[TUHS] First Unix-like OSes not derived from AT&T code?

[Clem Cole]

On Mon, May 2, 2022 at 12:16 PM Bakul Shah <bakul at iitbombay.org> wrote:

>  Thoth as students but QNX is not derived from it.

Interesting. Possible I suppose.  Derived is probably the operative word
here.  Of course, it is also quite possible that I could be miss-remember
the conversations, but as IIRC both Mike Malcolm and Dan Hildebrandt have
said to me about the influence of one on the other when I have spoken with
them socially.  Also, Kelly (who got the shirt and was at Waterloo during
that time), and was the person that introduced me to Mike in the late
1970s; also said something similar to me.

FWIW: In the late 1980s, I too used QNX (in C) in a production setting on a
386.  Before that, I had played with Thoth in a grad OS course, but I never
ran it significantly.

That said, my point was that Thoth was not trying to be a UNIX look/work
alike from an API standpoint.   Thoth, like V,  RIG, Accent, *et al*, were
all distinct developments that learned from the UNIX work but were not
trying to emulate it.   When QNX was birthed, the mK was not trying to be
UNIX, but they, like Mach later on (after the failure of Accent), did try
to supply an application layer UNIX (and later full POSIX) API.

The point that started this thread was when UNIX emulated.

BTW: I had Ieft out another important Pascal-based UNIX clone.  In 1983,
Michael Gien published his work in USENIX on Sol.  In the early 1990s, he
and his team rewrote that in C++ to create Chorus.

When OSF announced its long-term strategy for OSF/1 was to be based on
Mach; UI announced that the future SVR6 was to be based on Chorus.  While
the former was eventually released (and I think the sources can still be
found in the wild), I did not believe the latter was ever completed.

Clem



ᐧ
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20220502/dafe31ff/attachment.htm>