🚀 go-pugleaf

Submitted-by: eggert@twinsun.com (Paul Eggert)

njr@texhrc.uucp (Nick J. Rees) writes:

>I am also very interested to hear from anyone who has had
>positive experience with VMS POSIX. I have been using it here
>and can find nothing good to say about it.
>Problems range from the mundane (incredibly slow, hopeless
>on-line documentation) to the rather more serious like not
>being able to use 'ar' unless you have system privilege. I
>would probably have more complaints than this, except I have
>not been able to do very much with it.

RCS, a widely used free version control system, was made Posix-compliant
about a year ago.  In the past, VMS users ported old RCS releases to
native VMS.  You'd think that VMS Posix would save these users a lot of
work, since it should be much easier to port new RCS releases to VMS Posix
than to native VMS.  But so far as I know, nobody has yet ported RCS to
VMS Posix.  I'm not a VMS expert, so I don't know the details, but the
general sentiment seems to be that if you really want to use RCS under
VMS, you have to port it to native VMS; VMS Posix doesn't do the job.

There's a big difference between conforming to a standard and supporting a
standard.  In the past, skeptics have suggested that the widely advertised
Posix interfaces for non-Unix operating systems from DEC and HP (and
promised interfaces from IBM and Microsoft) are merely smokescreens to
fool government bean counters who require Posix.  Early experience with
VMS Posix suggests that the skeptics may be right.  If so, this is bad
news for the open systems community.  It'll be worse news if Microsoft's
promised Posix interface for NT also turns out to be useless in practice.


Volume-Number: Volume 29, Number 49