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
>> { ... 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.
>
> There is no conflict.  Let's take this in steps.
>
> I have accepted the code under the GPL v3.  I am bound by the GPL v3.
> Are we agreed about this?

So far, so good.

> I am allowed by the GPL v3 to license my contribution GPL v3 or later.
> Are we agreed about this?  (Note that my so licensing it does not
> change *my* obligations one bit, I still have to follow the GPL v3.)

> The GPL v3 allows me to distribute the modified result to a third
> person, say Carol.  Of course it does so under its terms, which I am
> obliged to follow.  Are we agreed about this?

With you so far.

> Carol has received code from me.  From the copyright licenses she has
> been offered part of it under GPL v2 or later from the original
> author, and some modifications (from me) under GPL v3 or later.  Are
> we agreed on this?

Not quite on the same page here.  See below.  I believe Carol is getting
a license grant from the original author but *only* under the terms you
have
chosen to convey your variant.

> As section 10 of the GPL v3 makes clear (there was a similar section
> in the GPL v2, and in fact I understand that this section is merely a
> restatement of copyright law), Carol's acceptance of the license is in
> no way, shape or form contingent on my acceptance of the license.
> Therefore she is allowed to accept the permissions of the GPL v3.  If
> a GPL v4 is out, she is allowed to accept that set of permissions
> instead.  If she can track down all of the copyright holders and
> negotiate her own terms, she can accept it under any copyright license
> she can get us to agree to.
> 
> In short I can take code licensed "GPL v2 or later", make a
> modification and license that modification "GPL v3 or later", then
> give it to someone and at that point the code really IS GPL v3 or
> later.

Here is where we part company.  Looking at the V2-or-later
license from the original copyright holders, if you view the "or later"
as
(in GPLv3 terminology) an additional permission, you are allowed to
remove that additional permission in copies which you distribute (convey
in
GPLv3 terms).  If you choose to remove the additional permission and
convey under V3 only, this is allowed.  However, in this
scenario, when Carol gets such a V3-only copy from you, the
permission-less
license is the one which Carol must follow, not the original V2-or-later
license with a permission which you have removed.  (Of course Carol may
always go back to the ID and get the original V2-or-later version if she
so chooses.}

Also, since you are not the copyright holder in the original code, you
may not add your own
new permissions to the original code.  So, here is my problem with your
scenario: I don't see how you have rights to remove the "or later"
permission, accept the original code under V3 only, and then add an "or
later"
permission back to the original code when you convey it.

BTW, I would actually like to be wrong here, because I don't
particularly like the
answer my analysis leads to!  So, if there is indeed a convincing
argument that
does not involve adding permissions to code where you are not the
copyright owner, I will be predisposed to accept it.

Cheers

Andy Wilson
Intel open source technology center