Re: Simple Proposal

[Etienne M. Gagnon]

Per,

I do really want that my 2 proposals be discussed separately.

I think that there is a problem with the question: "is carrying the exception to 
derivative work allowed?".

The LGPL proposal, on the other hand, is just a nicety to library developers. If 
you don't like that part, please DO NOT dismiss my other proposal along with it!

Per Bothner wrote:

> My opinion, which I suspect the rest of the SC would share:  While I
> don't have any real objection to this license, I don't see that there it is
> a real problem it solves, and don't see how discussing or implementing
> it would be worth the effort involved.  


I think there is an important ambiguity: one has to ask RMS to know whether 
carrying the exception to derivtive work is allowed (and maybe one lawyer's 
might disagree with RMS!). Here's the problem:

In the strict interpretation of copyright law, you are only allowed to do what a 
license tells you is allowed (unless some parts of the license cannot be 
enforced for some other reason (another law)).  So, the obvious interpretation 
of the exception clause is this:

1- Fact: Source code modification IS NOT the same as linking.
2- Fact: The exception says that anything other than linking is governed by the 
GPL.
3- Fact: The GPL says that modified versions should be GPLed.

Conclusion =>

If you modify Classpath, you have to license your modified Classpath under the 
"pure" GPL (no exception).

---

So, unless you see an abvious flaw in my reasoning, there is indeed an ambiguity 
(or the exception does not carry the sematics you would like it to carry).  So, 
I am proposing that you add to the license of ALL of Classpath, libgcj, libgcc, 
and libstdc++ the following notice:

--- BEGIN ---
Note that people who make modified versions of Classpath are not
obligated to grant this special exception for their modified versions;
it is their choice whether to do so.  The GNU General Public License
gives permission to release a modified version without this exception;
this exception also makes it possible to release a modified version
which carries forward this exception.
--- END ---

This text has been authored by RMS and acknowledged by (I am sorry I do not 
remember his name) the FSF's counseling lawyer (the professor).  RMS suggested 
to include it in the license of SableVM.


So, the license of ALL of Classpath, libgcj, libgcc, and libstdc++ would remain 
GPL+exc but it would include this clarification to eliminate the ambiguity (or 
buggy semantics) I described.

> Thus you need to 
> convince
> at least Stallman, the FSF legal council, and the Gcc SC that this change
> is enough of an improvement that it is worth the hassles.  I don't see that
> happening, when you consider that Stallman doesn't even like the LGPL.


OK. o if we throw away my LGPL proposition and keep the clarification one, do 
you think there is a chance to get it throug?

> I also fear that raising the issue at this time might be a distraction from
> the issue of the AWT license.


It is a timing thing

.  As we already have to deal with a licensing problem, and debate it, why not 
take advantage of it to settle down all license problems at once, so that we can 
then vacate to coding?


--
Etienne M. Gagnon                    http://www.info.uqam.ca/~egagnon/
SableVM:                                       http://www.sablevm.org/
SableCC:                                       http://www.sablecc.org/