Fedora Freedom and linux-libre

[Alexandre Oliva]

On Jun 18, 2008, Les Mikesell <lesmikesell at gmail.com> wrote:

> Alexandre Oliva wrote:
>> 
>>> Except when the separate parts are identifiable and not derived.
>> 
>> *and* *not* distributed as part of a single work derived from the
>> Program.

> Yes, those are true - it's not derived from 'the Program'

The whole is derived from the Program and the firmware.  You don't
want it to be, I know, but repeating it won't make it any less false.

> and it is an identifiable separate section.

This doesn't matter once you realize it's part of a single
copyrightable work.

>>> "If identifiable sections of that work are not derived from the Program ..."
>> 
>> "... when you distribute them as separate works"
>> 
>> that's the same sentence you started quoting, BTW.
>> 
>> Do you dispute that e.g. tg3.c as it stands is part of a single work?

> It is an aggregate of separate unrelated parts with origins and
> destinations in separate places.

This doesn't answer the question.

>> And that it's derived from a (theoretical?  unpublished?) earlier
>> version of tg3.c that was entirely under the GPL?

> Maybe, but the owner of the code can apply whatever license he
> wants.

But in order to distribute it as part of a GPLed work, there's only
one option for the license that applies to the whole.

> I think the old GPL marking on the tg3 code
> was a mistake and a red herring in this context anyway.

Agreed.  This doesn't ake the current marking (which is what I'm
talking about) any more acceptable for inclusion in a GPLed program.

> Why not pursue this argument regarding the CPU microcode for a more
> generic disposition?

Because the CPU microcode is not part of a single source file along
with GPLed code (even though it is distributed by Fedora as part of a
single package, allegedly under the GPL)

> Putting two things together doesn't necessarily create a derived
> work.

I know.  It's modifying them for integration that does.  Or arranging
them in a way that involves a creative process.  Or selecting a
sufficient number of them under creative criteria to the point that
the collection is a creative work in itself.  There are many ways to
step away from the 'mere aggregation' case into the 'single derived
work' case.

> It may still be two separate things.  What's your distinction
> between an aggregate stored together and a derived work?

It doesn't matter what I think it is.  What matters is what the law
says.

> Being contained in the same file making them derived seems as wrong
> as two separate printed works being on the same page.

Nobody claimed it was a hard and fast rule.  But then, this is not how
tg3.c came to be.  It would be trickier to make a call if it had been
that simple.

>>> "Collective work" is not a relevant issue.
>> 
>> ... because you it would show you may be mistaken?

> No, because aggregates are permitted collective works.

Adn their distribution is permitted only under the terms of the GPL,
unless it's "mere aggregation" (i.e., it doesn't amount to a single
derived work)

>> Do you dispute that tg3.c is derived from (i) earlier versions of
>> Linux, and (ii) the firmware it contains?

> Yes.  The firmware it contained was a separate item then, and so is
> the version it contains now.

Dude!  tg3.c *contains* the firmware, and it was modified to count on
its being there.  It can't possibly not be derived from it.

>>> We already know they are separate, since they get dumped
>>> into separate hardware.

>> As in, a movie is not a single work, but rather a mere aggregation of
>> independent separate works in a single DVD, because the video gets
>> processed by our eyes whereas the audio is processed by our ears?

> Does not compute...

The fact that it gets to different hardware is irrelevant.  The
relevant point is that the whole forms a single work, and claiming it
isn't just because portions of it are for your vision hardware and
other portions are for your listening hardware doesn't make a
difference.

> But if you mean that songs used in the movie are
> still owned by the same entity that owned them before the movie, I
> think that is correct,

I didn't mean that, but I agree it's correct.  But this has ZERO to do
with whether the movie in the DVD is a single work.  Surely the song
whose snippets play during the movie are not mere aggregation, they're
an integral part of the creative work.

> and using a song in one section of the movie in
> no way affects the rights of a different one in another section.

Nobody ever disputed that.  Please don't insist on this point any
further.  It's irrelevant.

This is not about any rights as to the separate work.

This is about rights over the single work that integrates that
formerly-separate work with a formerly-GPLed project by means of the
tg3.c driver, derived from both.

>> And, come to think of it, even the video is a mere aggregation in
>> itself, because even though it's compressed, it's a mechanical
>> transformation out of a sequence of clearly separate and independent
>> pictures.
>> 
>> Right? :-)

> Yes, the representation does not affect the underlying creative works.

I think this is enough to show that you're seriously missing
information as to how copyright works, and it's become clear that you
have no interest in obtaining such knowledge.

Thanks for playing, take care,

-- 
Alexandre Oliva         http://www.lsd.ic.unicamp.br/~oliva/
Free Software Evangelist  oliva@{lsd.ic.unicamp.br, gnu.org}
FSFLA Board Member       ¡Sé Libre! => http://www.fsfla.org/
Red Hat Compiler Engineer   aoliva@{redhat.com, gcc.gnu.org}