Subject: Re: GPLv3 draft
From: "Stephen J. Turnbull" <turnbull@sk.tsukuba.ac.jp>
Date: Wed, 05 Apr 2006 13:35:08 +0900

>>>>> "Norbert" == Norbert Bollow <nb@bollow.ch> writes:

    Norbert> Stephen J. Turnbull <turnbull@sk.tsukuba.ac.jp> wrote:

    >> Both the GPLv2 and draft GPLv3 require that redistributions
    >> take place under identical terms (sections 2(b) and 6 in
    >> GPLv2).  GPLv3 contains an explicit DRM clause, GPLv2 does not.
    >> Therefore, you may not combine GPLv2 code with GPLv3 code
    >> without separate, explicit permission from at least one of the
    >> authors to use the license specified by the other.

    Norbert> This argument is incorrect, because the DRM clause of the
    Norbert> GPLv3 draft does not impose any GPLv2-incompatible
    Norbert> restrictions.

You may be correct that the DRM clause imposes no new restrictions,
although I disagree.

However, I agree with Bernard.  GPLv2 Section 1 clearly envisions that
"the terms" will be word-for-word identical, as does the license for
the text of the GPL itself.  The argument you are making is valid only
for sublicensable licenses.  The GPL is not sublicensable; it makes it
clear that the license is from the copyright holder (aka "original
licensor"), not the distributor.  You cannot change a word.

We could be wrong, but *for my purposes that doesn't matter*.[1]  The
point is that the meme of verbatim continuation of the license runs
deep, and furthermore, you would be on *very* flimsy ethical grounds
to change upstream's wording.  If you're wrong, and the change you
apply does change "the terms", you would be opening *your* downstream
up to all kinds of annoyances.  Many of us would be unwilling to risk
it, even with a lawyer's OK (and who can afford to get that? easier to
get upstream's permission!)

Ie, you still get a fork (in Bernard's weak sense of having to ask
permission to combine works under the different licenses).

    Norbert> I assure you that I understand the relationship between
    Norbert> strong copyleft and license incompatibility.

I hope so!  It would mean XEmacs can avoid making some hard decisions.
I'll trade that for being forced to eat my words, any day.

But I'm quite dubious, for the reasons given above.


Footnotes: 
[1]  Except that of course I will eat my words if I'm wrong. :-)

-- 
Graduate School of Systems and Information Engineering   University of Tsukuba
http://turnbull.sk.tsukuba.ac.jp/        Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
        Economics of Information Communication and Computation Systems
          Experimental Economics, Microeconomic Theory, Game Theory