Having just posted that there didn’t seem to be too much of this we have a talk in the Social Sciences parallel session that covers exactly this.Pete Edwards talked (amongst other things) about ourSpaces, a tool providing a resource for sharing resources that can be tagged in all the expected Facebook style ways. He then went on to talk about how you record both data and how it is recorded i.e. methodology. This is obviously critical for making decisions about policy. There are obvious parallels here with the issues for metadata capture for laboratory experiments.
How do they help people populate the ontology? They have looked at connecting social tags to an ontology. Each community has its own set of social tags within the framework of the ontology. Developed a metadata interface. Web service for uploading data sets using a natural language interface. A text description of the service is clicaable with pop-up menus. The pop ups are driven by the ontology so assisting the user to use the correct terms. These then generate RDF triples. Tags can then be added in an arbitrary fashion using existing tags or new ones. These tags act as further information within the leaves of the ontology.
I asked the quetsion as to whether there are problems with persuading people to provide data to the upload system . The answer was that this was an issue and drove the generation of ourSpaces, which encourages people to start sharing. Thus Web2.0 resources actively encourage people to share more.