[Simh] pdp11 and unix

[Clem Cole]

On Fri, Feb 26, 2016 at 6:43 PM, Eric Smith <esmithmail at gmail.com> wrote:

> It would be interesting to know how they went from B to C .. but once
> they had a higher level language (C .. well, higher level compared to
> the assembler), things would become much easier.
>

​Dennis described it all in one of his papers.   NB was written in B.  C
morphed into being so as he said, there was never really a "C" compiler
from scratch.   At one point, they realized the language had diverged
enough to call it something else.

Also, as Doug has pointed out either here or on the TUHS list, there was an
early parser in TMG - which I believe spit out B at that point.   Yacc and
Lex do not appear until later in the cycle.

The point is that the kernel and the tools "matured" as time went on.   At
some point Dennis would collect up tools that people had and pick up the
current state of the kernel and "release" was come out. So it was a
ephemeral thing, not a big formal process we think of today with release
candidates et al.    The bad news for us trying to pick through the
history, is that it means in some cases we really do not have an
established date of references point.   Warren has done yeoman's work to
try to help establish such a timeline and probably has the most definitive
track of what was what - but in some cases it was hazy and frankly the
intermediate codes have been lost.

Clem
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.trailing-edge.com/pipermail/simh/attachments/20160226/12b0098f/attachment.html>