Re: Java, GPL and CDDL
"John Halton" <johnhalton@gmail.com> wrote:
> On 15/11/2007, Joerg Schilling <Joerg.Schilling@fokus.fraunhofer.de> wrote:
> > You first need to be very carefull to find out the license for this
> > software. It does not mention the GPL version number, which makes
> > it hard to find the authors will.
> >
> > Given the fact that a lot of the files have not been touched since the
> > GPLv3 has been published, the Author most likely intended to put the
> > software under GPLv2.
> >
> > Given the fact that it is illegal to agree on a license that is unknown at the
> > time a contact has been signed, the old files cannot ever be under GPLv3. As
> > GPLv2 and GPLv3 are incompatible, we need to asume that CaRMetal is under GPLv2.
>
> It is possible the licence uses the "version two or later" version of
> the GPL, which would allow the software to be distributed under GPL
> v.3, but I agree it looks more likely that the current version is
> under GPL v.2.
The Author seems to be French and in Europe, the clause "version two or later"
is void as I already tried to explain.
> If GPLv2 applies then the analysis is much the same as in my previous
> email, but the assessment that has to be made (based on clause 3 of
GPLv2 §3 tells you what to do in case you like to publish binaries.
If you cannot publish binaries, the software is still not illegal.
For other aspects, it seems to be important that the GPL was listed as
non-free license before the FSF declared that the ambiguous parts need to
be interpreted in a way that makes the GPL compliant to the OpenSource
definition from Bruce Perens http://www.opensource.org/docs/osd, in special
§9.
> GPLv2) is whether colorchooser is either:
>
> a. a module contained in CaRMetal;
This is obviously not true as it is not contained as far as I can see.
BTW: This is (if ever) a requirement of GPlv2 §2) !!!! not §3.
While GPLv2 §3) talks about "the whole source", GPLv2 §2) only
talks about "the work". As CaRMetal is not part of the work colorchooser
this does not apply.
> b. an "associated interface definition file" of CaRMetal; or
This is obviously a missunderstaning of the GPLv2:
GPLv2 §3) may force you to publish the colorchooser source together
with the CaRMetal source if you publish binaries but as CaRMetal
is not part of the work colorchooser, there is no requirement
to put colorchooser under GPL.
> c. a "script used to control compilation and installation of the
> executable" (I'm guessing "no" to that one).
colorchooser is most likely part of "the full source" of CaRMetal.
If you publish CaRMetal binaries, you need to publish colorchooser
together with CaRMetal. This does however not influence the license
of colorchooser.
> I'll leave it to the coders to answer that one. If the answer to any
> of those questions is "yes" then we next have to consider whether it
> falls under GPL v2's "major components" exception, which is narrower
> than that in GPL v3.
THIS is definitely a missunderstaning of the GPL. See above for more
information.
> > >CarMetal uses colorchooser https://colorchooser.dev.java.net/ wich is
> > >CDDL licensed.
> >
> > If colorchooser has been developed independently from CaRMetal, and only
> > CaRMetal calls colorchooser, it is indeed similar to what happens with mkisofs
> > and the license mix you describe is legal too.
>
> As set out above, the question is whether colorchooser fits within the
> GPL v2 wording covering (a) modules, (b) interface definition files or
> (c) build scripts. If CaRMetal's "calling" colorchooser brings it into
> one of those headings, then there is a licensing issue at that point.
> But I don't know the answer to that one.
You seem to confuse the fact that GPLv2 §3) requires to publish _more_
source than just "the work" with the requirement from GPLv2 §2) to
put all of "the work" under GPL. §2 and §3 are unrelated. If the intention
of the author of GPLv2 was to require GPL for "the full source", then
GPLv2 §3) would need to mention this explicitely.
Jörg
--
EMail:joerg@schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin
js@cs.tu-berlin.de (uni)
schilling@fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/
URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily
Reply to: