"Ben Wing" <ben(a)666.com> writes:
> The people who consider these issues for Debian have arrived
> at an extremely strict and mechanical way of interpreting
> their criteria, which I think is very mistaken.
>
> The FSF made this decision years ago, and it is not an open
> question now. Could you please move the discussion to
> gnu-misc-discuss(a)gnu.org?
This is COMPLETELY relevant to this list, because the bottom issue is
[a] XEmacs forked and kept the identical license
[b] 3 or 4 years ago -- i.e. 9-10 years after the XEmacs fork -- you changed
the license in a way that it cross-incompatible.
[c] I made a simple request -- will you cross-license the manual for us?
[d] you said no.
This is totally relevant to GNU Emacs.
Uh, no. It is relevant to XEmacs. Since you more or less stated that
XEmacs developers are not to be bothered with issues like licences, it
is somewhat unclear what factual problem you would have with changing
the XEmacs manual licence to the GFDL. That you don't feel like it is
not really sufficient.
To the extent that your problems are related to actual practical
considerations instead of "we liked the older licence better and did
not bother asking contributors whether they'd go with changes, since
they better be fine anyhow", they might also be relevant to Emacs.
It is my personal opinion that by far the most prominent distribution
channel for the Emacs manual is as part of Emacs, and so it should
also be available under the GPL (which would probably not help your
case, incidentally). I'd be somewhat less inclined to state the same
thing for the Elisp tutorial: that is more like being a separate piece
of software, not as tightly integrated as either the Elisp manual and
the Emacs manual are with Emacs.
Richard, please reconsider. You have a reputation of antipathy
towards XEmacs. If you're at all interested in mending fences a
little, this would be a very easy step -- simply declare that we are
allowed to use the code under our license (which was your license up
through 2000 or so). Otherwise you give the impression of actively
hindering the XEmacs project.
The question to solve for the licences of Emacs are what benefits
Emacs first, and free software in general second.
Licences like the GPL and the GFDL retain their force by having
relevant software with suitable copyright owners willing to defend
them.
The FSF maintains a large and significant body of free software as the
sole copyright holder, and it is paying a high price for it: the price
is that the FSF itself is unable to reap the benefits of the GPL. It
can't reincorporate third party contributions without explicit
assignment.
The GPL implies the freedom to fork. XEmacs developers have decided
to make use of that freedom, and have by the same token decided that
they would make use of that freedom also by relaxing their
contribution criteria and maintaining diversified copyrights, partly
not even tracking them. Which you were free to do. A different
significant fork from FSF-maintained software was egcs. The egcs
project, in contrast, decided that they would not simplify matters in
a similar way: copyright assignments to the FSF were diligently
maintained. As a result, this fork was able to get merged with the
FSF's own variant of GCC again. Of course, egcs developers already
had seen the consequences of the Emacs/XEmacs fork, so it was easier
for them to decide differently.
Now with regard to the discussion on this list, it has not made
apparent by you
a) why the GFDL is a bad choice for Emacs manuals (I myself have
offered some opinions about that, but those mostly hold equally for
the previous licence).
b) why it would be technically impossible for you to change the XEmacs
manual licence to the GFDL: it does not seem like you have handed out
any written assurances that the licence is never going to change.
My personal guess would be "we don't like the licence". Neither do I
for this purpose and have explained why. But my dislike would be for
the old version of the licence, too.
It is pointless to say "well, everybody else but us should be affected
by the GFDL, so please give us an exception". If there are arguments
good enough for you, they should also be good enough for everybody
else. So please explain in detail what concrete problems you would
have with a licence change that other users and developers of free
software could reasonably also be expected to have.
As long as this is not particularly connected with Emacs itself, but
rather a general dislike of the GFDL, I have to agree with Richard
that this group does not seem the right place for it.
--
David Kastrup, Kriemhildstr. 15, 44793 Bochum