When to evaluate dual licenses (was: license categories, was: I'm not supposed to use the ECL v2?)

[Wilson, Andrew]

Ben Tilly wrote:
>Andrew Wilson wrote:
>> This is a very interesting question.  Let's assume, for purpose of
>> discussion, that
>> it is axiomatic that the license of a derivative work of GPL code ==
>> effective license of the
>> original GPL code.  {Yes, I know, not everyone accepts this.  Stop
>> reading now
>> if you don't.}
>
> I have a sneaking suspicion that I've just been asked to stop reading.
> But I have to ask.  Why would this be axiomatic?

Let me try a more precise formulation: it is axiomatic that the
license of a derivative work of GPL code must be a license which
is permitted by the license of the original GPL work.  Better?
Just trying to prevent this thread from being sidetracked by those
who question the validity of GPL-imposed licensing constraints on
derivatives.  
For anyone of that mindset, this entire thread is presumably moot.

{ ... discussion follows about permitted licenses of GPLv2-or-later
derivatives ...}

> My theory is that you accept the licensed code under GPL v3 and
> release your modifications as GPL v3 or later.  That makes the license
> on the combined work be the intersection of the original (GPL v2 or
> later) and your code (GPL v3 or later) which is GPL v3 or later.

Hold on a minute, partner.  If you have accepted the V2-or-later
licensed code
under V3 and combined it with V3-or-later additional code, the
intersection
of these licenses is V3-only.  I do not see how you can both
"accept the licensed code under GPL v3" and then immediately revert
to the original V2-or-later inbound license. 

Andy Wilson
Intel open source technology center

under GPL v3" and pass