User: Password:
Subscribe / Log in / New account and contributor agreements and contributor agreements

Posted May 21, 2011 13:35 UTC (Sat) by webmink (guest, #47180)
In reply to: and contributor agreements by kleptog
Parent article: and contributor agreements

So you're saying whether a CLA requires copyright assignment or not is not important since it makes no practical difference? Sounds odd, it must make some difference.

No, it makes no practical difference. As I point out in my essay on the subject linked in the article, the problem with all CLAs is that they anoint one member of a community with broad rights to the whole work that no other member enjoys. The inequality that creates is the toxin. It makes no difference how the inequality is administered, whether by giving the copyright away to them (assignment) or granting them a license that conveys rights equivalent to ownership. They are then free to work on the code outside the community, make deals around it without including the community and generally violate the transparency principle. I've covered that in an essay too.

You also suggest that you can give permission to re-licence without assigning copyright. I didn't think that was possible.

Yes, the license these agreements use is broad enough to permit the recipient to relicense under any arrangement they like. That's the whole point in most cases.

Google's CLA seems fairly harmless, perhaps I'm missing something.

Clause 2 of the agreement you point to grants Google "a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license" - that's enough to allow them to do anythign at all they want with your contribution, including distributing it under a proprietary license in another work without advising you. If that's no problem to you then yes, it's just as harmless as all other CLAs.

(Log in to post comments) and contributor agreements

Posted May 21, 2011 14:50 UTC (Sat) by rahulsundaram (subscriber, #21946) [Link]

I don't think the Fedora agreement has this problem. It is not described as a CLA but I have seen others call it that and contributor agreements

Posted May 21, 2011 14:57 UTC (Sat) by webmink (guest, #47180) [Link]

Indeed, it is explicitly not a CLA - it merely asserts community norms for licensing and seeks consent to them from all participants. It does not create a community inequality. I have been suggesting calling this sort of agreement a "participation agreement" to avoid confusion with CLAs but there's some reason Spot opposes that terminology (perhaps because I coined it?). and contributor agreements

Posted May 24, 2011 2:59 UTC (Tue) by spot (subscriber, #15640) [Link]

I don't oppose that terminology because it came from you, but rather, because it is possible to participate in a community without making a copyrightable contribution of code or content. Thus, I felt that the choice of wording was poor and chose to stick with "contributor", because even though that term could also be misinterpreted, it has a reasonably well established meaning in FOSS. and contributor agreements

Posted May 25, 2011 19:42 UTC (Wed) by webmink (guest, #47180) [Link]

Thanks for the reply. Why would you only want the people who are contributing copyrightable materials to agree to your community norms as they participate? and contributor agreements

Posted May 25, 2011 22:24 UTC (Wed) by rfontana (guest, #52677) [Link]

The particular community norms in question here only concern copyrightable material. and contributor agreements

Posted May 21, 2011 19:19 UTC (Sat) by pphaneuf (subscriber, #23480) [Link]

I'm not sure what "sublicensing" means precisely, but at any rate, the Apache license (which is used with most Google projects, I think, and the Google CLA is pretty much the same as the Apache one) isn't a copyleft, it's more similar to a BSD in spirit, and you could always make a proprietary fork of those, so this is nothing new, CLA or not.

If you're a strong proponent of copyleft, then this might not be acceptable, but is the CLA making it any worse? and contributor agreements

Posted May 22, 2011 13:42 UTC (Sun) by kleptog (subscriber, #1183) [Link]

Well, IANAL but 'sub-licensing' means that any right you give Google, that they have the right to licence those to others.

If you look at the GPL, you'll see it talks about sub-licensing but says it's unnecessary since the GPL gives the receiver all the rights of the GPL from the original author so there is nothing to sub-license. So AFAICS it has no effect on strong-copyleft either.

It's certainly not assigning copyright. I don't think it can because in several jurisdictions you can't do that with an electronic form anyway. My theory is that the Google CLA is designed to be able to cover anything, not just open-source stuff, and that its effect on non-open source stuff is more profound. Documentation you submit comes to mind.

Still, I'd like someone with legal training to actually write something (you know, with citations and research) about the interaction between contributor agreements and copyright licences, because it vaguely feels like a dual-licensing arrangement, which always seemed like a dodgy part of copyright law to me anyway. and contributor agreements

Posted May 22, 2011 23:26 UTC (Sun) by pphaneuf (subscriber, #23480) [Link]

I just took a side-by-side read of the Apache and Google "individual" CLAs, and it seems to be almost identical, mostly changing "Foundation" with "Google", and I think the Apache one is older? So I doubt it's authors had much thought for non-open-source? I might be wrong here, this is just conjecture.

I'd be interested as well in such a article/paper.

Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds