Since the beginning we’ve provided our licenses for three separate, distinct audiences: humans, lawyers and machines. The machine audience has been served by metadata versions of the licenses. This metadata is encoded with the HTML you get from the license chooser, as well as for each individual license. For example, you can find the metadata for the Attribution 3.0 Unported license by appending /rdf to the URL.
While we’ve always provided this machine readable version of the licenses, we’re finally taking steps to begin eating our own semantic web dogfood. We’re doing several things to accomplish this. First, the RDF/XML we serve for each license is now considerably more informative. It includes:
- an explicit pointer to the license legalcode
- information on when the license was deprecated (for example, the Developing Nations 2.0 metadata)
- information about what license replaces this one (for example, the Attribution 1.0 Generic metadata)
- an explicit assertion about the license’s jurisidiction; this was previously encoded only by convention
We’re also starting to use this metadata to power our own applications. The OpenOffice.org Addin ships with a copy of the RDF and uses SPARQL to determine the license you’ve selected. As we continue to build out the tools around CC licenses we’ll be moving in a similar direction, looking for ways we can leverage this resource we already have.