> Yeah...I've never quite understood why so many people - at least on
> these lists - seem to feel POSIX compliance is important but that
> compatability with traditional BSD practice isn't.  Is it just because
> BSD didn't care to push some standards body into rubber-stamping their
> way of doing things?  (As far as I can tell that's all POSIX is.)

Considering that traditional "hard-core" BSD people _DID_ feel
strongly enough about many issues to "push some standards bodies" into
"rubber-stamping" their way of doing things, one has to wonder why
they didn't care about others.

Have you participated in any of the POSIX standardization process?
I've not personally done so, but i've spent an awful lot of time
around the former CSRG people, while they were working to get 'the
right things' through those "rubber-stamping" committees.

If the traditional BSD implementation was a bug or accident, why
propagate it?  (traditional BSD mapped the zero page...  and many
thought that it should be changed, but in BSD it never was...  is
NetBSD wrong for changing that?  why, then, for fixing other bugs,
especially when doing so _gains_ something?)


cgd