Requiring 1003.1 to be POSIX-compliant

[rja7m at helga1.acc.Virginia.EDU]

From: rja7m at helga1.acc.Virginia.EDU

I strongly feel that the moderator's position is correct.

[ What?  The moderator has expressed no opinion.
I'm the moderator.  I should know.  All comments
by the moderator are enclosed in [ -mod ] pairs
or appear in articles signed by the moderator.  -mod ]

I'm generally of the opinion that the whole POSIX and P1201
process has gotten out of hand with too much premature
standardisation and too many working groups and too much
breadth.  

If a real-time system doesn't have or need a file system then
it shouldn't try to be POSIX.  Much of my present work involves
real time controls and the notion that we should try to make them
POSIX-compliant is laughable.  Yes they are computers and they
are programmable by the user but POSIX compliance for real-time
controls that lack a file system is meaningless.

It seems that a lot of vendors want to be able to say that they
are "POSIX-compliant" without actually doing the work to make their
products truly open and interoperable.  The effort to water down
the meaning of POSIX compliant appears to be rooted in such
vendors' marketing desires rather than technical merit.

  Ran
  randall at virginia.edu


Volume-Number: Volume 19, Number 63