Re: More GPL questions

[Stefaan A Eeckels]

On Tue, 17 Oct 2006 17:49:53 +0200
David Kastrup <dak@gnu.org> wrote:

> Stefaan A Eeckels <hoendech@ecc.lu> writes:
> 
> > On Tue, 17 Oct 2006 14:51:48 +0200
> > David Kastrup <dak@gnu.org> wrote:
> >
> >> Stefaan A Eeckels <hoendech@ecc.lu> writes:
> >
> >> > An original program in source code format, and contains function
> >> > and/or system calls does not consist of "revisions, annotations,
> >> > elaborations or other modifications" to the libraries or the OS.
> >> 
> >> Sigh.  But a literary work consisting of annotations does not
> >> contain material from the original work.  It is, as a whole, an
> >> original work of authorship.
> >
> > You cannot "annotate, revise, elaborate or otherwise modify" without
> > anything of the original work.
> 
> I have here a secondary literary work covering "Ulysses", consisting
> pretty much exclusively of annotations.  Where there are citations,
> they are short enough not to count as copyrightable in itself.  But it
> certainly is a derivative work.

Because it's "raison d'être" is to comment and annotate "Ulysses". I
did not claim that such works are not derivative works - it's pretty
obvious that using the same cast of characters creates a derivative
work, even though not a single sentence of the original novel is
present in the derivative work (cfr. the new Peter Pan novel).

> > Programs that use a library or an OS are not "revisions,
> > annotations, elaborations or other modifications" of the library or
> > the OS.
> 
> Naturally.  So one has to translate the examples from the context of
> literary works to that of computer programs.

Where the idea that one could have a derivative work (or something else
covered by the copyright of a library) that contains no code from that
library is far less evident than in the case of literary works. 

> > In the specific case of "annotations, revisions, elaborations and
> > other modifications", which supposes that there is a work that is
> > revised, annotated, elaborated or otherwise modified". Are you
> > claiming that all programs are modifications, elaborations,
> > revisions and annotations of the OS and the libraries?
> 
> I am not claiming any such thing.  I am just saying that the lack of
> direct verbatim inclusion of a copyrightable amount of material is not
> a necessity in the explanation for literary works, and so it is not
> obvious why it would have to be a necessary criterion in the case of
> software.

I would go as far as to say that in the case of software, I cannot see
how a program (compiled or not) that contains no code from another
program or library could ever be covered by its copyright. Quite
clearly, some (like the FSF) believe that because static linking came
first, and resulted in executables that obviously were a combination of
the program and the libraries linked to it, dynamic linking must have
the same copyright implications. I for one, don't believe this to be the
case, but IANAL. 

-- 
Stefaan A Eeckels
-- 
"Treason doth never prosper. What is the reason?
For if treason prosper, none dare call it treason."