Re: Geant4 Software License, version 1.0
Joe Smith writes:
> "MJ Ray" <email@example.com> wrote in message
> > Kevin B. McCarty <kmccarty@Princeton.EDU>
> >> I would be interested to hear your opinions on the Geant4 Software
> >> License, version 1.0 . [...]
> >>  http://geant4.web.cern.ch/geant4/license/LICENSE.html
> > I think it is clearly GPL-incompatible (as you noted) for reasons
> > similar to the old BSD licence and it might not follow the DFSG because
> > of clauses 4 (automatic donation to upstream), clause 8 (no right to
> > dispute unauthorised inclusion of your code) and maybe 5 (discrimination
> > against fields of use) below.
> How exactly does automatic upstream licence violate the DFSG?
There are several perspectives on this. The clearest is that it is a
violation of DFSG#3 (same terms as the license of the original
software). One could make a case that it is a fee for redistribution
or that it discriminates against fields of endeavor. Since the nature
of the license-to-upstream is unclear, it is more of an issue.
However, because the license-to-upstream is merely the default state,
rather than a truly automatic state, I personally do not view it as a
> I think I see your point with number 8, but the idea of that clause
> fits the spirit of Software freeness. the idea is to prevent a big
> company from
> suing the developers without any good reason in an attempt to extort
> money from
> the developers. That sort of clause does discorage that. It is also a
> purely defensive clause.
> Can you suggest a better wording?
> I don't think 5 is really a DFSG freeness problem, as attempting to patent
> a derivitive work is questionable. With all likelyhood such a patent
> would be invalidated in a court battle, so the clause only serves to prevent
> the whole thing from happening. I see no feilds of endevor which require
> software patents.
My reading of section 5 was that it prohibits the inclusion of
portions of the Geant4 software itself in a patent application. I do
not believe this impairs the patent process -- if the patent depends
so strongly on the base software, its novelty is dubious. Similarly,
I think that including open source software in patent applications is
not defensible as a DFSG#6 "field of endeavor".