On Sun, Oct 31, 1999 at 01:59:24AM -0500, Raul Miller wrote:
> But the xterminal example is a bit more constrained. Here, you could
> still run into trouble -- but only if you were distributing both the
> proprietary x software and the GPLed software as composite parts of some
> larger work. [And, the "mere aggregation" clause of the GPL restricts
> the sorts of larger works which can get into trouble this way.]
I must say this thread is VERY disturbing. Have you people considered
what you're talking about? How is a non-GPL shell or environment spawning
a GPL app different than a GPL shell spawning a non-GPL app? Either way
it's the same sort of run-time connection, using the same interface. If
one is not allowed, the other is not allowed either.
If that is the case the GPL contaminates other software (like the Debian
distribution as a whole) by requiring that EVERY SINGLE THING we
distribute be GPL. A specific example of something that the GPL would be
trying to contaminate would be Apache.
Fortunately, the GPL does not do this. I think it's approaching
dellusional to believe otherwise, nothing in the GPL itself indicates that
simply running a program or having another program run it should be
considered a combined work. ANd in fact the GPL is careful to say that
mere aggregation of GPL packages is perfectly acceptable if we follow the
other restrictions in the GPL.
The reason Richard has not tried to close this apparently obvious loophole
in the GPL is probably quite simply that he couldn't do it--the law just
doesn't work that way. A Copyright license should not cover usage because
it can't legally enforce usage restrictions. And IM(NS)HO, if he tried
Richard would find he'd lost just about all respect anyone has for him.
If Richard came up to me and told me that Debian couldn't distribute
Apache anymore (advertising clause isn't GPL compatible) because it's
init.d script uses /bin/sh--bash on a Debian system and therefore violates
the GPL on bash, I'd tell him what he could do with the GPL... I have no
choice to believe he's got more sense than that.
I think we could do a lot better to focus on educating people as to why
free software is important---and not just why it makes for better
software either. Sure that's nice and all, but in the end it's that the
freedom and openness that make the software as good as it is so quickly.
Case in point, Netscape may have released their source but it wasn't until
they actually tried to open the development process that things started
moving at any measurable progress level. Having published source is nice,
but it's not enough on its own.
FWIW, I think this is a problem with Qt still today. They pretty much
want you to make whatever changes you want, diff them, and submit the
patch for their approval and possible future inclusion or not. Not
exactly a major encoaragement for people to want to work on it is it?
--
- Joseph Carter GnuPG public key: 1024D/DCF9DAB3, 2048g/3F9C2A43
- knghtbrd@debian.org 20F6 2261 F185 7A3E 79FC 44F9 8FF7 D7A3 DCF9 DAB3
--------------------------------------------------------------------------
<Knghtbrd> mariab - don't think Debian hasn't had some very stupid and
obvious bugs before
<Knghtbrd> of course, we usually fix ours BEFORE we release =D
Attachment:
pgpcJnHqJPU9m.pgp
Description: PGP signature