Re: GNU-Linux vs Linux naming [was RE: LAM/MPI Parallel processing]

[SL Baur <steve@example.com>]

Simon Cozens <simon@example.com> writes in tlug@example.com:

> On Wed, Sep 06, 2000 at 02:30:48PM +0900, SL Baur wrote:
>> I disagree with this statement, btw.  The critical component is libc
>> as alternative C compilers exist.

> That isn't logical. You're saying that gcc isn't critical as other
> compilers exist. You're saying that libc is critical *ALTHOUGH*
> other libcs exist.

No.  I meant "isn't as critical".  I'm basing my statement on the fact
that Gcc was usable starting in 1988.  Glibc has only gotten usable in
the last couple of years.  By the advertising hype we (the Linux
development community in 1997) were originally sold on[1] and forced to
abandon libc5, it has only gotten usable _this week_.

> I believe the BSDs have their own libcs.[1]

I was not aware of that.  I was under the impression they were going
glibc.  This is very good news.  If there is a usable non-GNU libc
available, then I would retract my statement above.

> The FSF contributed roughly 30% of the average userspace

It's not even close to that.  The last count I did put the total
GNU[2] SLOC[3] somewhere under 10% and the margin of non-GNU to GNU
software has grown since then.

> if you discount software such as "GNU Perl" and "GNU TeX" which
> they rudely appropriated without the knowledge of the original
> development communities.

... and XFree86/X11 ...

I do.  There's a much stronger case for considering XEmacs part of
GNU, and a much stronger case for insisting on the name GNU XEmacs,
however, Stallman has disowned that project.

> To me, that doesn't make them the major stakeholders.

O.K.  We're in agreement.

> You can run a Linux system with zero GNU components; I've done it.

Oh really?  I guess I'm going to have to take a close look at *BSD
someday soon.

> [1] And if my belief is incorrect, then why doesn't RMS shout about
> GNU/*BSD as well? Could it be because Linux is the popular one, and
> he's more interested in popularity than principles?

He's not a principled person, witness the hypocritical nature of
accepting (a *huge* amount) non-FSF copyright assigned code into
Emacs at the drop of an XEmacs 20.0 release at a time when he was
railing at me for not being strict about FSF Copyright assignment in
XEmacs.  Of course, when I did manage to get developers to sign the
Copyright assignment, the first thing he did was try to talk them out
of developing for XEmacs.  Bah!

Footnotes: 
[1]  Internationalization and in particular, internationalization for
Asian locales.

[2]  As defined as being software available at prep.ai.mit.edu/pub/gnu
aka ftp.gnu.org/pub/gnu.

[3]  Source Lines Of Code.