Still more License issues

[Fare Rideau]

>: Martin Cracauer <cracauer@wavehh.hanse.de>

> Look at BSDI and FreeBSD/NetBSD. The free versions are not technically
> inferiour, as long as people invest their time to support the free
> version. If we don't have people who do, the project would be stagnant
> at that point anyway.
>
Precisely. Look at all the redundant work lost for BSD systems
since BSDI became commercial! Had BSD choosen a GNU license,
this work would have benefitted everyone. BSDI could still have
sold commercial packages and applications on top, as well as support,
so they would not have lost anything.
Instead, what do we have? Division of forces among BSD users!
(not to talk about the many divergent free branches;
but I'm not sure how this relates to Licensing;
well perhaps the requirement to provide the original source or patches
would have eased the merging of divergent enhancements?).
Proprietary software vs free software is BAD
for the same reason as protectionism vs free trade is BAD.
Allowing a few to plunder the free tradists and protect themselves,
is only a waste of energy and time for everyone.

> So what would you loose if someone make a commercial version from your
> work?
It means that from now on, there will be redundant effort done
about this software, so that everyone will lose from such thing
Again, this does not apply to new commercial packages that
might be written on top of the OS (though for similar reasons,
they would benefit from the GPL, too, but that's an independent issue).

> I'd say we even *gain* from it since more people will hear about
> the LispOS thing and then can choose between the free and the
> commercial version.
No. Now they loose half the effort spent in LispOS,
and prevent people from each user camp from helping the other one.

> If people loose interest in deveoping the free
> version further, I'd prefer to have the commercial one improving than
> no improvement at all.
With the GPL, improvement would be done in the free version,
which is better than improvement in a commercial version.
Instead of being able to market existing code
with low-quality in improvements,
commercial companies would either sell real computer services,
or build high-quality software to run on top of the free software;
in all cases, the competition will be fairer,
and the overall code quality higher.

The whole GNU philosophy is that software itself is NOT
a service or a resource, and shouldn't be owned.
Software *writing*, *supporting*, is a service,
and they will always be a great need for it,
and that's how computists should honestly earn their living.

> The BSD and Linux camps discussed this to death.
Indeed. And until we take a decision [BTW, how will we know?],
this subject might continue to be discussed.
Maybe it's time to setup a minimal infrastructure?

> But a number of BSD developers (and very powerful ones, they are)
> stated they wouldn't support a GPL system. Anyone interested should
> check dejanews and search for it, especially John Dyson's comments.
>
Ok. Now, our BSD/CMUCL patches are not going to be GPL,
so John can sleep on both ears.
Maybe if/when LispOS is successful,
John can understand that he was wrong after all, and join us.
I think that patches to GPL code can be made BSDish
(or else, we can adapt the GPL to our needs),
so that people like John can still contribute...

Again, nobody ever pretended that software
that already have some BSD-style license (CMUCL, *BSD)
should be stolen from their authors are put under the GPL.
It's all about the new software that we'd contribute.

> Suppose we have a handful of people who would invest time in the free
> version of LispOS until they could sell it and then place their own
> branch under a commercial copyright? What's wrong with that?
It will cut in two further LispOS progress.
If they want to sell services about LispOS,
they can very well do so without stealing code away from the rest of us:
RedHat, Caldera, Cygnus, NeXT, ARDI, and more, do it.
They'll live on their software, and will benefit from free software,
without there being any conflict of interest between the two.

> They would make money out of code written by the free-license people and
> those wouldn't get a penny out of their work. Maybe they never do the
> step to found the company and contribute for many years?
>
I don't judge personal choices. These will go in both directions.
What I judge is the dynamics: GPL clearly gives better software dynamics.

> So what? My interest is a) to get a system running at all and b) to
> get as much free high-quality software as possible. None of these are
> violated if someone takes a branch of my code and sells it.
>
b) is violated, because it prevents free exchange of free code
between buyers of the commercial branch and free software users/writers.

> My only way to choose one of the licenses (BSD, GPL or public domain)
> is to choose the one that gets the most developers into the boat. My
> own goals are not related to a license, but to the performance of the
> development team, which is soemwhat f(license).
>
I fully agree. The purpose of my messages have been
1) to debunk the myth about all these software licenses
2) to give my personal opinion (that's only an opinion, however strong)
 that the GPL is indeed what, on the long run, will give a better
 performance in the development team.
I agree that my arguments as for 2)
might not be absolutely convincing to everyone,
as they rely on some economical theories that not everyone would agree with.
I hope that at least my explanations as for 1)
will make our choice a voluntary, conscious, choice, whatever it be.

== Fare' -- rideau@ens.fr -- Franc,ois-Rene' Rideau -- DDa(.ng-Vu~ Ba^n ==
Join the TUNES project for a computing system based on computing freedom !
                TUNES is a Useful, Not Expedient System
URL: "http://www.eleves.ens.fr:8080/home/rideau/Tunes/"