[Fwd: Re: CDDL+GPL and stuff]

[Arnoud Engelfriet]

For some reason Alex can't get through the list, so I'm forwarding his response.

Subject: Re: CDDL+GPL and stuff
Date: Tue, 17 Jul 2007 21:05:06 +0200
From: Alex van der Wolk <avdwolk at xs4all.nl>

On Jul 17, 2007, at 2:36 PM, Arnoud Engelfriet wrote:

> The only one who can create a bright line is the copyright holder
> to the work you're basing yourself on. Linus Torvalds for example
> said that applications that run on Linux are not derivative works.
> That's a bright line - but only because he said so.

At the risk of being too much of a lawyer: this is not entirely true.

Something is a derivative work, because against (some) objective
standards it is considered to be, or not. The objective standard is
set by the "average person who is representative for industry group".
But the copyrightholder himself does not have a casting vote in
whether something is a derivative work. The copyrightholder has the
sole right to sue on copyright infringement, and so he may _choose_
not to sue certain derivative works.

Such is the case with Torvalds. Legally speaking, one can argue about
the merit of such 'promises'; should Torvalds decide to sue on
applications that run on Linux - assuming they are derivative works -
I would say he is entitled to.

However, the discussion on whether something is a derivative work can
be held to the end of times. Arnoud is right that there is no clear
rule of thumb. There are some examples (most of which are not from
software cases) as to what constitutes a derivative work, but most
often you'll have to go on common sense. If a piece of software is
sufficiently autonomous (i.e. in code, not in application) and
created from scratch, it's probably not a derivative work, but a new
and original work that may apply for copyright protection. However,
if you use pieces of existing code (which is copyright protected),
and what you've created does not constitute a new work, then it is
most likely a derivative work of the pieces of code you used.

Sorry to say, it doesn't get much clearer than this.

Alex

-- 
Arnoud Engelfriet, Dutch & European patent attorney - Speaking only for myself
Patents, copyright and IPR explained for techies: http://www.iusmentis.com/