Wikipedia gets a little more semantic
A pretty simple idea, but these guys did all the hard work of implementing it: dbpedia. While the idea of actual semantic data within Wikipedia is probably a ways off (must stop linking to my own sites!), there are certainly templates widely used within Wikipedia that could be extracted, converted into semantic data externally, and then queried like a database. Which is just what these German researchers did. What’s a template? It’s the code that creates an info table to the side of an article, like the one for, say, “Dots and Loops” (picking an awesome album at random). Since the fields are standardized, you can turn them into semantic data on your site in a straightforward way.
The dbpedia project doesn’t have all the template information from Wikipedia, but it already has quite a lot, especially on subjects like entertainment, geography and nature. I created a query to get a list (including image) of all extinct birds whose name contains the letter “d”.
It’s kind of a pointless list, though on the other hand this is, as far as I know, the first time in human history that someone could create such a list without doing any installing, computing, or any research on the actual subject matter.
Hopefully this will also convince people to fill in more information in Wikipedia’s templates (up until now, there hasn’t been a really compelling reason to do so), to ensure comprehensiveness.
Que sera, as they say, but I think this is the shape of things to come.
(Via Mike Linksvayer)
March 31st, 2007 at 12:30 pm
Larry Sanger’s Citizendium will, I think, make good use of this tech but yes, really deserves credit.