[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Sheflug] RMS Talk
>>>>> "Perry" == Perry Ismangil <ismangil [at] acm.org> writes:
Perry> Last March I attended a LinuxWord Conference/Expo in
Perry> Singapore, where Bob Young (RedHat) was a keynote speaker,
Perry> also with Bob Chassell from FSF (Treasurer, I think).
The interesting thing is that most of the members of the Free Software
Business mailing list (http://www.crynwr.com/fsb/) are very much
pro-free-software (that is, they believe in the freedom as much as in
the economic advantages of the open source model of development and
maintenance). But RMS only shows up to whinge when someone mentions
the "viral" clause. ESR, of course, is nowhere to be found. I don't
really think there are two movements; I think that the FSM and the OSM
(as such) are really fringes of one movement.
Perry> Strangely, the guy from FSF was nowhere near as evangelical
Perry> as RMS. Oh, he talked about ethics, philosophy, principles,
Perry> but it came across as rather bland.
That's probably because even among the fringe of free software
advocates Richard is unique. He's the Abbie Hoffman (_Steal This
Book_) of the network society. He definitely does not believe in
property rights in software, and is not at all sure about intellectual
property in general.
When Linus talks about World Domination Now! he means via technical
superiority. OTOH, Richard is the Lenin at the top of a network of
revolutionary cells, and intends to accomplish his goals via political
and/or legal action.
Richard believes that "more options" can mean "less freedom", unless
_all_ options are free. That is, using a Debian system plus
(commercial) Netscape leaves you with _less personal freedom_ than
using a Debian system without Netscape gives you. Not something an
economist can easily sympathize with, I'll tell you.
Richard carefully calculates every action with respect to political
effect. When XEmacs came out with integrated "Mule" multilingual
capability, he immediately dropped all his well-founded technical and
legal objections---which he had stood by for _ten_ years---and merged
Mule into the mainline GNU Emacs _without_ getting a copyright
assignment to the FSF. Richard claims that he has legal papers for
Mule (true, but they are not an assignment and were an option in the
current form all along) and that it simply was time for this
technology (but his technical objection that Mule does not understand
Unicode was more valid in 1998 than it was in 1993, and Mule still
does not understand Unicode). I see no way to interpret this other
than the need to avoid conceding a dominating edge to the leading
competitor to the FSF's flagship product.
None of this has anything to do with the validity of his advocacy of
free software, or even with the validity of his arguments for it. But
it does point up his uniqueness.
Perry> I see nothing wrong with that. Despite what RMS says about
Perry> OSM, in the end I personally believe that both OSM and FSM
Perry> are a good thing for the software community as a whole.
I disagree. The Free Software Movement, as defined by RMS's
leadership, has about outlived its usefulness. It has accomplished
its goal, and in spades: not one, but at least two full-fledged free
software platforms (the original BSD people admit being inspired by
RMS and the GNU Project, although they disagreed on means == license).
All I ask is that it drop its emphasis on the "purity" definition of
freedom, and rather emphasize the "choice" definition, but RMS will
never do that, I think.
If you want to argue that the Free Software Movement is the aggregate
Brownian-motion-with-trend of the free software community, then I
agree whole-heartedly that the FSM is making great contributions, ones
which the OSM cannot duplicate. (But we did start this thread with
"RMS".)
>>>>> "Richard" == Richard <richard [at] sheflug.co.uk> writes:
Richard> Yes, this is a bit like myself I suppose :) There have
Richard> been times when I would have liked to have seen more
Richard> action from the FSF and other times when I've thought
Richard> that a bit more from the Open Source people would have
Richard> been helpful or useful.
I'd be interested to hear examples.
Richard> I have noticed that there are times when the Open Source
Richard> people were more acceptable to business users. Although,
Richard> in a situation where the FSF is now the steward of the
Richard> copyrights for the Linux S/390 platform it could well be
Richard> the case that things will change.
I don't think so. That was a _very_ OSM kind of thing to do. Very
few in the OSM (except ESR) object to the FSM, nor the FSF, as such;
it's RMS they can't work with, and that only at the philosophical/
political level. But IBM is not going to assign copyright in its
improvements to Apache or XFree86 to the FSF; it's going to assign
them to the project foundations. If you're going to assign a
copyright in GPLed software, what's the logical place to go to? Not
Linus, who had to be dragged kicking and screaming to apply for a
trademark on "Linux"! This
And it's not the "Linux S/390 platform" that they own the copyrights
to, either. It is the (relatively small) set of patches and new files
that allow the (huge) body of Linux code (very little of which is FSF
assigned) to be used to create a Linux S/390 platform. It's a great
symbolic gesture, good press for both IBM and the FSF, and basically
meaningless in practical terms.
