[Foundation-l] An argument for strong copyleft

[Andrew Whitworth]

On Mon, Apr 7, 2008 at 4:14 PM, Anthony <wikimail at inbox.org> wrote:
>  Most people have no clue what the term "derivative work" means, but I
>  would assume that most people who do have a clue would agree that a
>  newspaper article which contains both photos and text is a derivative
>  work of both the photos and the text.

To be honest, I've always considered such a mash-up of objects to be
an "Aggregation", which is defined in section 7 of the GFDL. Including
a picture with some text, neither of which actually cause the other to
be modified, and both of which use different licenses, form a single
aggregate document and not a derivative.

>  The FSF has confused this point by trying to claim that in some
>  instances the text is a derivative work of the photos.  That part's
>  generally nonsense.  But the *combined* work is pretty clearly a
>  derivative work.  Of course, there's very little case law on this,
>  because it almost never matters.

If such a combined document is an aggregate and not a derivative
(which, as I said before, makes more sense to me), then the license of
the document does not necessarily have to be GFDL, so long as the GFDL
license on the free component is not impeded. The license of the
non-GFDL component is likewise not affected by the presence of the
GFDL component in the document.

Of course, this all ignores the possibility that the GFDL document
(assuming it's a short text or a single image) could be used as fair
use, and then the licensing of it matters even less as far as the rest
of the composite document is concerned.

--Andrew Whitworth