Subject: Re: (c) permission on multi-authored works
From: "Stephen J. Turnbull" <stephen@xemacs.org>
Date: Tue, 19 Apr 2005 16:27:28 +0900

>>>>> "Joe" == Joe Corneli <jcorneli@math.utexas.edu> writes:

    Joe> The issue I was concerned about was the case in which not all
    Joe> copyright holders agree about relicensing.  (Perhaps this
    Joe> concern is irrelevant to the secrecy-options license that was
    Joe> under discussion.)

You seem to be confusing _joint_ work (which is a very limited aspect
of "community owned" projects like XEmacs or the Linux kernel) with
_derivative_ work.  Under copyleft, unilateral relicensing of a
derivative work is simply not allowed.  If you can identify an
individual author's contribution as say "joe.diff", then you may do
patch -R -p0 <joe.diff and you can relicense what's left without Joe's
permission.  If you can't effectively do that, you _must_ get Joe's
permission.

The _joint authorship_ issue is restricted to the case where Joe and
Gumby work _together_, either on an original work or on a derivative
(ie, a patch).  Then either can relicense (in the case of the
derivative, only the patch, of course).  But "community owned"
generally means layer upon layer of "I derive from your derivative
derived from Gumby's derivative derived from your derivative of my
original work".

As I understand it, we are not thereby joint authors, unless we make a
formal contract pooling our copyrights in the ultimate derivative.
Since it is typically not effectively possible to do the required
patch -R (although Tom Lord is working on that), in practice something
like XEmacs cannot be relicensed without wholesale replacement.


-- 
School of Systems and Information Engineering http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
               Ask not how you can "do" free software business;
              ask what your business can "do for" free software.