GLADE and GPL (was: what DOES the GPL really say?)

[fielding@kiwi.ics.uci.edu (Roy T. Fielding)]

>Perhaps I am confusing David Weller and David Wheeler.  I am referring
>to the Ada Community License--much shorter than the GPL, just as clear
>if not clearer, and which does not impose its terms on other code by
>association.  It is completely independent of the GPL--unless of course
>you try to mix GPL code with ACL code.  :-)

Which is exactly my problem.  I am a free software developer without the
attitude problem (witness wwwstat, MOMspider, libwww-perl, Apache, etc.).
Traditionally, I have used either the UC (Berkeley) license or the
Artistic License to distribute my source code, because I actually *want*
to see commercial providers reuse it (with acknowledgement).

I am working on a project that needs an Ada95 binding to sockets, and I
happen to know that there is one enmeshed within the DSA code of Garlic
(the guts of GLADE, which is distributed under the GPL).  I could very
easily extract that code and distribute it with my project's source,
but in doing so I'd infect my better-than-GPL free source code
with the GPL virus.  Unless the GLADE folks can be convinced to use the
Ada Community License, which is just the Artistic License updated to
refer to Ada libraries instead of C and perl, then the best I could
do is create a separate Garlic-lite package with GPL terms,
distribute the two packages separately (which wastes my time), and
inform third-parties that they'll have to write their own sockets binding
if they want to use my library without distributing their own source.

The question for the GLADE developers is why are you distributing under
the GPL?  Is it really your intention to prevent proprietary use of
the GNAT Distributed Systems Annex?  If not, then LGPL is more appropriate,
though even the LGPL has some bone-headed provisions.

If you just want to receive credit for your work, and don't have the
attitude that all derivatives must be LGPL/GPL'd, then just use the
Ada Community License.  Your project will be much more useful to others,
and more successful in the long run.

 ...Roy T. Fielding
    Department of Information & Computer Science    (fielding@ics.uci.edu)
    University of California, Irvine, CA 92697-3425    fax:+1(714)824-1715
    http://www.ics.uci.edu/~fielding/