This is a good introduction to the options. I’d only make two additions.
- Authors needn’t retain full copyright in order to provide OA to their own work. They only need to retain the right of OA archiving—which, BTW, about 70% of journals already give to authors in the copyright transfer agreement.
- Charles mentions the author addenda from SPARC and Science Commons, but there’s also one from MIT.
Peter is right on both points; however, my document has a broader rights retention focus than providing OA to scholars’ work, although that is an important aspect of it.
For example, there is a difference between simply making an article available on the Internet and making it available under a Creative Commons Attribution-NonCommercial 2.5 License. The former allows the user to freely read, download, and print the article for personal use. The latter allows user to make any noncommercial use of the article without permission as long as proper attribution is made, including creating derivative works. So professor X could print professor Y’s article and distribute in class without permission and without worrying about fair use considerations. (Peter, of course, understands these distinctions, and he is just trying to make sure that authors understand that they don’t have to do anything but sign agreements that grant them appropriate self-archiving rights in order to provide OA access to their articles.)
I considered the MIT addenda, but thought it might be too institution-specific. On closer reading, it could be used without alteration.