Re: Clean up glibc manual references to "GNU system" (bug 6911)

[Sandra Loosemore <sandra at codesourcery dot com>]

On 02/18/2012 06:07 AM, Joseph S. Myers wrote:
> I am given to understand that many of the statements about properties of
> the GNU system were aspirations when they were written in 1991 - the
> extent to which the variants of the GNU system have achieved them since
> 1991 varies.

Diving back into ancient history.....  When I wrote the original glibc 
manual in 1991, there was no "GNU system" yet.  Well, folks were working 
on the HURD, but there was nothing usable yet.  Instead, glibc ran on 
various flavors of Unix, and perhaps other operating systems that were 
in common use at that time, like VMS.  I personally was using an HP 
workstation running the U of Utah version of BSD Unix to develop all the 
examples.

Somewhere along the line, support for running glibc on all those other 
operating systems was dropped.  Also, at the time I was using an early 
version of the POSIX standard for portability guidelines, and I believe 
more recent revisions have tightened up the requirements on some things. 
 I cannot say I have been tracking any of this for the past 20+ years. 
 :-P  OTOH, I'm sure the manual has become bit-rotten in many ways.  In 
particular, it's certain that glibc as it exists now is not either the 
thing it actually was in 1991 or the thing RMS was imagining in 1991 
that it would become.

On 02/18/2012 04:05 AM, Robert Millan wrote:
> IMHO you could avoid all this by simply referring to actual
> components.  There's no shade of doubt on what's the meaning of
> statements like "GNU C Library", "GNU C Library when running on the
> Hurd", etc.

I agree that this is the most pragmatic approach.  The glibc manual 
should document the GNU C Library, not the "GNU system", and indicate 
when glibc behavior depends on the operating system kernel.

-Sandra