[Info-vax] Anyone interested in another public access system

[sapienzaf]

On Apr 11, 2:38 pm, billg... at cs.uofs.edu (Bill Gunshannon) wrote:
> Face it, it is not so much a "Unix" standard as a "common API" standard.
> And it is not the first time this has been tried.  The last time was a
> very unix-like API as well.  Should tell you something.
>

a) If it's not a Unix standard then why does it include "for Unix" in
its name?  That seems pretty specific to me.

b) What it tells me is that there have been (and are) many variations
in how Unix and its derivatives are being implemented, and somebody
was trying to make sense of it all by creating a set of standards that
they should strive to meet.  The goal of POSIX was to allow
portability of applications from one Unix variant to another.  It has
failed in that goal, so it's kind of interesting that you are trying
to use it to demonstrate something positive about Unix-like systems.

> Yeah, because thet refuse to pay some third-party money to tell them they
> are what they are.  But then once the IEEE gets their hands on anything
> it becomes all about the money.

Everything is about the money.  Why else would it be so important to
hype the "free software" b.s. that is creating all this activity?

And, it's not just to get a label.  POSIX is about application
portability.  It's about making sure operating system features behave
the same regardless of platform.  It's about trying to stick to the
platform independence claims that shows up in so much marketing gunk.

The reality is that GNU/Linux (and other "free" Unix variants) can't
meet the standard.  It's very well documented.  It's also kind of
funny to read some of the stuff that tells you how to write "portable"
applications on GNU/Linux by trying to avoid the incompatibilities
rather than actually using the POSIX libraries.

> I have better things to do with my time.  In any event, I believe VMS
> dropped POSIX a few versions back so it is really moot.

If it's moot then why did you bring it up?

And no, they weren't dropped.

> And being as the original argument was what the standard for features
> offered by an OS should be, just which standard does VMS comply that
> the other OSes do not?

You brought up POSIX as a standard to which Unix systems comply, and
which VMS doesn't.  Or to be more specific that GNU/Linux was closer
to the standard than OpenVMS.  The burden is on you to stand behind
your argument.