🚀 go-pugleaf

In article <398@aupair.cs.athabascau.ca> lyndon@cs.athabascau.ca (Lyndon Nerenberg) writes:

   rms@AI.MIT.EDU (Richard Stallman) writes:

   >Since our main goal is to produce a single complete operating system,
   >supporting its components on other systems is secondary at best.  We
   >can't afford to support all of them.

   This makes no sense to me. What good is a "single complete operating
   system" if it only runs on a few select hardware platforms?

Do not take this too litterally.  GNU will certainly run on many
platforms, because there is a painful need for this.  However, there
is also an urgent need to get GNU running first, so RMS selected a
goal environment for the main effort stem.  The team cannot get too
distracted by all the possible environments, too soon.  This is what
he meant, at least for me.

If you look carefully, you will see there are a *lot* of hooks already
in most published tools.  Despite RMS own's assertion, he is making a
great deal of effort so the GNU C compiler will run on many machines,
and this is certainly not the easiest piece of software to port.  So,
once again, understand that RMS is making this strong assertion simply
to protect the project against flurry of distracting requests.  The
project is still too young to accomodate all systems in an orderly
fashion, yet.

You could get an hand on GNU coding standards.  They are available on
prep.ai.mit.edu:pub/gnu/standards.text.  I think this is the best
reference to see exactly how GNU stands about portability.

Please note that this letter represent my personnal opinion, I do not
speak officially for the GNU project.
--
Franc,ois Pinard          ``Vivement GNU!''         pinard@iro.umontreal.ca
(514) 588-4656    cp 886 L'Epiphanie (Qc) J0K 1J0    ...!uunet!iros1!pinard