloning the mailbox device driver code and > tweaking it to implement true Unix pipe behavior. It did not place > your process into an uninterruptible resource wait state if the pipe > filled up--you got a conventional wait-for-I/O-completion. I/O > channels assigned to the pipe were tagged as read-only or write-only > based on the first I/O operation done to them. This allowed for > "broken pipe" detection. This pipe device driver shipped along with > DEC Shell, which otherwise was non-privileged. I wonder if Dave Cutler et al. brought VMS mailboxes to Windows NT with this design flaw, remedied it, or went another direction with IPC. Regards, Branden ----- End forwarded message ----- From tuhs at tuhs.org Tue Dec 30 07:10:27 2025 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Mon, 29 Dec 2025 16:10:27 -0500 Subject: [TUHS] Release Dates, Systems History - (was Did System V Really Prevent 5BSD?) Message-ID: Look at my second email with this subject line; I already corrected it, but here is a more extensive list of the release dates. In my original message, I was referring to SVR2, not System V. But first, the background for all of it: - January 1982: AT&T agreed to divest its local operating companies to settle the government's antitrust suit. - August 5, 1983: Judge Greene signed the final order approving the breakup plan, clearing the way for the Jan 1, 1984 date. - January 1, 1984: The divestiture took effect, creating independent regional companies for local calls, while AT&T kept long-distance, manufacturing (Western Electric), and research (Bell Labs). AT&T UNIX System III had a single primary commercial release in 1982. An earlier, internal release internal to the Bell System is believed to have occurred around 1980. - Initial Internal Release: Some sources indicate an initial release in 1980. This was likely a pre-commercial version known internally by names such as UNIX/TS 3.0.1 or UNIX Release 3.0. - Official Commercial Release: AT&T formally announced System III in late 1981, and it was first made publicly (commercially) available outside of the Bell System in 1982. AT&T's UNIX System V had major releases starting with the initial System V in 1983, - followed by System V Release 2 (SVR2) in 1984 - System V Release 3 (SVR3) around 1986-1987 - System V Release 4 (SVR4) in 1989, - Later releases under Novell/USL included System V Release 4.2 (UnixWare 1.0 ) in 1992 - Release 4.2MP (UnixWare 2.0 ) in 1994/1995. On Mon, Dec 29, 2025 at 2:15 PM Rik Farrow wrote: > I have to disagree with Clem about the date of the SysV release. As an > example, see this PDF from a set of manuals for System V: > > > https://bitsavers.org/pdf/morrow/tricep/Uniplus+_System_V_Administrators_Manual_1983.pdf > > I was running SVR3 on a Motorola 68010 by 1984, and used it to write a > book about administering to System V. > > > On Sun, Dec 28, 2025 at 6:12 PM Clem Cole via TUHS wrote: > >> >> ATT would not have had anything to do with it. System III was released Nov >> 1981 and System V late 1984 - plus Jan 1, 1984 is when Judge Green breaks >> allows ATT to be in the computer business so ATT could not have had >> anything to say before then. >> >> Rik > From tuhs at tuhs.org Tue Dec 30 11:33:29 2025 From: tuhs at tuhs.org (Brian Stuart via TUHS) Date: Tue, 30 Dec 2025 01:33:29 +0000 Subject: [TUHS] Photo of old Unix manuals In-Reply-To: References: Message-ID: I love this idea. I'm going to steal ^W pay homage to this idea and share some of these pictures in my OS class that starts next week. BLS On Sun, Dec 28, 2025 at 9:44 PM Brian "Beej" Hall via TUHS wrote: > > Hey all, > > I'm looking for an impressive photograph of a bookshelf containing the > Unix man page books. (I remember when I had an AT&T Unix PC it had a > bunch of them. Not sure which weighed more--the PC or the books. Sadly I > no longer possess it.) > > In the first week of the OS class I'm teaching, I talk about the > different sections of the Unix manual and I thought it would be fun to > show the students what it was like when you had to read "books". :) > > My