Subject: Re: Standing on multi-authored works
From: Brian Behlendorf <brian@collab.net>
Date: Tue, 19 Apr 2005 10:52:33 -0700 (PDT)

On Mon, 18 Apr 2005, cody koeninger wrote:
> More to the point, in the U.S. at least, in order to
> bring suit on a copyright claim, you need to be the
> owner or exclusive licensee (nonexclusive isnt
> enough).  Otherwise you have to join everyone who has
> an interest in the work.  See 17 usc 411 & 501(b)
>
> In other words, the FSF's practice of getting
> assignments is necessary in order to enforce the GPL.

IANAL, but my understanding is as follows.  You can enforce the collective 
copyright without being the exclusive licensee of the copyrights within 
the collection.  It means that an infringer can infringe an isolated work 
from within your collection, but you can enforce the copyright on the 
combination of those works as found in your collective work - or even a 
subset that contains more than one work, as combined in the collective 
work.

In the example usually given of an anthology of poems where the poets 
license non-exclusively - an infringer could copy an individual poem and 
only the original author would have standing to sue.  But copying a whole 
set of poems as found in the book would give the book publisher standing.

Translated to an Open Source setting... we are preparing anthologies of 
works where the original work is delineated from the others only by 
examining the revision control history.  If someone creates a derivative 
work of the Apache web server, this new work will almost certainly contain 
the contributions of more than one owner, as combined in the Apache web 
server.  In that case, Apache would have the standing to sue for 
infringement, even though we do not get exclusive assignment from our 
contributors.  A determined infringer could jump into the revision control 
history and isolate a contribution made by a single owner - perhaps the 
initial load of some work done elsewhere by that contributor.  That 
isolated contribution would have to not be a derivative work of the 
collective work - so no bugfixes, for example, and probably no patches 
that change behavior.  If that isolated work were to be infringed, only 
the original author would have standing.  But far more often we're worried 
about the collective work - the Linux kernel, for example, rather than 
simply someone's individual kernel module.

As I understand it, Harald Welte's successful actions to enforce the GPL 
in Germany were based on the inverse - a single contributor suing to 
enforce the copyright terms of his license on his code in the Linux 
kernel.  In this case, the "distributed" nature of the copyrights in Linux 
could actually be seen as an advantage - it gives many more individuals
standing to sue to enforce it.  Harald's targets could have avoided 
problems by stripping out netfilter (Harald's work) and using some other 
packet filtering technology; this is the main weakness of such a 
distributed ownership setting.  I don't know if this is why one would say 
the U.S. would require "everyone else who has an interest in the work to 
join" in a similar suit; my hopeful guess is that you'd only need 
"everyone" if you wanted to prevent the infringer from being able to use 
the works of someone who didn't join you in enforcing the copyright.

Anyone know if there's been a case already that has considered the 
interwoven nature of contributions to Open Source works when compared to 
"anthologies of poems" in considering who has standing in an infringement 
of the collective work?

 	Brian