[TUHS] ACM Software System Award to Andrew S. Tanenbaum for MINIX

[arnold at skeeve.com]

Dave Horsfall <dave at horsfall.org> wrote:

> On Tue, 18 Jun 2024, segaloco via TUHS wrote:
>
> [...]
>
> > The story I've gotten is AT&T policy was to release odd-numbered 
> > versions, so PWB 1.0, System III, and System V made it out into the 
> > world, PWB 2.0 and Release 4.0 stayed in the labs.  In the most 
> > technical sense, System IV never existed, what could've become it 
> > remained a Bell System-only issue.
>
> Thanks; I'd forgotten about AT&T's "odd-only" policy.
>
> -- Dave

In 1982 I did some contract C programming on Unix 4.0 on a PDP 11/70
at Southern Bell.  At the time, C programmers were not so common.

The "odd only" policy may be true, but it's not what I was told; I
was told that the policy was to release externally one version behind
what was being run internally.

With the consent decree done and Divestiture in the works, AT&T was
going to be allowed get into the computer business. So at some point,
someone decided that for System V, the current system would be released
externally.

I doubt we'll ever know the exact truth.

Interestingly, there was no printed reference manual for Unix 4.0;
I was given a 3.0 manual. The documents for Unix were for 4.0, these
were the equivalent of the Volume 2 doc in the research releases.
It seems that the major changes in 4.0 were kernel improvements.

Arnold