> > > The only relevant thing I can find is this part of the Open Source
> > > Definition:
> > >
> > > "5. No Discrimination Against Persons or Groups
> > > The license must not discriminate against any person or group of
> > > persons.
> >
> > We are *not* talking about the "Open Source Definition" here,
> > but about the "GNU General Public License". From a juridical
> > point of view, additional restrictions on your license should
> > be compatible with the GNU GPL.
>
> That's just it - compatibility with the GPL is a non issue, because US
> Law would restrict Maxima being exported whether or not we said
> anything in the distribution itself, and regardless of the terms we
> use. Even if we relased public domain, in theory it couldn't be
> exported.
Nope, the DOE letter gives you permission to distribute the software
under the GPL *provided that* you include the paragraph with
the export conditions. But if you do that, then it is not
a GPL license anymore!
> > However, if you decide to change the license, don't forget that you
> > will no longer be able to include GPL-licensed pieces of software in
> > Maxima.
>
> That's why we want to stay GPL.
As I understand it now, you currently are not GPL.
> > No, I think that the DOE letter states that the restrictions
> > should be mentioned in the license itself:
> >
> > "... the previous paragraph should be included in the GPL and ..."
>
> But that paragraph is not a condition of the license, it simply informs
> the user as to a state of law that exists beyond our control. It's not
> a restriction we or the GPL place on the software with the license.
I disagree: the paragraph clearly restricts the right to distribute
the software under the condition that you include an additional paragraph
in the license. This is not an end-user restriction, nor an informative
message, but a restriction on the distributor.
> > Once again, I think that the Distributor has to meet two
> > restrictions:
> > the GPL restrictions and the DOE restriction that the GPL license
> > should include the addional export restriction (which does not seem
> > to be compatible with the GPL).
>
> The Distributor has to meet those restrictions from the DOE as a
> consequence of US Law, not as a consequence of anything we have any
> control over.
So you claim that other GPL'ed software (made in USA)
is bound to the same restrictions? So why should these restrictions
not be explicitly mentioned in the licenses of other software?
Indeed, I have never seen such restrictions mentioned before.
I would also suspect them to be incompatible with the GPL.
In other words, if you were really bound to do so,
then the GPL would be an illegal license in the USA.
> What is bothering you here - are you saying we can't distribute GPL
> because of U.S. Law restrictions, or because they ask us to include the
> notice those restrictions exist with the distribution? If the former
> is true, then depending on what those laws actually say there is a very
> real chance that NO software of this type written in Amercia could be
> distributed under GPL, which would impact projects far beyond Maxima,
> and maybe even beyond Mathematics.
Yes, that is what I think. But I also think that there is a difference
between your obligation to include an additional paragraph in the GPL and
the abstract obligation of every US citizen to meet the law.
> > I think that there are several things you may do:
> >
> > * Contact the GNU folks and their legal experts.
> > Drawback: if they decide that the restriction is incompatible
> > with the GPL, then they will ask you to quickly change
> > the license or get things straight.
>
> Which we don't have the power to do. We can try to straighten things
> out, but the distinct possibility of vast consequences for all GPL CASs
> is not to be taken lightly.
The fact that you fear consequences is not a good reason not
to examine them. First of all, I think that the GNU lawyers
have probably considered this issue before, so they may give
you valuable advice. Secondly, I think that there is a difference
between "GPL + export restrictions" (maxima) and "GPL" (other CASs).
Thirdly, if the GNU folks really discover a problem, then article 9
of the GPL allows them to modify the license so as to eliminate
the problem.
> Joris, I'm curious what consequences you forsee if we just keep going
> without banging our head against this - at worst, wouldn't we be told
> to change the license and have to go through exactly what you are
> proposing we leap into? Are you saying dive face first into it now to
> avoid the possiblity of running into it later?
Well, the worst scenario is that all work done until now and
in the future on the free version of Maxima has no legal basis.
In that case, court may forbid you to distribute Maxima any longer.
A solution would then be to modify the license so as
to meet all legal restrictions. But the later you do this,
the more copyright holders you have and the harder it
will become to get all necessary permissions.
Furthermore, if the current license is not really a GPL,
then it would be fair to say so to potential contributors.
As you noticed, some French contributors don't like certain
US export restrictions. Incorrect information given to contributors
may also cause you legal trouble: if some of the copyright holders
give you permission to distribute contributions under the GPL,
then you may not use them as long as you are not GPL yourself.
> In any case, I am not the final authority on whether to take steps or
> not on this. Jim has the final say and as leader of the project is the
> logical guy to contact DOE - I have just tried to explain the situation
> to the best of my ability. Please clarify this point through: are you
> saying the GPL doesn't work because of the DOE having us include the
> notice about US Law that relates to Maxima, or because of the law
> itself? They are two different issues, and the second one is FAR more
> serious - potentially a nuclear bomb on the scientific free software
> world, in the worst case scenario.
I rather think that the first point is the real issue here.
Maybe the second one too, but that would surprise me,
since the GPL has a solid reputation.
> I agree it would be nice to know the GNU folks position on this, but if
> they raise cain then we are in for a world of trouble.
Yep, but you do need to clarify things,
in the first place by honesty towards potential contributors.