[ILUG] (un)link

[Rick Moen]

Quoting Philip Reynolds (phil at rfc-networks.ie):

> Like I mentioned previously in my post, POSIX compliance has helped
> Linux's popularity, especially in industry. Box-checking may not be
> relevant to you, but when businesses look to choose a new platform
> for X (development, mailserver, web server) it allows them quickly
> and efficiently to rule in and out certain aspects of the Operating
> Systems features.

It might be worth mentioning that, as it used to say in the front of 
Linux Journal, "Linux aims at POSIX.1 compliance".  In the early days
of the Linux kernel, Linus adopted the policy of implementing support
for POSIX calls only as they proved to be needed to write or port actual
real-world applications.  Why?  Because Linus knew that, while POSIX.1 
included and properly specified many vital system functions, it also
incorporated vast amounts of bureaucratic rubbish that might never be
needed -- and that the end result of supporting the whole lot was
Slowaris.

I'd be very surprised if later additions to the kernel team have done
otherwise.

Furthermore, the Linux approach has been quietly vindicated, of late, in
that the IEEE's Austin Group, in deciding what to drop entirely from 
future POSIX revisions, have reportedly been following advice from Linux
kernel developers.

-- 
Cheers,   The difference between common sense and paranoia is that common sense
Rick Moen     is thinking everyone is out to get you.  That's normal; they are.
rick at linuxmafia.com      Paranoia is thinking they're conspiring.  -- J. Kegler