I think the contribution of the IBM Classes for Unicode (ICU), now
renamed the International Classes for Unicode, is far more
significant. I haven't checked copyright ownership or license on
that, yet, though.
We'll see whether the symbolism affects the movement(s); I predict
not, except to further marginalize the few truly "anti-FSM" OSM types.
I don't think it changes the FSF's (already pretty high) credibility
(as a copyright steward) among OSM types in general.
Richard> I find RMS to be a breath of fresh air. I've spent most
Richard> of my life dealing with business people and engineers and
Richard> scientists who spend something like £500M without
Richard> blinking or even thinking. The idea that a complete
Richard> project could be assembled and made to work without too
Richard> much formality does appeal to me.
That was genius on RMS's part, although he has yet to learn the lesson
that he taught the rest of us---he still doesn't believe in an open
development model. He thinks distributed "cathedral" development is
good enough.
But remember, dropping an irritant particle into an oyster produces a
pearl, a thing of beauty and elegance. I have yet to see a Boeing 747
built that way. (On the plus side, I can't imagine the Millenium Dome
getting built that way, either! Yay!) I don't think it's an accident
that RMS goes about throwing screaming fits trying to repossess the
name of the OS that we all love, while the HURD is hurtin' baaaaaad.
>>>>> "Perry" == Perry Ismangil <ismangil [at] acm.org> writes:
Perry> [RMS said] "... after the next release of Emacs we will
Perry> consider using open development model... " which implies
Perry> that Emacs current is using a closed development model.
"Implies"?! Surely you know which project is _the_ model for ESR's
"cathedral"?
As long as XEmacs exists, RMS will be reluctant to open up Emacs
development because of the "unfair competition" it opens GNU up to.
Cf. http://www.xemacs.org/About/XEmacsVsGNUemacs.html and
http://www.jwz.org/doc/lemacs.html.[1] The former contains an
unedited statement from RMS, and the latter is a long thread with some
context missing from one of the GNU newsgroups. You can think of the
latter as forming part of the background for RMS's distrust of XEmacs
as expressed in XEmacsVsGNUemacs.html.
But I think that RMS also has philosophical reasons for preferring the
cathedral model in general. To RMS, software development is primarily
a social activity. If it provides benefits to the world at large,
that makes him happy. But (my contacts with RMS have led me to
believe that) what he values about it is the network of cooperative
relationships among developers that a focal piece of software like
Emacs or an OS kernel creates. It truly saddens him that most XEmacs
developers do not work with the GNU team (despite his intense dislike
for folks like Jamie Zawinski and Ben Wing). Every project needs
management and coordination; he was the incumbent for Emacs, and he
doesn't understand why anybody would get their shorts in a bunch about
it. He has said many times that there's room in Emacs development for
anybody who can contribute, and XEmacs has clearly proven that there
are plenty of XEmacs developers who can contribute and would benefit
from the sharing that goes on in the Emacs project.
On the other hand, the sharing displays and requires commitment. Thus
the cathedral---it's not appropriate for people to wander in,
rearrange the furniture, and wander back out. Disturbs the church
members at worship, you know.
Disclaimer: I interact with RMS at intervals, not daily, and those
interactions are rarely pleasant on either side. I respect his work
and his philosophy, although I disagree with the latter in many ways,
and sometimes violently so. I've thought a lot about his ideas,
partly from professional interest, and partly because coordinating
with RMS is often very useful in accomplishing FSM and/or OSM goals
that I subscribe to. Everything is IMHO and YMMV.
Footnotes:
[1] There is much other mucho cool stuff on Jamie's web site; I
strongly recommend browsing it.
--
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091
_________________ _________________ _________________ _________________
What are those straight lines for? "XEmacs rules."
---------------------------------------------------------------------
Sheffield Linux User's Group - http://www.sheflug.co.uk
To unsubscribe from this list send mail to
- <sheflug-request [at] vuw.ac.nz> - with the word
"unsubscribe" in the body of the message.
GNU the choice of a complete generation.