Why Share?
Fun and easy to demonize publishers and drug companies, but we're all in this together.
The "tragedy of the commons":
sheep grazing on common pasture:
advantage - more sheep for each shepherd to shear
disadvantage - more grass eaten, eventually no grass - a justification for fences
The enclosure movement - lawyers believe a second enclosure movement is ongoing.
Old models of rivalrous-based publishing are being applied to non-rivalrous models (digital networks).
Current copyright law is created because of entertainment lobby. Scientific papers by Einstein are not in the public domain. The public domain keeps moving to protect popular entertainment - uses Mickey as example.
Creative Commons - 5 versions from most restrictive to most permissive. Language is readable by lawyers, humans, and machines.
140M works under CC licenses w/ exponential growth. Why? He believes humans want to share.
Science Commons tries to put together science law and economics to create licenses that make sense for science research. Asks, "where are imbalances in scientific research cycle?"
Policies for author self-archiving - some publishers make it easy; many do not.
Incentives to share research - many authors don't see or don't have
Science Commons is releasing a Scholar's Copyright to easily generate an author's addendum that specify various kinds of author's rights. Librarians, take note and link soon.
Discusses difficulties in transferring and sharing scientific materials - DNA, cell lines, bones, rocks, etc - this is a culture problem. Scientists are rewarded for hanging on to their materials and sharing only rarely - certainly not with little-known labs in foreign countries.
Science Commons is working on changing this culture by creating a clearinghouse of scientific materials and standardizing licenses to cover exchanges.
Impossible to process all of information that we already know (uses research on p53 protein as an example) under current usage models - print, reading, article-based.
Text-mining works better, but will work even better with semantic web implementation - author-provided keywords and associated hyperlinks. The goal is a model of reading/processing that works as well as Amazon.com - they are really good at processing tons of data, knowing what you like to read, and making recommendations to you. Why can't scientific research be this easy?
A problem with many major publishers - like Elsevier - licenses forbid crawling, spiders, robots - so enabling this kind of Semantic Web work over a current subscriber's web pages would be illegal under the DMCA.
http://sciencecommons.org
willbanks@sciencecommons.org
[edited for spelling and URLs 10/12/06]