I’ve mentioned Semantic MediaWiki several times as having the potential to tremendously increase the value of Wikipedia by unlocking (in the sense of making queryable) all of the data in the encyclopedia.
dbpedia.org has taken a different approach to “Querying Wikipedia like a Database” (their excellent tagline) — extract datasets from Wikipedia, presumably with a manual mapping of relevant categories and data populating infoboxes to triples (described in What have Innsbruck and Leipzig in common? Extracting Semantic from Wiki Content).
I suspect Wikipedia implementation of Semantic MediaWiki would only help dbpedia.org, but the latter is already impressive, requiring no changes at Wikipedia. In addition to making some of the data in Wikipedia queryable they’re exposing non-Wikipedia datasets.
The Semantic Web is so here, now. Doubters repent! ;-) Like I said before:
Once people get hooked on access to a semantic encyclopedia, perhaps they’ll want similar access to the entire web.