searching is turning up nothing. Does anyone have one they wouldn't > mind me sharing? > > -Beej > > -- > Brian "Beej Jorgensen" Hall > beej at beej.us https://beej.us/ From tuhs at tuhs.org Tue Dec 30 11:36:33 2025 From: tuhs at tuhs.org (William H. Mitchell via TUHS) Date: Mon, 29 Dec 2025 18:36:33 -0700 Subject: [TUHS] Photo of old Unix manuals In-Reply-To: References: <19830742.3904.1767019390932.JavaMail.root@zimbraanteil> Message-ID: <8FB23986-F715-4BA7-97A4-8A110E30218A@msweng.com> I remember many times forty years ago extolling the virtues of the latest 4.XBSD to Professor Chris Fraser, my then would-be supervisor at U of Arizona CS. Dr. Fraser would grab his Sixth Edition manual, a single binder maybe two inches thick, shake it at me, and say, "This is all you need!" He was pretty much right, of course. FWIW, I learned a lot about UNIX in grad school by watching TV: I’d flip through printed 4.1BSD documentation during commercials. At the time I was tempted to merge sections 2 and 3 of the loose-leaf manual. "What's the difference?", I thought! I later "got it", but that still makes for a good exam question in a UNIX-related OS class. William Mitchell Mitchell Software Engineering Software Development Consultant, Educator, and Contractor Occasional Adjunct Lecturer at U of Arizona CS 520-870-6488 (m) x77686d (Discord) linkedin.com/in/x77686d twitter.com/x77686d (X, schmex!) wyzant.com/Tutors/whm From tuhs at tuhs.org Tue Dec 30 11:54:13 2025 From: tuhs at tuhs.org (segaloco via TUHS) Date: Tue, 30 Dec 2025 01:54:13 +0000 Subject: [TUHS] Photo of old Unix manuals In-Reply-To: <8FB23986-F715-4BA7-97A4-8A110E30218A@msweng.com> References: <19830742.3904.1767019390932.JavaMail.root@zimbraanteil> <8FB23986-F715-4BA7-97A4-8A110E30218A@msweng.com> Message-ID: On Monday, December 29th, 2025 at 17:36, William H. Mitchell via TUHS wrote: > FWIW, I learned a lot about UNIX in grad school by watching TV: I’d flip through printed 4.1BSD documentation during commercials. > > William Mitchell I can see my favorite breakfast diner from my back door. One of my favorite pasttimes is to wake up in the morning, grab something from the shelf, and wander over to sip some coffee and read for a bit. This is one of the key reasons I keep a well-curated library rather than just a directory of PDFs. I've also got several Bell System/Western Electric coffee mugs which I select from for the morning. There's something satisfying and almost ritualistic about the whole affair. You don't get that with pixels on a screen! - Matt G. From tuhs at tuhs.org Tue Dec 30 11:57:54 2025 From: tuhs at tuhs.org (Luther Johnson via TUHS) Date: Mon, 29 Dec 2025 18:57:54 -0700 Subject: [TUHS] Photo of old Unix manuals In-Reply-To: References: <19830742.3904.1767019390932.JavaMail.root@zimbraanteil> <8FB23986-F715-4BA7-97A4-8A110E30218A@msweng.com> Message-ID: Well-worn paper with good typography, in natural sunlight, is so much easier on my eyes, something just feels better about the whole reading experience :) On 12/29/2025 06:54 PM, segaloco via TUHS wrote: > On Monday, December 29th, 2025 at 17:36, William H. Mitchell via TUHS wrote: > >> FWIW, I learned a lot about UNIX in grad school by watching TV: I’d flip through printed 4.1BSD documentation during commercials. >> >> William Mitchell > I can see my favorite breakfast diner from my back door. One of my favorite pasttimes is to wake up in the morning, grab something from the shelf, and wander over to sip some coffee and read for a bit. This is one of the key reasons I keep a well-curated library rather than just a directory of PDFs. I've also got several Bell System/Western Electric coffee mugs which I select from for the morning. There's something satisfying and almost ritualistic about the whole affair. You don't get that with pixels on a screen! > > - Matt G. > From tuhs at tuhs.org Wed Dec 31 02:10:51 2025 From: tuhs at tuhs.org (Paul Winalski via TUHS) Date: Tue, 30 Dec 2025 11:10:51 -0500 Subject: [TUHS] Pipes on Eunice In-Reply-To: <20251229194327.7i6ix2vp6mp5dcnm@illithid> References: <3df19765-f201-14c4-712c-ef906719f373@bitsavers.org> <20251229194327.7i6ix2vp6mp5dcnm@illithid> Message-ID: On Mon, Dec 29, 2025 at 2:43 PM G. Branden Robinson < g.branden.robinson at gmail.com> wrote: [regarding the inability to detect the "broken pipe" condition with VMS mailboxes] > > I wonder if Dave Cutler et al. brought VMS mailboxes to Windows NT with > this design flaw, remedied it, or went another direction with IPC. > > The Windows NT API has a CreatePipe() function. It creates the pseudo-device and returns a read-only and a write-only file handle to it. So Windows NT implements true Unix pipes. AFAIK NT does not implement VMS-style mailboxes. Just as well--IMO the inability to detect a broken connection condition is a serious design flaw in mailboxes. User-level programmers have to roll their own disconnect detection mechanism; they get no help from the VMS OS. -Paul W. From tuhs at tuhs.org Wed Dec 31 03:37:40 2025 From: tuhs at tuhs.org (Heinz Lycklama via TUHS) Date: Tue, 30 Dec 2025 09:37:40 -0800 Subject: [TUHS] Did System V Really Prevent 5BSD? In-Reply-To: References: <5fb3d6a9-890c-4d24-88f9-7a98654370fd@osta.com> <71af2301-7de7-4b4d-abdc-b5b119dce3ff@gmail.com> Message-ID: ISC also did not make the source for our UNIX on VMS product available to customers. The possibility of finding a copy of the ISC product is slim to none. Remember that the ISC product part of the company was bought out by Sun Microsystems  at the end of 1991. I left the company at the same time  and went into full time consulting for the next 20 years. Sun was bought out by Oracle later on. So finding a  copy of what we had in 1991 will be very hard to do now. Heinz On 12/28/2025 9:42 PM, Warner Losh via TUHS wrote: > 9 > > On Sun, Dec 28, 2025, 9:11 PM Wesley Parish via TUHS wrote: > >> Regarding Eunice, was the source actually distributed in any way? And >> even if not, are there any source tapes still surviving? >> > Eunice was marketed by Wollongong, but it was Kashtan's work. It never had > a source distribution. Internally, source was tightly controlled. TWG's VMS > TCP/IP product had part of Eunice inside. Maybe some of the engineers from > the Falls Church office might have a copy, but I doubt it. TWG was sold to > attachmate, and the IP scattered I was told. Maybe Frank is around and can > say... > > Warner > > > Wesley Parish >> On 29/12/2025 16:55, Heinz Lycklama via TUHS wrote: >>> Regarding the reference to "Eunics" below - it was actually >>> called "Eunice" - a product introduced by the Wollongong Group >>> In 1981. We (ISC) also introduced a product that provided the UNIX >>> environment running on top of the DEC VMS system in 1979. >>> I took the lead on that project/product for ISC when I started with >>> ISC in early 1978. It turned out to be a successful product by >>> ISC for many years in the 1980's. >>> >>> Heinz >>> >>> On 12/28/2025 6:53 PM, Jon Forrest via TUHS wrote: >>>> Some very minor notes: >>>> >>>> On 12/28/25 5:12 PM, Clem Cole via TUHS wrote: >>>> >>>>> And getting back to BSD the key differences between 4.0 and 4.1 are >>>>> pretty >>>>> small and the time between them was short (Oct 1980 and June 1981). >>>>> The >>>>> primary differences are the #ifdef FASTVAX stuff that Joy did over the >>>>> winter after the dust up that the Stanford folks started in the >>>>> fall 1980 >>>> All the talk about Stanford should make it clear that it was actually >>>> the Stanford Research Institute (SRI), not Stanford University. >>>> SRI was also one of the first 4 nodes on the Arpanet. >>>> >>>>> - Joy had to demonstrate that Unix was just as fast as VMS (which >>>>> had been >>>>> written in assembler). He instrumented a bunch of the kernel and if a >>>>> couple places dropped into assembly and got Unix to perform within a >>>>> very >>>>> small epsilon on everything that DARPA cared about. So the issue >>>>> became >>>>> that ATT nor DEC was supporting Unix. CSRG does not yet exist. >>>> In the early 1980s David Kashtan at SRI wrote Eunics, which was a way to >>>> run Unix (I don't recall which version) on top of VMS. It was an >>>> emulation layer, not a virtual machine. (I'm not aware of any VMs that >>>> ran on top of VMS). I used an VMS executable version of 'vi' that worked >>>> just fine for most of my VMS career. >>>> >>>> Kashtan and Joy went back and forth for a while. I think both >>>> communities (VMS and Unix) benefited from their work. >>>> >>>>> [For a >>>>> historical prospective, Stanford had counter proposed using DEC/VMS and >>>>> Australian Wollongong’s Unix for VMS and the Tek/CMU IP/TCP stack >>>>> for VMS - >>>>> two commercial products and the later FOSS. >>>> I don't recall how Kashtan's Eunice, done at SRI, became part of the >>>> Wollongong Group. I do know that Kashtan and Ken Adelman wrote a >>>> IP/TCP for VMS called Multinet that was quite popular for a while. >>>> I'm not aware of it ever becoming FOSS. >>>> >>>> (For a fascinating non-Unix anecdote, check out >>>> https://www.californiacoastline.org/streisand/lawsuit.html about how >>>> Adelman presumably used some of the money he made when he and Kashtan >>>> sold Multinet to Cisco to fund his successful lawsuit defense against >>>> Barbra Streisand.) >>>> >>>> Sorry to mention VMS so much. However, early Unix history was often >>>> buffeted by what was happening in VMS and DEC. >>>> >>>> Jon >>>> From tuhs at tuhs.org Wed Dec 31 05:25:47 2025 From: tuhs at tuhs.org (Thalia Archibald via TUHS) Date: Tue, 30 Dec 2025 19:25:47 +0000 Subject: [TUHS] Photo of old Unix manuals In-Reply-To: <-RY8UG8lpwIabK6tLNho_zNbsgXacsn9nplASrw_6-PX2qI3SVQ1IQpFWOzxpQ2vmX3CTv2NsYkQyrcKNEH_UBunY1dPJ0oAWDt8SghQsh4=@protonmail.com> References: <-RY8UG8lpwIabK6tLNho_zNbsgXacsn9nplASrw_6-PX2qI3SVQ1IQpFWOzxpQ2vmX3CTv2NsYkQyrcKNEH_UBunY1dPJ0oAWDt8SghQsh4=@protonmail.com> Message-ID: segaloco wrote: >> https://imgur.com/a/0XawAnx >> >> First image is V6 (+BTL Lions Commentary), V7, 3.0, 4.1, 5.0 (BTL), and early standards (/usr/group, SVID 1, X/Open, POSIX draft) and SVID 2-based User's Manual. >> >> Second image is 5.0 (WECo), SVR1 (+WWB), 4.2BSD, 4.3BSD, and SVR4. > > Covers here: https://imgur.com/a/z59pJjO Jim Capp wrote: > http://www2.anteil.com/My_UNIX_Bookshelf.JPG > The two large volumes are from the Seventh Edition. Volumes 1-5 are from AT&T System V. Wow you guys have me so jealous. My UNIX shelf is very thin. I’m too young to have built up a collection like this back in the day and whenever I’ve searched for a V7 or earlier manual, I’ve had no success. Thalia From tuhs at tuhs.org Wed Dec 31 05:38:10 2025 From: tuhs at tuhs.org (segaloco via TUHS) Date: Tue, 30 Dec 2025 19:38:10 +0000 Subject: [TUHS] Photo of old Unix manuals In-Reply-To: References: <-RY8UG8lpwIabK6tLNho_zNbsgXacsn9nplASrw_6-PX2qI3SVQ1IQpFWOzxpQ2vmX3CTv2NsYkQyrcKNEH_UBunY1dPJ0oAWDt8SghQsh4=@protonmail.com> Message-ID: On Tuesday, December 30th, 2025 at 11:26, Thalia Archibald via TUHS wrote: > segaloco wrote: > > > > https://imgur.com/a/0XawAnx > > > > > > First image is V6 (+BTL Lions Commentary), V7, 3.0, 4.1, 5.0 (BTL), and early standards (/usr/group, SVID 1, X/Open, POSIX draft) and SVID 2-based User's Manual. > > > > > > Second image is 5.0 (WECo), SVR1 (+WWB), 4.2BSD, 4.3BSD, and SVR4. > > > > Covers here: https://imgur.com/a/z59pJjO > > > Jim Capp wrote: > > > http://www2.anteil.com/My_UNIX_Bookshelf.JPG > > The two large volumes are from the Seventh Edition. Volumes 1-5 are from AT&T System V. > > > Wow you guys have me so jealous. My UNIX shelf is very thin. I’m too young to have built up a collection like this back in the day and whenever I’ve searched for a V7 or earlier manual, I’ve had no success. > > Thalia Mine is certainly not a "vintage" collection, rather, years of watching auctions and making connections with folks. Literature like this pops up on eBay from time to time, especially as lots of stuff from someone's estate or storage unit. Either way, don't give up hope, there's a lot of stuff floating around out there. Just know if it's not on my shelf already you may be competing with me for it :) Long term my library is a gift to some museum whenever the time comes. - Matt G.