451 CAOS Theory 
A blog for the enterprise open source community
Copyrights and wrongs
Matthew Aslett, February 8, 2010 @ 11:16 am ETOne of the issues I have with the Free Software approach is that advocates have habit of throwing the baby out with the bathwater when discussing issues that they see as in any way negative to free software.
I was reminded of this while reading Bradley M. Kuhn’s criticism of Mark Shuttleworth’s reported views on copyright assignment.
Having read the original interview with Mark, and then Bradley’s response, it is pretty clear that the two have very different perspectives on copyright assignment: Mark is speaking from the perspective of a commercial business, Bradley form that of a non-profit foundation.
The two entities have very different reasons for enforcing copyright assignment policies, and Bradley is right to point out that a potential contributor should approach a copyright assignment policy from a commercial entity with a great degree of caution.
However, the ultimate reason for enforcing copyright assignment is about control. From a vendor’s perspective the desire for control is often to produce closed versions of the code. From the FSF’s perspective the desire for control is about keeping the code, and derivatives of it, open.
However, the fact that the FSF “promises to never proprietarize its versions of the software assigned to it”, does not support Bradley’s assertion that Mark “wants to confuse us about copyright assignment so we just start signing away our software”.
This claim is especially problematic given that Mark appeared (and it must be said we are reliant on the reporting of his statements to understand what he meant by them) to be attempting to reduce confusion around copyright assignments by, if possible, introducing some sort of standardization.
This is a suggestion that deserves more consideration. However, Bradley is so busy protecting the FSF from being maligned by Mark that he completely ignores the point raised by Mark – that copyright assignment policies are confusing, complex, and potentially problematic.
As the iTWire report demonstrates, the issue of copyright assignment is not just one that impacts contributions by individual developers (which is a common way of looking at it) but also of contributions from employees of Canonical to projects led by the likes of MySQL, Zope, Novell, Red Hat, Intel and others.
As previously noted, Oracle’s acquisition of Sun, and with it MySQL, has highlighted the issue of copyright control in encouraging/restricting community development in vendor-led development projects and providing acquirers with the potential to close an open source project.
Clearly, the issue is not as problematic for non-profit foundation-led projects, but the issue of copyright assignment needs more thoughtful assessment than a response that amounts to “non-profit=good, for-profit=bad”.
For more considered analysis of the issue of copyright assignment see:
Dave Neary: Copyright assignment and other barriers to entry
CAOS Theory: On the importance of copyright assignment
Daniel Chalef: OSBC, Community Engagement and Contributor Agreements
Michael Meeks: Some thoughts on Copyright Assignment
Tarus Balog: More on Copyright Assignment
Comments (4) Categories: Licensing,Software
The issue of copyright assignment rears its ugly head again.
Under US copyright law, only “the legal or beneficial owner of an exclusive right is entitled … to institute an action for any infringement of that particular right….” 17 USC 501. So if all you have is a non-exclusive license, or indeed if all you have is joint ownership, you cannot enforce that copyright in court without the other owners joining in.
The blogosphere and conversations among open source lawyers are full of debates about whether copyright assignment allows greater “control” over the work than does a mere license. As far as control by the *licensee*, I’m not sure what that could possibly mean. An appropriate open source license allows a licensee to control his own uses of the copyrighted work just as effectively as if he owned the copyright in the first place. For example, the Apache CLA and Apache License 2.0, which are merely non-exclusive licenses, allow the Apache Software Foundation to make copies, prepare derivative works, and distribute any of the contributions our friends give us. So too the Academic Free License 3.0, the MIT license, and the BSD license, are non-exclusive licenses that allow recipients of the software to exercise control over their uses of the software given to them. Licensees under these licenses can even distribute their copies and derivative works under different licenses, and can enforce their own copyrights on their own collective and derivative works in court even if they don’t own the copyrights on the components.
What greater degree of control does the licensee need? Why should the original copyright owner of the original work want to give away his ownership of his own copyright to help a licensee exercise greater control than he needs?
As for control by the *licensor*, US copyright and contract law give you that. You can enlist lawyers (like SFLC or many other FOSS-friendly attorneys in the US and around the world) to help you enforce your licenses as long as you haven’t assigned your copyright ownership to someone else. And if the time comes that you want to give ownership away to someone who loves your work more than you do, copyright assignment can be done then.
Psychologically and financially, our capitalist society places great value on the ownership of things we needn’t actually own in order to benefit from them. In the FOSS and open standards communities, however, and as our friends at Creative Commons remind us, ownership is not as important as the freedom for everyone to take advantage of our intellectual creations.
/Larry Rosen
Thanks for that insight Larry.
[...] Copyrights and wrongs One of the issues I have with the Free Software approach is that advocates have habit of throwing the baby out with the bathwater when discussing issues that they see as in any way negative to free software. [...]
[...] should be noted, incidentally, that Bradley and Mark have some previous when it comes to copyright assignment, which was also, I believe, caused by confusion rather than [...]