I tried very hard to get the front end of pcc released to open source (we didn't call it then) because after K&R was printed, everyone and their cat started writing C compilers based on the appendix. I had strong management support for this move, but the lawyers were still in their "lets study this for 10 years and then it will be clear what we should have done" mode. So we ended up with far pointers and ten years of standards committee agony. It's so obvious to me now, as then, that such specs should be executable (although not necessarily product-quality in speed or things like error messages). But it's also obvious that the desire to compete by adding glitter and icing runs strong nontheless.

Steve

----- Original Message -----
From:
"Larry McVoy" <lm@mcvoy.com>

To:
"Richard Salz" <rich.salz@gmail.com>
Cc:
"The Eunuchs Hysterical Society" <tuhs@tuhs.org>
Sent:
Wed, 11 Sep 2019 11:54:18 -0700
Subject:
Re: [TUHS] PWB vs Unix/TS

On Wed, Sep 11, 2019 at 02:18:08PM -0400, Richard Salz wrote:
> >
> > It would have been
> > much better if Sun had licensed their source base to AT&T and then
> > AT&T could have leveraged the industry standard.
>
>
> Interesting to speculate if that would have sped up the creation of OSF or
> delayed/prevented it. I think the former.

You're probably right but it wouldn't have mattered. SunOS was very popular
and had a good VM system with a working mmap. Once it became official
AT&T source everyone would have moved to it over time.

Sort of obvious in retrospect. Nobody, that I know of, considered it at
the time. I proposed open sourcing it.