Not logged in
Log in now
Create an account
Subscribe to LWN
LWN.net Weekly Edition for May 23, 2013
An "enum" for Python 3
An unexpected perf feature
LWN.net Weekly Edition for May 16, 2013
A look at the PyPy 2.0 release
Mark Shuttleworth on companies and free software
Posted May 17, 2011 20:25 UTC (Tue) by kleptog (subscriber, #1183)
But ISTM the issue is more bad agreements than that all such agreements are bad. Can't someone come up with a few good examples that can be used as models.
That said, what does a good contributor agreement give you (or anybody) that requiring all contributions to be BSD licensed doesn't.
Posted May 17, 2011 20:40 UTC (Tue) by jspaleta (subscriber, #50639)
When codebases are BSD license, copyright assignment doesn't create any special relicensing privileges for any contributor or contributing entity. When its BSD code, anyone can take the code a make a proprietary fork and compete with everyone else for the same market.
Mark Shuttleworth continues to over simplify the issues. The issue is not copyright assignment. The issue is the interaction of copyright assignment with copyright licensing choices in a way that deliberatively creates an unfair business advantage for one contributor over all others. He wants to paint opponents as a fundamentally oppose to assignment because he needs to point opponents as unreasonable in order for his point of view to seem relatively more reasonable. This is naked rhetorical manipulation of the discussion. At this point, he should know better. It's time to stop manipulate perception and to start having a conversation on the merits.
Posted May 17, 2011 21:54 UTC (Tue) by kleptog (subscriber, #1183)
There are contributor agreements that don't do copyright assignment, Google's for example. They just want a statement that you own the code and won't assert any patents.
But there's your statement: "... in a way that deliberatively creates an unfair business advantage for one contributor over all others". Besides that being the whole point of copyright, I have a hard time seeing this as automatically unreasonable. If some company is investing in a project, why shouldn't they get some benefits? As long as all contributions are available in an open source release, you still get to fork if you really want to. What's the catch I'm missing?
Posted May 17, 2011 22:16 UTC (Tue) by jspaleta (subscriber, #50639)
However if your employer happens to be in the business of competing with that company and a proprietary version of the codebase you want to contribute to helps that other company more effectively compete with your employer... make really really sure your employer is okay with your contributing your expertise in helping out a competitor get a specialized advantage that your employer can't equally benefit from.
Or imagine a situation where the original copyright owning company completely and utterly fails to execute on its business plan. Just utterly drops the ball because of gross incompetence at the management level after like a decade of chewing through venture capital. You and a few other independent contributors who have the technical skills to maintain the project want to create a new managing entity with better management and try your hand at the proprietary relicensing business because you want to take a shot and doing the business thing better. If the code is GPL with copyright assignment, You'll have to _buy_ _back_ the copyrights you originally _gave_ _away_ to the first corporate entity. If the code is BSD (without without assignment) you can just go for it and build the better business.
Posted May 18, 2011 0:43 UTC (Wed) by dlang (✭ supporter ✭, #313)
if there is a company doing major development and you don't do copyright assignment and they go under and you want to try your had at proprietary licensing, you have to buy all the copyrights that you don't own. It may be harder to buy copyrights from many different people than to buy them all from one entity (even if it includes things that you wrote to begin with)
there is a fundamental difference in view between the people who assume that assignment (even if joint) is valid and those who do not
those who see it as valid see the organization that manages the copyrights as creating the vast majority of the code, with the other contributers being , if not minor, at least significantly less significant.
those who see it as invalid see the contribution from the outside as being worth at least as much as that generated by the organization.
different projects will have different ratios, but at least initially, almost every project where their is a team of people paid to work on it full time, that team will out-produce the outsiders. Over time the outsiders may become a much larger portion of the development, but if the development really is very lopsided, is it really so unfair?
one thing that assignment avoids is any arguments over if a particular patch is significant enough to warrent copyright on it's own.
Posted May 18, 2011 1:00 UTC (Wed) by jspaleta (subscriber, #50639)
It's only when you mix copyleft licensing which strongly implies a co-development model with corporate copyright assignment where things get problematic and inequity arises between contributors. In this case the specific wording of the contributor agreement can matter a lot in terms of weighing the trade-offs...especially if you ever ever want to use the code you are contributing in another project or in another context.
Posted May 18, 2011 1:14 UTC (Wed) by dlang (✭ supporter ✭, #313)
but is such an extreme contract even common? much less the norm?
anything that I've seen that was an actual copyright assignment also included a license back to the author to use the code for any purpose or any context.
Posted May 18, 2011 2:00 UTC (Wed) by foom (subscriber, #14868)
Posted May 18, 2011 2:07 UTC (Wed) by dlang (✭ supporter ✭, #313)
that's a very good reason for pushing for either joint copyright assignment, or to explicitly give the project the right to dual license the code, or some other mechanism that can give the organization the rights that it is really looking for.
Posted May 18, 2011 3:13 UTC (Wed) by jspaleta (subscriber, #50639)
Posted May 18, 2011 3:34 UTC (Wed) by dlang (✭ supporter ✭, #313)
isn't that why Cannonical pushed for project harmony? It makes sense to me that while they see a problem with the current agreement, rather than trying to tweak the current agreement they instead try and work out a better document through wider discussion and only after that change their version.
it takes time to figure out how to fix things, but there is plenty of evidence that they are working on this area.
Posted May 18, 2011 4:16 UTC (Wed) by jspaleta (subscriber, #50639)
As far as I'm concerning Harmony under Chatham House rules was lost time and effort. Harmony is rebooting now with a public mailinglist. We'll see if Mark shows up on that publicly archived list and makes the case for assignment in his renewed effort to show leadership in this area. The continued lack of public discourse from him continues to be disturbing.
Posted May 25, 2011 10:26 UTC (Wed) by markshuttle (subscriber, #22379)
Harmony is not "rebooting", it's got a draft which is appropriate for discussion.
FTR, Jake's article fairly represents my commentary. Under the circumstances, with me speaking fast and him taking notes, it's a very reasonable rendition.
Posted May 17, 2011 20:51 UTC (Tue) by mjg59 (subscriber, #23239)
Posted May 18, 2011 8:05 UTC (Wed) by michaeljt (subscriber, #39183)
Perhaps you don't want anyone to be able to use your code, but you do want it to get into this one commercial product. VirtualBox lets you choose between MIT licencing and a contributor agreement.
Copyright © 2013, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds