Re: Licensing Problems with Debian Packages (Was Re: Copyright lawyers analysis of Andreas Pour's Interpretation)
Raul Miller wrote:
>
> Raul Miller wrote:
> > > You've been implying that the irreversible change mentioned in Section
> > > 3 is a requirement for cases where libc is used as a Library.
>
> On Mon, Feb 21, 2000 at 12:27:45AM -0500, Andreas Pour wrote:
> > Well, it seems you misquoting me is becoming an annoying ritual. What
> > I said is that under the interpretation -- an interpretation that
> > obviously is not mine -- that Section 3(a)/Section 2(b) of the GPL
> > require the "complete source code" -- including library source code
> > (like qt) -- to be licensed under the GPL, libc would have to be
> > licensed under the GPL. The next step in the reasoning is that if libc
> > has to be licensed under the GPL, you have to go through the Section 3
> > (LGPL) conversion.
>
> Except that libc (or any other LGPLed library) is not the complete source
> code for grep (or any other GPLed program which might use libc or some
> other LGPLed library). You do realize that, I hope.
>
> When applying the GPL when multiple licenses are involved means that the
> work as a whole is a collective work, and the individual licenses remain.
OK, perhaps we are making progress after all. It appears
that you have now
abandoned the argument that Qt itself must be licensed under
the GPL. So if
that is true, all you require is that the collective work in
kghostview/Qt
be licensed under the GPL, with "collective work" having the
meaning provided
in the Copyright Act.
Please clarify/correct any mistake in the above summary, and
we can proceed
from there.
>
> > > But this third paragraph clearly indicates that Section 3 is
> > > optional. That the use of Section 3 is for cases where the code (for
> > > example, the lgpled libc code) would not be not used as a library.
> >
> > It's just an example. Anyway, I don't think libc has to be licensed
> > under the GPL when linked with a GPL Program, for the very same reason
> > I don't think Qt has to be licensed under the GPL when linked with a
> > GPL Program.
>
> It's statements like this that make me think you don't understand
> collective copyrights. Why else would you bother making such assertions?
I find this statement puzzling. Is it not your position
that libc is not
part of the "complete source code"? You stated that above.
If that is
true, then Qt must also not be part of the "complete source
code". If you
disagree with this, please explain why libc is not, and Qt
is, part of the
complete source code, for purposes of Section 3(a) of the
GPL.
>
> > > [For that matter, the first sentence of the first paragraph also
> > > clearly indicates that Section 3 is optional, but the third
> > > paragraph has the advantage of showing why the option is available.]
> > >
> > > To use a phrasing similar to what you've been using in other
> > > contexts: a Library need not be a Program.
> >
> > Good. So you now agree that the "complete source code" of Section 3(a)
> > of the GPL does not have to be licensed under the GPL. But no, of
> > course you don't.
>
> Once again, libc is not the complete source code. In some examples,
> it's a part of the complete source code, but the individual parts get
> to retain their own licenses.
By some examples you mean if you take some libc source code
and use it in
a GPL'd program. I am referring strictly to the case of
dynamic linking.
In that situation, do you see libc/Qt being part of the
complete source
code for purposes of GPL Section 3(a), or not? A simple
"Yes I do"/"No I
don't, for the following reasons . . . ." answer would be
nice :-).
Ciao,
Andreas
Reply to: