Re: Bug#227159: ocaml: license conflict in Emacs Lisp support?

[mru@kth.se (Måns Rullgård)]

Don Armstrong <don@donarmstrong.com> writes:

> On Tue, 13 Jan 2004, Sven Luther wrote:
>> And, were is the problem ? The GPL is especifically against
>> distributing the result of the linking of GPLed code with
>> uncompatible code.
>
> No. The GPL restricts the creation of derivative works. Linking is
> considered by the FSF to be a case of derivation, but it is certainly
> not the only way that a derived work can be generated.

And since when does the FSF have the final word in legal matters.

>> It doesn't say anything against distributing GPLed and GPL
>> incompatible but free code in the same tarball or package, as long
>> as it is not linked together.
>
> From §2:
>
>      But when you distribute the same sections as part of a whole
>      which is a work based on the Program, the distribution of the
>      whole must be on the terms of this License,
>
> As you can see, linking is not the metric used. Only derivation is.

Yes, and I say linking isn't a case of derivation.  I can easily find
any number of people that disagree with RMS about this, so who's
right?

>> If that was not the case, then all our GPLed packages would have
>> been undistributable, since at least the GPL document itself is
>> clearly non-free.
>
> No. The license itself is _not_ a work based on a GPLed
> program. Therefore its aggregation with the work in question isn't at
> issue. Furthermore, it's likely that licenses may not actually be
> copyrightable... and regardless, licenses generally exist in a gray
> area anyway, so they're not a particularly useful example.

You guys seem to dismiss any argument you don't like as belonging to a
gray area.  *Everything* about licenses or other legalities is gray.

>> I don't know, i have the impression that the response i am getting
>> here are not based on legal theory and reading and interpreting the
>> licence, but on some pre-decision made because of that RMS quote.
>
> The question is: Is the code a derivative work of Emacs? RMS seems to
> believe that it is. As the FSF is the group in a position to prosecute
> such a case, we generally will acquiese to their viewpoint unless you

So why don't you just blindly believe it when (possibly evil)
companies make claims beneficial to them?

-- 
Måns Rullgård
mru@kth.se