[cc-community] A case against CC-BY-SA 4.0 -> GPL

Christopher Allan Webber cwebber at creativecommons.org
Thu Dec 15 12:17:16 EST 2011

[Sorry for posting this late, I thought I posted it earlier, but did so
under the wrong account so it probably bounced.]

These are valid concerns, and I have two responses:

Maciej Pendolski <beholder0x100 at gmail.com> writes:

> After having written some rather pro-[A]GPL posts I'm going to write
> something against CC-BY-SA 4.0 -> GPL.
> GPL option could be very divisive:
> - GPL-licenced content contributors could refuse to contribute back to
>   corresponding CC-BY-SA-licenced content
> - CC-BY-SA-licenced content contributors could get frustrated at
>   inability to use contributions to corresponding GPL-licenced content
>   (and in the most extreme situation could decide to contribute
>    only under GPL or to use custom licence in future)

A serious issue, but I think the use cases of CC->GPL are peculiar
enough where one-way compatibility is mainly about non-software content
(say, something off of opengameart.org) becoming integrated with a
hybrid resource (level files that contain simultaneous code + content).
I'm not entirely sure how this compatibility would work (I am going to
admit I don't have a super great understanding of compatibility) but my
recently increased understanding is that the cases of things like the
MPL the licensed work is not converted to the GPL... it still *is* the
MPL, but now it's compatible.  In a BY-SA meets GPL scenario, I imagine
that content pieces will continue to be released in the BY-SA section,
and code pieces will continue to be written in the GPL section.  The
only potential issue I see here is that if you combine a BY-SA work with
a GPL'ed work, that combined work may be able to skirt the attribution
provisions.  But that seems like a pretty rare risk?

Note: I'm definitely not an expert in this area.

> In a software world similar situation have happened at least once (BSD
> -> GPL) resulting in a heated debated:
> http://kerneltrap.org/Linux/Continuing_Dual-Licensing_Discussions,
> http://kerneltrap.org/Linux/Clarifying_the_ath5k_Licensing,
> http://kerneltrap.org/OpenBSD/Atheros_Driver_Developments
> Another things is reliance on FSF taking wishes of CC community into
> account when writing GPLv4 (and later). GPLv3 was very divisive when
> it was release and some groups (e.g. Linux kernel developers) have
> decided against upgrade from GPLv2. Of course similarly controversial
> changes could be made to CC-BY-SA but at least there would be less
> uncertainty about what future will bring with Creative Commons still
> being the only entity able to decide about future of CC-BY-SA-licenced
> content.

That's true, we will probably have to do it as an "v3-or-later" type
setup, but I think the risks of that are pretty low.  I have a strong
amount of trust that the FSF will not work to make copyleft
significantly weaker than it currently is, and that means the upgrade
options for users are only:

 -> stronger/stricter copyleft in further revisions
 -> just precisely the same amount of copyleft as exists in v3

I doubt the GPL is going to get more permissive than it currently is.
And if we have a risk of the FSF becoming corrupted as stewards of the
GPL, BY-SA->GPL corruptability is probably the least of the free-*
communities' concerns.

More information about the cc-community mailing list