When I said that POSIX was an attempt to make a one-size-fits all, I meant
that, as I understand it, POSIX compliance is a standard to achieve for any
UNIX system at a minimum. And my knowledge (or lack thereof) comes from the
hp tutorial:

;;;;;;;;;;;;;;;;;;;;;W       :
::: POSIX is one of these industry standards.  Like many standards,   :
:
::: POSIX came out of the UNIX world.                                 :
:
:::                                                                   :
:
::: In the 1980s, many hardware vendors began offering systems based  :
:
::: on the UNIX operating system.  Theoretically, an application      :
:
::: program written for any one UNIX-based system should be portable  :
:
::: to any other UNIX-based system.  That is, you should be able to   :
:
::: move the source code from one hardware platform to another,       :
:
::: recompile it, and the program should work in the same way on both :
:
::: platforms.

 That, at least, is how things are supposed to work.  In actual    :      :
:::: practice, however, many of the hardware vendors included          :
:
:::: "enhancements" with their own particular implementations of UNIX. :
:
:::: Programmers who took advantage of these enhancements (often       :
:
:::: unwittingly) tied the programs to a particular version of UNIX.   :
:
:::: It became increasingly difficult for application programmers to   :
:
:::: keep track of what interfaces were "standard", (available on all  :
:
:::: UNIX platforms) and which were "proprietary" (only available on a :
:
:::: few).

::::: By 1990, it had become clear that what was required was a set of  :
:
::::: vendor-independent standards.  These are the POSIX standards.     :
:
::::: POSIX stands for Portable Operating System Interface for UNIX.    :
:
:::::                                                                   :
:
::::: In the early 1990s, POSIX.1 and POSIX.2 were published, defining  :
:
::::: a standard set of interfaces for programs written in the C        :
:
::::: programming language (POSIX.1), and for a command line interface  :
:
::::: (POSIX.2).


As for feeble and other words that may have offended, I assure you that was
not my intent.

Have a happy Thanksgiving.


> -----Original Message-----
> From: Steve Dirickson [SMTP:[log in to unmask]]
> Sent: Wednesday, November 21, 2001 4:35 PM
> To:   [log in to unmask]
> Subject:      Re: OT: Linux
>
> > > Posix compliance, if 100%, would enable 100% portability of
> > > code between
> > > any two 100% Posix compliant OSes regardless of hardware,
> > > assuming that
> > > only POSIX compliant APIs were used, and any third party
> > > stuff called was the same on both systems.
> >
> > True...it would seem that the POSIX act was a feeble attempt
> > at truly making a one-size-fits-all OS.
>
> Uh, not even close. First off, Posix isn't an OS of any kind; it's a
> (large) set of interfaces (which is what the "i" stands for) which can
> be implemented on pretty much any OS--something which should be
> obvious from the fact that we have "Posix" on MPE boxes. As for
> "feeble attempt", it is both inappropriate and uninformed to criticize
> the standards for the failure of various vendors to properly/fully
> implement those standards. Finally, a phrase like "the POSIX act was"
> indicates substantial ignorance of the ongoing efforts of entities
> like ISO/IEC JTC 1/SC 22, JASC, and The Open Group.
>
> * To join/leave the list, search archives, change list settings, *
> * etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *