Re: GNAT GPL Edition Maintenance and Upgrades

["Ludovic Brenta" <ludovic@ludovic-brenta.org>]

Jean-Pierre Rosen a écrit :

> Ludovic Brenta a écrit :
> > Jean-Pierre Rosen a écrit :
> >>I think that this argument goes *against* the spirit of free software.
> >>As far as I can understand, the basis of free software is "you can do
> >>anything with this software, *except* deny to others the rights you have
> >>received". Anything, including proprietary software.
> >
> >
> > Precisely.  With the GMGPL, you receive the right to see and modify the
> > source code of the GNAT run-time; but you can also deny others this
> > same right.  The GPL is more "free" than the GMGPL, since with it you
> > cannot deny others this right anymore.
>
> Not exactly. With the GMGPL, you certainly cannot prevent anyone from
> seeing the source code  of the GNAT run-time; you are just not required
> to provide it.

But you can modify the run-time, ship it in binary only as part of your
program, and prevent anyone from seeing your modifications.  Thus, you
can make a closed-source version of the run-time.

> In the case of a compiler, it really does not make sense to restrict
> what the compiler can be used for. Do you imagine distributing GIMP with
> a notice saying that if you distribute any image produced with it, it
> must be exempt of rights? This would certainly be perceived as a
> restriction on freedom!

Run-time library /= compiler.  You yourself pointed that out today in
a previous post.  Therefore, your comparison with GIMP is incorrect,
because GIMP does not include any "run-time library" into images.

If you want independence from the run-time library and its license,
you know what to do: pragma No_Run_Time.

But this has been explained many times over already, here and in
countless other places where a library is placed under the GPL (e.g.
the Qt library, or the MySQL client library[1]).  There seems to be a
widespread perception that all libraries, especially ones that come
with a compiler, must always allow writing proprietary software with
them.  This is not true: the authors of the library alone decide on
that.

You may object that "the GNAT run-time library is part of the Ada
standard"; it is not.  It is an implementation of the Ada standard,
and as all implementations of any standard, it has its own license.

I agree with you that this may have been a bad move, from a marketing
perspective, both for AdaCore and for Ada in general (this is the
"marketing argument").

[1] See myth #6 in
http://dev.mysql.com/tech-resources/articles/dispelling-the-myths.html
I think it would be nice if AdaCore granted a FLOSS exception similar
to MySQL's, thus addressing the "other free software argument" which
concerns me quite a lot.

[2] I've summarised all arguments earlier on this group; for reference
see
http://groups.google.com/group/comp.lang.ada/browse_frm/thread/24ac770ebf312b7a/586cefe12b7f4066?hl=fr#586cefe12b7f4066

-- 
Ludovic Brenta.