[Axiom-legal] Re: GCL used commercially?

[Page, Bill]

Stavros wrote:

> I do not claim to have such a superior ethical and moral
> compass as to decide for others what an appropriate license
> for *their* work is.

I have no trouble agreeing with Richard (and the FSF) about
the undesirable nature of closed source software and I also
agree about it being an ethical issue. But I disagree
completely with the proposed solution. Philosophically it
does not make sense to me to use legal means (software licenses)
to "enforce" freedom. This is inherently self-contradictory
and risks weakening the original argument in favour of freedom
in the first place. In my mind it is similar to "fighting for
peace" - a concept that as far as I am concerned has been
amply disproved by our generation. (Please pardon me if this
seems like an inappropriate allusion to some of you.)

But I think we can, with no contradiction use legal means to
*enable* freedom. Then we can use education and example to
encourage people (and commercial software developers) to choose
freedom. I think that the GNU foundation, through tools like
gcc and linux have already demonstrated (to many people's
surprise) that this strategy actually works very well.

We can (and do) also use legal means to protect the rights
of authors (developers, designers, etc.) in return for free
access to their creations. The intent of the law is to
encourage free innovation and accessibility by making the
work public.

I agree with the FSF that the use of copyright law is *not*
appropriate for software. Writing a program is not like
writing a book. Authors (usually) do not want their work
plagiarized. (Most) artists want access to their work
controlled and do not wish it to be copied and used as the
basis of the art of other people. And certainly not without
due credit.

Developing software on the other hand is rather more like
"invention" or industrial design. Inventors do want their
work to be used by others (but of course usually wish to be
paid for their efforts). And the government and the public
do want to encourage this kind of innovation which has so
clearly contributed to our collective quality of life over
the last few centuries. However I am not really convinced
of the appropriateness of software patents either.

Pure mathematicians, theoretical physicists and many other
scientists (usually) do not try to patent or copyright their
theorems and results. But they want to publish their results,
sometimes simply for their own sake, and sometimes to
establish their credentials among their peers (and thereby
help to ensure their personal job security, etc.). But they
do want "credit" and recognition for there work, both
academically and socially. There are well established social,
cultural and even economic conventions which make this
possible.

I think it is one of the great social innovations of the
Web is that it makes attribution of authorship and allocation
of "credit" so straight forward and equitable. We simply
"link" to other people's web pages. In a sense it is easier
to reference someone else's work than it is to copy it and
mis-represent it as your own. And of course this has become
part of the Web culture. I think this is one of the reasons
that the Web grew (and continues to grow) at such an
unprecedented pace.

I think the motivations of those who develop free software
is most similar to the motivations of the scientists. And
I think they need a similar kind of social and cultural
protection and encouragement for their efforts. Luckily,
it seems to me, the Web is just the kind of tool that makes
this possible.

Of course there will be people who use software for
unethical purposes just as their are those scientists who
use science in this manner. But public awareness and
established cultural norms should help to limit this
harmful activity. But I still do not wish scientific
ethics to be enforced by law because I think science
requires exactly the same kind of freedoms that the FSF
argues is required for (ethical) software development.

So the bottom line: For axiom and for GCL I am very much
in favour of the status quo. I think going to GPL would
be a mistake (though probably not a disastrous mistake).
Camm's recent proposal that GCL follow the gcc model
makes the most sense to me. I think the modifications
required to GCL to make conformance to the gcc developer
model possible (i.e. make it possible to use GCL to
develop applications which do not by default inherit the
full functionality of GCL in the resulting product)
would not be difficult. So GCL can be more like BSD in
that respect, i.e. it should be legally possible to
use GCL to develop software that is not free provide
of course that it no longer embodies GCL itself.

On the other hand, I am a little worried about Axiom's BSD
license - especially since Axiom (as it exists now) makes it
possible to "escape" to the underlying GCL lisp environment
and this environment is (apparently) an integral part of the
usual means for the extension of Axiom through the compilation
of the algebra code library. Does the design of Axiom also
need to change to be compatible with Camm's proposal for
GCL i.e. can Axiom be used successfully with the version
of GCL that does not allow the GCL functionality itself to
be cloned? Will the GCL functionality that is embedded in
Axiom be sufficiently well separated and/or separable from
the Axiom code itself that separate and somewhat incompatible
licensing terms coexist for the two parts?

Apparently the terms (BSD) under which Axiom was released
specifically include the possibility that closed source
extensions of Axiom might be developed for commercial or
other purposes. If Axiom must be viewed as an extension
of the full underlying lisp environment, then such closed
source development would presumably only be possible for
the version of Axiom based on a lisp with a compatible
licensing scheme (e.g. CCL, but maybe not GCL).

Perhaps the best approach would be to do nothing at all
and to allow the social and cultural mechanisms that I
mentioned above develop along the lines that already
seem rather well established and in the "right direction".
I think it is reasonable to expect that the availability
of an advanced and well supported open source computer
algebra systems like Axiom (and of course like MAXIMA
and YACAS and a few others) will establish new examples
that will encourage the traditionally closed source
computer algebra developers to become more "open-minded"
themselves. In fact I think there is some evidence that
this is already happening.

Regards,
Bill Page.


> -----Original Message-----
> From: Stavros Macrakis [mailto:address@hidden
> Sent: Monday, July 28, 2003 12:35 PM
> To: address@hidden; 'Bill Page'
> Cc: address@hidden; address@hidden; 
> address@hidden;
> address@hidden; address@hidden; address@hidden;
> address@hidden
> Subject: RE: [Axiom-developer] Re: [Maxima] Re: GCL used commercially?
> 
> 
> Stallman says:
> 
> > ...But when we make such a decision, it is a strategic one;
> > it is not because we think that non-free projects deserve our
> > cooperation. They never deserve our cooperation.  They deserve
> > to be replaced with free software.
> 
> This puts it quite clearly.  The GPL is not about defending our
> work (against proprietary forks etc.), it is about constraining
> others to license *their* work under the FSF's preferred terms
> whenever possible.
> 
> The LGPL defends our work very nicely without constraining others'
> work.  That is good enough for me.  I do not claim to have such a
> superior ethical and moral compass as to decide for others what an
> appropriate license for *their* work is.
> 
>         -s
> 



> -----Original Message-----
> From: Richard Stallman [mailto:address@hidden
> Sent: Sunday, July 27, 2003 2:40 PM
> To: Mike Dewar
> Cc: address@hidden; address@hidden;
> address@hidden; address@hidden; address@hidden;
> address@hidden; address@hidden;
> address@hidden
> Subject: Re: [Axiom-developer] Re: [Maxima] Re: GCL used commercially?
> 
> [Mike Dewar wrote:]
> >     That is a very broad and misleading statement.  Many 
> > so-called "free" licenses, in particular the GPL, require
> > computer users to give up their freedom as well.
> 
> The GPL only stops you from denying other users their freedom.
> See http://www.gnu.org/philosophy/freedom-or-power.html for
> more explanation.
> 
> 
> _______________________________________________
> Axiom-developer mailing list
> address@hidden
> http://mail.nongnu.org/mailman/listinfo/axiom-developer
>