Many thanks to Fred Giasson, Niklas Lindström, and others for their kind advice in developing the university ontology.

SPARQL me this:


PREFIX dbpedia2: <http://dbpedia.org/property/>
SELECT ?drink
WHERE { ?drink dbpedia2:bourbon "yes"@en}


(For the non-semantic webby, this is a basic semweb query for asking DBpedia for info in wikipedia about drinks with bourbon)

Alexandre Passant has produced a really nifty exporter of SIOC and FOAF data from one's flickr account. Type in your flickr username, and it grabs the publicly available profile info (username, groups, etc.).

Two of the strengths of RDF are the ease with which one can express relationships between things and the ease with which one can mix together different sources of data. Throw into that mix the power of a tool like SIMILE's Solvent to scrape data off different web pages, and we have all we need to take data that people didn't think were (or should be) interrelated and make some more useful relationships. How to represent the relationships? Make use of another of SIMILE's excellent tools: Exhibit.

Coming up all too soon, Jim Groom and I will be giving a talk on social tagging and the semantic web, and that's been my prod into playing with using the two together, and to finally get around to learning to use the SPARQL query language for the semantic web.

Stumbled across The 325 Factlog, an effort to "document the year 325". Read that again, soak it in: "document the year 325." As in, here's everything we know about the people, places, and events if you are interested in the year 325. Okay, it isn't there yet, and it's really concerned with Rome and the West, but it's still a massive undertaking with an enormous amount of information.

Fantastic and exciting developments in one of my favorite worlds, that of rdf and the semantic web! I haven't had a chance to surf around and read material for a while, so there's too much for one blog post and I'll get to them all eventually. The big ones to know about are version 3 of Piggybank, the SIOC ontology (where was I that I missed this coming up?!?), and Talkdigger.

Talkdigger makes a nice move in describing itself as "Semantic Web Ready." It's an apt description, I think, in pointing up the new ways of envisioning the relationships between content on the web that the semantic web will bring while tacitly acknowledging that it's not in the air yet. But with Talkdigger, we're getting to liftoff velocity.

Talkdigger lets users pick out a conversation taking place on the web by focusing on a URL that's a node in that conversation. It then searches around, both in a general web search and within other conversations already in the system, for other nodes in the conversation and starts putting together the relationships between them. In that way it is similar to co-comment (I confess I haven't used co-comment, though, so I hope someone who has will help make the comparison). It uses the SIOC (Semantically Interlinked Online Communities) ontology to produce the connections. The result is a way to make connections between online conversations and the people producing them.