Re: testing before a release

[Mark Wielaard]

Hi,

On Fri, Nov 09, 2001 at 02:41:03PM +1300, Bryce McKinlay wrote:
> Tom Tromey wrote:
> 
> >I've always assumed that since libgcc has used this license for years
> >(probably more than 10 years), that it is well understood and
> >acceptable (for its purpose).  If it has problems, I think they should
> >be addressed through RMS and the gcc steering committee.  If this
> >license is unworkable for libgcj then it is a hundred times worse for
> >libgcc -- libgcc is required by every gcc-compiled program.
> 
> Its unfair to lump classpath/libgcj and libgcc together here. The libgcc 
> license has and does seem to work well - for libgcc. This is a small 
> compiler support library that makes little sense outside the scope of 
> supporting object files compiled gcc. But classpath and libgcj are much 
> larger and broader in scope than libgcc or even libstdc++.

Since most runtime support librarys in gcc such as libgcc, libstdc++
and libobjc for example use the GPL+Exception is was not a strange request
from the GCC team to have the java runtime libraries also under the same
license. But I do agree with you that libgcj (Classpath) is much bigger
in scope.

Is there a way we could define the runtime parts of Classpath and put
those under the GPL+Exception and still have a meaningfull small subset.
Then all other classes could be put under the normal LGPL.

I always hoped that that was a way to solve the AWT issue. Just relicense
the libgcj parts under the LGPL as they are currently in Classpath but
that didn't seem to be such a hot option for the GCC people. But I assumed
that this was because everybody really didn't like to have two different
licenses. Which I think is indeed really confusing.

Cheers,

Mark
-- 
Stuff to read:
    <http://www.toad.com/gnu/whatswrong.html>
  What's Wrong with Copy Protection, by John Gilmore