Adrian Pickering, Christopher Gutteridge, and David De Roure have self-archived "A Networked Registration Scheme to Support Open Science" in the ECS EPrints Repository.
Here's an excerpt:
The Open Source and Open Science movements have demonstrated the success of distributed collaborative experimentation and intellectual property (IP) development. While those contributing to the effort may do so without seeking to secure IP rights, it is clear that credit and attribution are crucial to the scholarly lifecycle because they underpin reputation—when IP is created it is only fair that 'credit is given where credit is due'. We propose that there need to be systems in place, independent of the project, where the evidence of 'prior art' can be registered. The authors' thesis is that simply having such a system available will ensure proper behaviour between collaborators and foster higher productivity.
Repositories such as EPrints and myExperiment, which focus respectively on publications and digital 'research objects', can readily use such a system—the intellectual assets stored digitally in the repository can be registered by their owners. To achieve this with the necessary guarantees we need an appropriate registration scheme and architecture.