Tres Seaver wrote:
Martijn Faassen wrote:
If the GPL is one of those included licenses, the whole package falls
under the provisions of the GPL, not just the dependencies. This is what
the GPL requires.
I'd prefer to have somebody at the foundation pay for advice on this: I
have consulted to one very Zope-and-Python savvy IP lawyer (Ron
Chichester) who has subsequently made his analysis of the interaction of
GPL and Python's import public (at the Plone Symposium in New Orleans
[one lawyer's interpretation of GPL's interaction with Python]
Not disrepecting Ron Chichester's opinion, but I have the suspicion that
each lawyer we talk to will have a different one... I'm obviously not a
lawyer so what I'm saying is not legal advice, let that be understood
for all time. I do think that my interpretation is the interpretation
more commonly made, and more in line with the original intent of the
GPL. Whether that's correct legally I cannot say.
I bring this up not to argue for Ron'd analysis, but only to say that
assuming that you know what the GPL means in the context of Python might
need to wait until the issue has been adjudicated.
Agreed: this would become more clear once there's jurisprudence in the
courts, and it may very well be this jurisprudence will also turn out
differently in different countries...
In the meanwhile, it
is probably *not* going to be within the ZF's IP policy to allow
checking in code which forces users of the repostiory to deal with the
GPL at all;
Agreed again: we could save the hassle and just sidestep the issue
(instead of asking for legal council): avoid the GPL and thus we won't
have to worry about it. :)
I would consider such a checkin now, in the interregnum
period, to be particularly ill-advised.
My reading of the ZF's IP policy is that it tries to avoid the
provisions of the GPL. Whether it *strictly* forbids the checking in of
code that depends on GPL-ed code elsewhere I have a hard time saying,
but I'm more confident of my understanding of the intent of it.
Remember, we are talking only about a dependency here, not even an
inclusion. This case is much weaker than a lot of others.
I know we're talking about a dependency here. I'm not saying what you
did was wrong, but I do also think Benji brought up a good point that
should be carefully considered.
The Zope repository as managed by ZC has had a clear anti-GPL policy; I
don't think that the foundation's policy is likely to be more favorable
to code which might, in theory, trigger the provisions of the GPL.
I think that this is correct: it's more explicit in the ZF's policy than
before, actually. This is why I was talking about the intent of the
rules as opposed to the exact letter.
Zope3-dev mailing list