Re: PHP-Nuke License Conclusion?
On Wed, May 21, 2003 at 01:59:36AM -0500, Branden Robinson wrote:
> Good. Then perhaps you'll agree that saying "This is licensed under the
> GPL with the additional restriction that" is an invalid statement,
> because such a thing is not licensed under the GPL at all.
I think that you've misparsed it. It's not saying that it's licensed under
the GPL, it's saying that it's licensed under the (terms of the GPL with
the additional restriction that... ). This, to me, is as obvious and
evident as the fact that the statement "I eat only eggs and ham" means that
I eat both eggs and ham, but nothing else.
> What happens when one constructs a cut-n-paste license is
> inclusion-by-value. Whatever inclusion-by-reference is, (I have to
> assume it's "This is licensed under the GPL with the additional
> restriction that") it certainly cannot be inclusion-by-value. In
> inclusion-by-value, the meaning of "these terms and conditions" is clear
> and unambiguous. In inclusion-by-reference, it's vague, and clause 6 of
> the referenced GNU GPL renders the work undistributable.
No it doesn't. To me this seems a ridiculous interpretation.
[snip]
> The result looks EXACTLY like:
>
> Copyright 2003 Joe Blow.
>
> TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
>
> 1) You can do foo.
> 2) You can do bar.
> 3) You can do baz.
> 4) Each time you redistribute the Program (or any work based on the
> Program), the recipient automatically receives a license from the
> original licensor to copy, distribute or modify the Program subject to
> these terms and conditions. You may not impose any further restrictions
> on the recipients' exercise of the rights granted herein. You are not
> responsible for enforcing compliance by third parties to this License.
>
> END OF TERMS AND CONDITIONS
Now what are you going to do with the overriding requirement that you can't do
baz? Let's see...
The result looks EXACTLY like:
Copyright 2003 Joe Blow.
TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
1) You can do foo.
2) You can do bar.
3) Each time you redistribute the Program (or any work based on the
Program), the recipient automatically receives a license from the
original licensor to copy, distribute or modify the Program subject to
these terms and conditions. You may not impose any further restrictions
on the recipients' exercise of the rights granted herein. You are not
responsible for enforcing compliance by third parties to this License.
END OF TERMS AND CONDITIONS
Or, in a murkier case, where the permission to do baz was joined to
something else, or implicit in something else, something like:
The result looks EXACTLY like:
Copyright 2003 Joe Blow.
TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
0) You may not do baz. This clause overrides and takes precedence over
any subsequent contrary clauses.
1) You can do foo.
2) You can do bar and baz.
3) Each time you redistribute the Program (or any work based on the
Program), the recipient automatically receives a license from the
original licensor to copy, distribute or modify the Program subject to
these terms and conditions. You may not impose any further restrictions
on the recipients' exercise of the rights granted herein. You are not
responsible for enforcing compliance by third parties to this License.
END OF TERMS AND CONDITIONS
> license immediately falls over dead. The licensor has simultaneously
> told me I am allowed to do baz, and I am not allowed to do baz.
NO HE HASN'T. See above.
> He
> needs to get his story straight, and this sort of garbage is not
> acceptable in Debian. Not main, not non-free, not anywhere. It's "all
> rights reserved". When a licensor cannot even make a coherent statement
> of license, it is imprudent in the extreme to make assumptions about his
> intentions.
<magical +3 sigh of hyperbole deflection>
Maybe Henning or I should package something really trivial with a license
such as we are debating, just to make it completely obvious what we are
talking about. At least then we'd have something concrete to talk about.
I might do that if I find some time somewhere.
> As I said before: when the GNU GPL says "this License" and "herein",
> these terms are not variables. They are constants. They always and
> forever will refer to the terms and conditions laid out within the same
> document. It is not semantically possible for a different meaning to be
> superimposed on these words without changing the content, and thus the
> identity, of the document.
You can define a dog to be a cat if it pleases you -- as indeed some
gentlemen's club does ("No dogs. For the purposes of these rules, a
dog leading a blind person shall be deemed to be a cat..." or similar).
> As a matter of policy, I think it is prudent to discourage licensors
> from riding on the back of the GPL's success and goodwill by misleading
> people as to both the license's freeness
I don't believe that such a licensor would be either "riding on the
back of the GPL's success" or "misleading people as to the license's
freeness".
However, should a maintainer be worried about that, there's nothing
to stop them from adding bloody great warnings that "ALTHOUGH THE
LICENSE OF THIS WORK MENTIONS THE GPL, IT ADDS EXTRA RESTRICTIONS,
AND THEREFORE WORKS DERIVED FROM BOTH THIS AND ANOTHER, GPL'ED WORK
MAY NOT BE DISTRIBUTED".
> Maybe people shouldn't be misled, maybe they should be paranoid and
> school themselves in free software licensing. But if so, what function
> have we on this mailing list?
We are here to help make sure that Debian distributes only things that
it may distribute, and that packages with restrictions that, whilst not
rendering them undistributable, cause "problems", do not get distributed
outside of non-free.
If you think that our users are likely to be misled by a "GPL+restrictions"
license and merge it with GPLed works before distributing the result, then
perhaps we should consider flagging *all* packages as GPL-compatible or
not -- I'm sure that people are almost as likely to make the same mistake
with any other "free but not GPL-compatible" license.
> I wouldn't say that people should *substitute* debian-legal's judgement
> for their own, but they should permit debian-legal's judgements to
> *inform* their own, to the extent that we merit respect for the
> closeness of our reasoning.
Then we might consider adding warnings as described above.
Cheers,
Nick
--
Nick Phillips -- nwp@lemon-computing.com
Bonus fortune cookie: buy 10 get 1 FREE!
Reply to: