[TUHS] Some UNIX/TS Info From John Mashey

[Clem Cole via TUHS]

Yes, he was an early Masscomp hacker. How did that happen?  Easy, trb was a
friend of mine from a few years previously. I had joined Masscomp as the
second Unix hacker to help Tom with the O very early on. But shortly there
after we started expanding as there was a lot to do, and I introduced trb
to our VP of Eng (Lorin Gale) and Lorin hired him. IIRC: He worked
extensively on the command system, for maybe two or three years.

As I recall he was living closer to Boston proper and had a social life
centered around Cambridge.  While Masscomp was out on 495
(Littleton/Westford), so he had a bit if a haul. Eventually he took a job
closer Cambridge. At some point he landed in the Boston office of
Interactive Systems Corp, which is when he wrote that note, but I don’t
remember if he had a stop somewhere else before that. I’ll have to ask him.


Clem


Sent from a handheld expect more typos than usual

On Sat, May 9, 2026 at 1:54 AM Andy Kosela via TUHS <tuhs at tuhs.org> wrote:

> On Sat, May 9, 2026 at 3:30 AM Larry McVoy via TUHS <tuhs at tuhs.org> wrote:
>
> > On Sat, May 09, 2026 at 11:07:30AM +1000, Jonathan Gray wrote:
> > > On Fri, May 08, 2026 at 10:46:30AM -0700, Larry McVoy via TUHS wrote:
> > > > Huh, I just learned that Andrew was part of Masscomp.  Clem, how did
> > that
> > > > come about?
> > >
> > > The Andrew Tannenbaum (trb) who worked at Masscomp and ISC is not
> > > the Andrew Tanenbaum (ast) associated with MINIX.
> >
> > But it was the Masscomp guy who did the history writeup?  I could see
> that
> > with Clem at Masscomp, Clem knows the history.
> >
> > Thanks for the clarification Jonathan, appreciate it.
> >
> > In the information you didn't ask for department, I'm half Dutch,
> > have lived there, learned the language, some, and I interviewed with
> > ast to go to grad school there (I think, it was a long time ago, when I
> > went there is not clear but I did interview with him).  I really liked
> > the Minix idea, it wasn't as good as QNX but it was mostly (I think)
> > just him.  It's a significant chunk of work.  I decided to pass because
> > the building he was in was gray, the furniture was gray, the weather
> > was gray.  That's trite but where he was working just didn't suck me in.
> > While I don't regret that choice, I think working with him could have
> > been fun.
>
>
> While MINIX 1 and 2 were mostly AST’s work and aimed at education, MINIX 3
> was a different beast. It was intended to be production-grade and developed
> as a collaborative effort. I know this because I contributed some code to
> it. It also leveraged the NetBSD userland.
>
> That included:
>
> - NetBSD libc
> - NetBSD build infrastructure
> - many NetBSD userland utilities: ls, cp, sh, etc.
> - NetBSD bootloader components
> - pkgsrc package system
>
> In practice, later MINIX 3 felt closer to a small NetBSD-like system
> running on top of the MINIX microkernel than to the older, fully
> self-contained MINIX 1/2 systems. It was really nice and I am still amazed
> why it did not become more popular.
>
> --Andy
>