Banter before the meeting

jar: There was the Gene Ontology (GO) project, which was extremely successful. Anyone doing a high-throughput experiment will compare their results to the GO.
... People want to mark up articles with things you wouldn't find with a simple text search. There are about 30 people manually reading articles and annotating them in the database. It's a major investment and they want to extend it beyond genes.
... Such as 'which papers using the following method showed negative results for blood cancers'?
... So there's been an explosion of ont development. GO is the most well know of a consortium of ontologies called OBO. As it grew they realized that they needed to connect to each other and resolve overlaps and combine annotations.
... As they realized they needed a more rational method for making ontologies, that's how OBO Foundry was created..
... So this is industrial strength ontology way beyond what most SW people have used. When they tie them together and their are independent study or method descriptions, they're very interested in provenance.
... Less highbrow is the UMLF, which is a family of ont used in biomed. Many but not all are closed -- you need a license to see them, which is ridiculous, but that's how they were developed.
... UMLF is also huge -- ~100k terms.
... There are external ont efforts that already have ideas about how to do it, so if you want to use RDF you have to have an explanation that makes sense to others.

MEETING CANCELED