Subject: Re: improving project maintainership
From: "Stephen J. Turnbull" <stephen@xemacs.org>
Date: 11 Feb 2002 17:25:58 +0900

>>>>> "rms" == Richard Stallman <rms@gnu.org> writes:

    rms> sjt writes:

    I've advocated the goal of maximizing the "volume" (in some sense) of
    free software that is available, regardless of the use of non-free
    software outside that sphere.

I'm sorry, I should have said explicitly that I mean to abandon that
formulation.  It never accurately represented what I really meant, and
therefore is a needless obstacle to discussion.

    rms> We cannot join you, because if said that was the goal, we
    rms> would be contradicting our principal message.  We evaluate
    rms> the success of free software by the amount that you can do
    rms> using a system that is 100% free.

I've never gotten that impression.  I've always gotten the impression
that you evaluate success in terms of the number of people who never
use non-free software, not in terms of the capability of free
software.

If you are serious that capability of free software, rather than the
elimination of proprietary software, is the definition of success,
then we have no important theoretical disagreements left.  But I
really don't think you are.

    rms> Once in a while it can be a good idea to take a leap and
    rms> assume the gaps will be filled in, but not always.  I've seen
    rms> numerous non-free packages worm their way into places of
    rms> importance which they occupy for many years.  Motif took
    rms> around a decade to replace,

Er, since when has Motif been replaced?  Sun Motif continues to a
thorn in the paw of XEmacs support.

And your initial entry into this thread was in reply to a post about
TRAMP.  You successfully managed to make use of a non-free program
from within X?Emacs difficult.  You did not fill the gap with
anything.  In fact, the gap was filled by OpenBSD, not the GNU
Project.  I contend (obviously it can't be proved) that the interest
in ssh generated by the convenience of using scp as a drop-in
replacement for rcp in Emacsen would have generated a strong movement
for a fully free replacement, and it would have been developed sooner.

    rms> and look how long we have struggled with Java.

If you mean the long gestation of Kaffe et al, well, life is like
that.  It takes time to age cheese and whiskey, and it takes time to
create a programming language system, even if you have a full spec to
work against.  But an unclimbable pit?  I don't think so.

If that's not what you meant, where should I look?  My browser of
choice is links; I don't visit sites that require Java support, at
least not twice.  AFAIK I have no Java support on my machines outside
of Mozilla, where it is turned off.  I haven't been aware that Java is
a problem for my free system, except that every so often Mozilla
trashes its registry and I have to turn it off again.

So AFAICT supporting Java is basically about supporting programmers
who want open source tools for writing their market-oriented, often
proprietary, Java programs.  Except for Star Office there's nothing
important on GNU/Linux systems written in Java.  Am I missing out on
something?

BTW, I don't see anything obvious in gnu.org/philosophy.

    rms> So we must not adopt a blanket policy of accepting this
    rms> approach.

Right.  I am proposing that GNU (1) adopt a policy focussing on making
replacement as efficient as possible, accepting use of non-free
software in "special circumstances", and (2) explicitly dealing with
the issue of "special circumstances" by a policy that says we must
devote resources with top priority to eliminating the special
circumstances.  I wouldn't even be averse to requiring the "GNU
Steering Committee", or whatever serves that purpose, having to approve
any such projects case-by-case at first.  This would be a really big
change for GNU, even so.

    rms> We have to teach people to be on guard against falling into a
    rms> pit we can't get out of.

If your faith were but the size of a mustard seed ....  Remember, free
software advocates are all programmers.  Make it convenient for them
to work with a proprietary system, and the first thing they're going
to think of is making one for themselves.  Most won't, but sooner or
later somebody will.

Now, software patents are an unclimbable pit.  You have to wait until
they expire, and by then they're probably hedged with many more.

But use of Aladdin Ghostscript is not such a pit.  You can get out of
it by simply waiting a year, or by sacrificing a few advanced
capabilities today.

And use of Microsoft Office is not such a pit.  Though it be a deep
pit and the walls extremely steep, we can get out of it if we want
to.  And it's pretty clear that we do want to and are making progress
on climbing the walls.

-- 
Institute of Policy and Planning Sciences     http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
              Don't ask how you can "do" free software business;
              ask what your business can "do for" free software.