[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Re (2): testing before a release
From: |
Etienne M. Gagnon |
Subject: |
Re: Re (2): testing before a release |
Date: |
Fri, 16 Nov 2001 17:16:21 -0500 |
User-agent: |
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.5) Gecko/20011023 |
Stuart Ballard wrote:
It's possible to modify a GPL'd app to use a proprietary library, and
run that, too, so long as you don't distribute the result (the fact that
the modified app is a derived work is clearer in this case, too).
The GPL only kicks in for distribution. Period. If that's not the case,
then I've been very much misled in every licensing discussion I've ever
had, and by the GPL FAQ and other FSF documents on the GPL. And if
that's true then someone ought to really raise a stink, because I doubt
I'm the only one.
Here's an edited extract of an email between RMS and myself about the subject.
...
RMS: You're assuming that "making normal use" cannot be making a derived work.
That assumption is not right, though. Making normal use of a module
CAN be making a derived work. For example, for a library, normal use
is ALWAYS making a derived work.
Etienne: So, what would be your answer to the question:
"Can I run proprietary Java applications and native JNI libraries under
Kaffe?"
RMS: My answer would be, no you can't.
Etienne: How come "running" a non-free Java program on Kaffe would violate
the GPL license? This is a point I have difficulty to understand.
RMS: Because it is linking the two together.
...
Etienne: Just to confirm that I now understand correctly...
Linking (dynamic and/or static) of an GPL work "A" with another work "B"
is addressed by clause 2 of the GPL, namely: "you may *modify* your
copy...". This is because linking "A" with "B" does *modify* "A".
The analogy would be to say that "linking" a work by Shakespeare (which
is in the public domain) with an "introduction and commentary" does
create a derived work, even though no word has been changed in
Shakespeare's text. So, the simple act of "linking" constitutes a
modification of the original work. Right?
RMS: Right. It makes a collective or combined work.
Now, you can ask RMS for clarifications, if you want.
Etienne
--
Etienne M. Gagnon http://www.info.uqam.ca/~egagnon/
SableVM: http://www.sablevm.org/
SableCC: http://www.sablecc.org/
- Re: Re (2): testing before a release, (continued)
- Re: Re (2): testing before a release, Etienne M. Gagnon, 2001/11/07
- Re: Re (2): testing before a release, Brian Jones, 2001/11/07
- Re: Re (2): testing before a release, Per Bothner, 2001/11/07
- Re: Re (2): testing before a release, Fred Gray, 2001/11/08
- Re: Re (2): testing before a release, Mark Wielaard, 2001/11/09
- Re: Re (2): testing before a release, Etienne M. Gagnon, 2001/11/09
- Re: Re (2): testing before a release, Mark Wielaard, 2001/11/16
- Re: Re (2): testing before a release, Etienne M. Gagnon, 2001/11/16
- Re: Re (2): testing before a release, Stuart Ballard, 2001/11/16
- Re: Re (2): testing before a release,
Etienne M. Gagnon <=
- Re: Re (2): testing before a release, Brad Cox, 2001/11/16