On Mon, 2008-04-28 at 13:35 -0400, max bianco wrote:
> On Mon, Apr 28, 2008 at 11:47 AM, Matthew Saltzman <mjs@xxxxxxxxxxx> wrote:
> >
> > On Mon, 2008-04-28 at 10:41 -0400, max bianco wrote:
> > > On Mon, Apr 28, 2008 at 12:23 AM, Les Mikesell <lesmikesell@xxxxxxxxx> wrote:
> > > >
> >
> > > > Which is a bizarre thing to be concerned about because the only thing they
> > > > could possibly do to diminish the value of the original copy would be to
> > > > improve it so much that no one would want the original. As a potential user
> > > > of that improved version, I think that restriction is a bad thing. And most
> > > > bizarre of all is the notion that I can't obtain my own copy of a GPL'd
> > > > library, and someone else's code under their own terms separately.
> > > >
> > > The hard work is done by the original author. So if I understand you
> > > correctly, its ok with you if i use your code, improve it, and
> > > relicense it so what you freely contributed is now going to cost you
> > > money. So your hard work now belongs to someone else.
> > >
> >
> > I don't think anyone is talking about modifying your code and
> > relicensing it. That would clearly be a derived work, and there's no
> > question you can impose conditions on its redistribution.
> >
> > You write a library. I write a program that calls routines in your
> > library. Now the question is whether your license can impose conditions
> > on my distribution of my own code. That's a fuzzy, gray area, but (to
> > mix a metaphor) it's just the tip of the iceberg of complexity.
> >
> > ChipCo creates a piece of specialized hardware and releases a
> > proprietary driver. I write code to interface your library and the
> > ChipCo driver. Can your license prevent me from distributing my code?
> > If so, you and I might have a reasonable disagreement about whether
> > that's a good thing. But you can't deny that some people who might
> > benefit from my code (and by extension, your code) are prevented from
> > doing so. You can only argue that some greater good is served by their
> > suffering. Note that I want to be generous with my code and release it
> > under an open-source license; I'm not trying to unfairly benefit from
> > your work.
> >
> > You write a library and distribute it under an open-source license. I
> > write a library and distribute it under a slightly different--but
> > incompatible--open-source license. Les writes a program that links to
> > both libraries. If your license can impose conditions on Les's
> > distribution of his program, then users who would get value from Les's
> > program are SOL. Note that nothing here violates the spirit of OSS.
> > Everyone involved wants to be generous. Nobody is trying to unfairly
> > benefit from anyone else's work. But due to a technicality, nobody can
> > benefit from Les's work at all! That seems like a shame, doesn't it?
> >
>
> Yes it does but what then is the answer?Everybody argues that A is
> right or B is wrong or c....you get the idea. What is the solution?
> Let's stop going over the same ground and come up with some kind of
> solution. The end user is ultimately the only one that matters, i
> think everyone can agree on that, if the end user cannot get their
> work done then everyone suffers, so what should we as end user's
> do?should i have to pay for a brand new office suite when nothing
> substantial except the companies desire to support it has changed?That
> is an example not a way to drag M$ into this, so please lets leave the
> M$ bashing where it belongs. this will of course create another debate
> but at least we will subtly change the content of the conversation.
I don't know that there is a retroactive "solution". I know there is no
"one license to rule them all"--GPL partisans will insist on serving the
Greater Good, whomever is inconvenienced along the way. Other OSS
licenses exist precisely because other developers don't want to subject
themselves or others to the GPL restrictions (though they agree with the
sharing spirit of OSS).
Going forward, using a license that doesn't impose these kinds of
restrictions would help. As an example, MySQL uses a GPL with a clause
exempting other FOSS licenses, which resolves the issues in my last
scenario. There are several other variants on the idea of GPL with a
FOSS exception. LGPL, CPL, and several other licenses that explicitly
do not extend to linked code (but do cover modifications to the library
itself) resolve all three scenarios.
Case law reconciling the definition of "derived work" may or may not
resolve the issue another way.
>
> Max
>
>
--
Matthew Saltzman
Clemson University Math Sciences
mjs AT clemson DOT edu
http://www.math.clemson.edu/~mjs
--
fedora-list mailing list
fedora-list@xxxxxxxxxx
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
