[ILUG] The old CD thread again {RHEL this time}

[Rick Moen]

Quoting Paul Jakma (paul at clubi.ie):

> For the benefit of Moen (i couldn't resist looking at the 
> online archives):
> 
> "Of course, as stressed elsewhere in this thread, Red Hat, Inc. can 
> speak _only_ to the licensing of properties it _owns_, an elementary 
> aspect of copyright law that seems to elude Paul."
> 
> I suggest Rick considers that collective works (like an ISO) can have 
> their own distinct copyright from works which comprise it.

Oddly enough, I've already covered compilation copyrights -- which is
what Paul probably means if he means _anything_ substantive -- separately,
but Paul seemingly did not notice.


The term "collective work" actually refers to a codebase (or other
creative work) with multiple authors, where legal ownership is
construed in a particular fashion denoted by that particular bit of
jargon.  The alternative theory of copyright law that can be used to
interpret such situations is "joint work".

Depending on which gloss a judge applies to a multi-author work,
different sets of legal rights and obligation will apply.  That is why
the Licensing HOWTO devotes a fair amount of coverage to the
distinction, and urges maintainers of multi-author codebases to consider
declaring their projects to be (specifically) collective works, and not
joint works.  (See the HOWTO for details.)

But none of that is relevant to "an ISO", which is _not_ a collective
work, because it is not a single codebase.  It is a multitide of
codebases that all happen to be on a CD.

As described separately, the person (or company) That assembles such a
collection can assert a type of title over the selection/arrangement of
other people's (and, where applicable, their own) property called a
"compilation copyright".  The law would recognise that property claim as
valid if the act of assembling/arranging is judged sufficiently
creative to justify recognition as a creative work in itself.  For the
standard example of compilation copyright being _denied_ by a court on
account of insufficient creativity, see Feist Publications v. Rural
Telephone Service Company, Inc.:
http://www.bitlaw.com/source/cases/copyright/feist.html

(Essentially, Feist assembled and published telephone directory listings
created by others alphabetically, and claimed that doing so allowed it
to enjoin Rural Telephone from duplicating its "creative work" in
sorting listings in A-Z order.  The court said, "Sorry, you lose.")

In any event, RHEL's "EULA" GPL-grant statement in the top-level
directory of its CD images applies to its compilation-copyright title --
if that title is actually valid, a matter that has never been
adjudicated and probably never will be.



> "I've already explained -- as has Red Hat, Inc., for that matter -- 
> the company's _conditioning_ of its licence for two RPMs' contents on
> observance of the company's trademark policy. "
> 
> The text of Rick's which I responded to claimed that RedHat were 
> intermingling the contents of those two non-GPL packages which are 
> not licenced under the GPL grant for the collective work either, into 
> other packages "during compilation" (i presume he meant compilation 
> of the ISO, not of the packages themselves - latter doesnt make sense 
> given the packages are seperate). My understanding is that this 
> simply not the case, either way.

I've explained fully, in a prior post, the manner in which a
trademark-encumbered image file (drawn from one of RHEL's two
trademark-encumbered SRPMs) might well cause a binary RPM compiled
from otherwise open-source SRPM source code (using
rhn-applet-2.1.17-5.src.rpm as a hypothetical example) to become, in
effect, proprietary in the sense of installing a logo file licensed only 
under proprietary terms.

If Paul wasn't paying attention, that's really not my problem.


> Further, even if it is the case, and RedHat do stuff images into GPL 
> packages which are trade-marked, there still isn't /yet/ a problem. 

I not only didn't see a "problem", I said I admire the strategy, and
consider it entirely benign.

In short:

Paul essentially seems to want to argue against dumbass straw-man
assertions of his own devising.  I have no objection to this; I'd just
ask that he _cease_ claiming that those assertions have anything to do
with me.  And thereby cease wasting everyone's time.

That, of course, is not likely to occur, but I can hope that he'll
eventually find some other hobby that distracts him.