On Thu, 20 Jul 2000, George Greer wrote:
> I'm the same way.
As long as the development team is chiming in as pro-GPL, I have to agree,
but with some minor pontificating. RMS has rather unfortunately bound the
GPL to a lot of idealogy and politicking that I don't like and don't feel
has a place. The license's definition of derivative works is a major
problem, and should be removed. It probably would be if the GPL wasn't
being used as a political platform. The problem comes with RMS's notion
that including a library in a program makes the result a derivative work.
Hence, including a GPL library or object files in a non-GPL program, or
vice versa, is not permitted.
This is ultimately a hypocritical move, taking away freedom and lending
GPL its well-known viral nature. There's nothing to be gained from
restricting GPL'd library code to only GPL projects (the code is already
available, and if they distribute a modified version, the changes are
still made under the GPL, and will have to be distributed in source form),
or from restricting non-GPL'd library code to non-GPL projects (if it's
really a problem, people will write a GPL replacement or not use it --
witness Lesstif). True, though, that the LGPL was created for this
purpose.
> UC Berkeley retroactively removed the "advertising clause" from the
> true BSD license (software from UC Berkeley).
Well, this isn't an injurous retroactive change (that is, it doesn't make
the license more restrictive), so many people would say its permissable.
For what it's worth, I don't believe retroactive licensing to be possible
at all. What Cal did is relicense its work without that clause, thus
immediately removing the requirement to include it in immediate
derivatives. An immediate derivative would be any in which the copyright
was signed, in exclusiveness, to the Regents of UC Berkeley. In a case
where the license bore the weight of other copyrights, the clause could
not be removed until the license from which it derives was changed. Some
people made the argument that this case was exceptional, reasoning that
Cal has the right to its own name and thus can remove the advertisement
clause retroactively and immediately from all existing BSD licenses. I
can't see that being anything but common-sense-lawyering: law is oft more
complex and rigid than we assume.
> I'm all for retroactive GPL-ing if the people from Diku agree.
We wouldn't need to retroactively GPL. As copyright holders, we can
relicense at any time (subject to the restrictions placed on us as being a
derivative work). If retroactive changes are possible at all, but only
without increasing restrictiveness or being injurous, we can't GPL. The
GPL requires you to distribute source code if you distribute the binary,
the possibility of existing binary distributions without source code means
a retroactive change could cause some people to be in violation of the GPL
automagically. This is decidedly injurous.
Like I said, though, we don't need retroactive changes to make this thing
work. We simply need to make it clear that we permit relicensing
CircleMUD under the GPL. This doesn't change the existing licenses, it
simply permits people to change licenses with our permission. This is
most certainly within the rights of a copyright holder.
Of course, this is all moot. The least of our concerns on this issue
would be how we would go about relicensing CircleMUD. As a derivative of
DikuMUD, we are bound by the terms and conditions of the DikuMUD license.
Unless we can get the Hans-Henrik, Tom, Sebastian, Michael, and Katja to
permit relicensing Diku, this is all talk and no walk.
-dak
+------------------------------------------------------------+
| Ensure that you have read the CircleMUD Mailing List FAQ: |
| http://qsilver.queensu.ca/~fletchra/Circle/list-faq.html |
+------------------------------------------------------------+
This archive was generated by hypermail 2b30 : 04/10/01 PDT