[TUHS] Some UNIX/TS Info From John Mashey

[segaloco via TUHS]

Hi folks, I recently contacted John Mashey with some questions about UNIX/TS, he sent along the following information:

> Sure, real quick.
>
> I assume you’ve seen Andrew Tannenbaum’s history, which is pretty good:
>
> https://www.rro.rs/unixarchive-usenet/comp.unix.questions/1988-July/005999.html
>
> although it is ambguous on some dates. My UNIX/TS manual is dated November 1978.
>
> By late 1976, Bell labs management had realized:
>
> a) Almost every project in Bell Labs wsa either using UNIX for development, text-processing or other support activies, OR delivering/planning operations support systems running UNIX.
>
> b) But there was a multiplicity of versions.
>
> So, to start consolidatation,
>
> 3 of us moved from PWB dept to Berk Tague’s UNIX Support Group in early 1977.
>
> Dick Haight Supervisor
>
> Larry Wehr & I, MTS.
>
> Goal was take the not-yet-released but well-along V7, make changes as needed (especially from PWB ruggedizing, accounting, performance, etc),
>
> and get a version suitable for time-sharing, BUT NOT with the CB-UNIX=> USG PG 3 IPC, semaphores, messaging, etc operations support systems stuff.
>
> I.e., we wanted a kernel that PWB dept didn’t have to change to make suitable for the large-user-community, computer-center environments that now existed.
>
> Mid-1979 was PWB/UNIX 2.0, i.e., UNIX/TS kernel+commands, with PWB tools, subsystems like RJE, LEAP, MRCS, etc.
>
> Mid-1980 was System III released outside.
>
> There was no outside System IV, internally, that was when the CB-UNIX features got folded into the mainline, released outside as System V.
>
> https://en.wikipedia.org/wiki/CB_UNIX
>
> Thus:
>
> V7
>
> 1 UNIX/TS
>
> 2 PWB/UNIX 2
>
> 3 System III
>
> 4 System IV (internal only) =<CB-UNIX/USG
>
> 5 System V
>
> I may some day take you up on that offer. I’m slowly going through old documents, trying to clean out house, may find things of interest.
>
> Thanks for kind words. Those were interesting times, and for some more related history, you might have fun with:
>
> The Long Road to 64-bits – double, double, toil & trouble
>
> https://spawn-queue.acm.org/doi/pdf/10.1145/1165754.1165766original in Qqueue, 2006
>
> https://dl.acm.org/doi/10.1145/1435417.1435431then CACM wanted it, 2009
>
> Weirdly, the 2 versions of the above rarely get cited in literature, but together have been downloaded >100K times:
>
> https://dl.acm.org/action/doSearch?AllField=long+road+to+64-bits
>
> One conclusion: ~1975 UNIX was on 16-bit PDP-11s, butC was getting ported to 32-bitters like IBM S?360, XDS Sigma5, etc.
>
> Had we been thinking 15-years ahead, we could have done better with typedefs & programming styles that would have saved much pain later.
>
> More recently:
>
> Interactions, Impacts and Coincidences of the First Golden Age of Computer Architecture
>
> https://techviser.com/wp-content/uploads/2022/06/Mashey.IEEE_.Micro_.2022.pdf
>
>> From:segaloco
>>
>> Sent:Thursday, May 7, 2026 9:42 PM
>>
>> To: John Mashey
>>
>> Subject:RE: UNIX/TS ?
>>
>> Thanks for the reply John! Frankly the thing that interests me the most is what relationship that system had with the Program Generic efforts USG was engaged in before, if the prior system was transformed into TS, if research was resampled, and/or if the earlier work in PWB was incorporated in any of what that system looked like.
>>
>> For some background, I recently helped archive and resurrect a copy of Program Generic Issue 3 from 1977 along with the documentation:
>>
>> https://www.tuhs.org/pipermail/tuhs/2026-April/033604.html
>>
>> This successful effort as well as preserving several UNIX 4.x documents from 1981 has renewed my interest in precisely what sort of rebundling UNIX/TS involved and how it differs from what was around it. It represents a bit of a historical hole which I've been able to flush out a fair deal of documentation surrounding.
>>
>> I'd also like to extend the offer that I am pretty constantly engaged with scanning and archiving lost and obscure UNIX literature, especially BTL/WECo materials. I'd gladly extend my services (and shipping) gratis if you have any notable materials you feel should be preserved.
>>
>> Finally a thank you John for the work you contributed. You and so many others delivered a system that shapes computing to this day and that countless lessons still emerge from.
>>
>> - Matt G.
>>
>> On Thursday, May 7th, 2026 at 21:30, John Mashey wrote:
>>
>>> Sure, that’s me. & I might be able to answer a few questions on UNIX/TS ,although I’d departed the UNIX Support Group in mid-1978 for Whippany, shortly before UNIX/TS released.
>>>
>>>> From:segaloco
>>>>
>>>> Sent:Thursday, May 7, 2026 7:17 PM
>>>>
>>>> To: John Mashey
>>>>
>>>> Subject:UNIX/TS ?
>>>>
>>>> Hello, my name is Matt Gilmore and I hope I am reaching the inbox of John Mashey!
>>>>
>>>> I was wondering if you might be the very same John Mashey of UNIX fame who played a key role in the Programmer's Workbench branch of the software. I'm hoping to get in touch regarding some study I am engaged in documenting USG UNIX developments from the mid 70s. Among other things I've been working to document details concerning the UNIX/TS effort.
>>>>
>>>> - Matt G.