Re: [tlug] Bill Gates and the GPL , let the flames begin

["Stephen J. Turnbull" <stephen@example.com>]

Daniel A. Ramaley writes:
 > I'm more likely to contribute code to a GPL project though, knowing
 > that some commercial entity can't take it without contributing back
 > their changes.

Well, "can't" is awfully strong.  The internal deployment loophole
(Morgan Stanley has a really nice version of XEmacs, I hear, but I've
never even seen it) and ASP loophole are always available (although
the latter is closed by the Affero GPL).  And consider Curt's example
of libreadline.  What is libreadline but an input method?  So run it
in a separate process and communicate with it via XIM.  GPL can't
touch that!

Note that if the commercial entity actually would trigger the GPL (by
distributing the software), then it *is* contributing back.  If the
modified software does something visibly useful, then it is
contributing that feature to the community, even if the implementation
is proprietary.

 > Sure, the original code would still be free, but if it becomes 
 > abandonware it won't be useful after a few years when the software 
 > world moves on. If code is still worth using at all, i'd rather 
 > maintenance of it be contractually enforced on those wanting to keep 
 > using it.

I'm not sure what you're saying, here.  If you look at what actually
happens with GPLed software, you see that it gets dumped on the
community after the initial contribution.  In the case of the Linux
kernel, this has worked really well.  In the case of Infodock, a (for
its time) very slick upgrade of XEmacs, the company went bankrupt and
Infodock is too big to update on volunteer time.  In the case of OOo,
I have to wonder whether Sun really thinks its a good idea any more,
since "the community" really has been unable to take over maintenance.

I would say that GPLing your contribution substantially increases the
probability of its abandonment in many cases.