Concurrent Licenses?

[W. Yip]

On Mon, 10 Apr 2000 10:51:16 -0400, "Rod Dixon, J.D., LL.M."
<rod at cyberspaces.org> wrote:
>I have noticed this A to B to C argument before, but I am uncertain whether
>it is applicable if the GPL has a strong copyleft provision. The GNU GPL is
>better viewed as A to B; A to B/C; A to C; A to D; and so on. 

But doesn't this omit the fact that B's derivative 'bits' may evoke
copyright subsistence, thereby requiring a concurrent license from B to C
in addition to the license from A to C? Otherwise, C only has license to
the initial program, but not the derivative 'bits'?

I think this situation is particularly so in Berne signatories where
copyright does not have to be registered in order to subsist.

The bottom line is that A cannot license what he does not own. And A
certainly cannot own B's copyright to the derivative 'bits' if these indeed
do subsist.

>If you keep in
>mind that the GNU GPL's purpose is to recapture copyright in instances where
>the public domain could not, then you should be better able to focus on why
>A is the only party who may bring suit if the GPL is violated. This fact
>changes dramatically under open source public licenses that do not have a
>strong copyleft. Indeed, your example, generally,  is correct for GPLs with
>no copyleft.

Please explain what is "GPLs with no copyleft".

I do not agree that only the copyright holder can sue. Surely an owner of
copyright to the derivative bits has sufficient interest at stake to have
standing to sue? No doubt he first has to show copyright subsistence, but
once he has established that, I don't see a reason why he can't sue.