Recently we spent 90 minutes putting up a Pelorus app to enable faceted browsing of the data.gov dataset catalog.
That's notable not only because it took so little time, but also because of the long list of things that we weren’t required to do in order to make it happen:
- We didn’t have any meetings with data.gov people; we know some of them—Hi George, Hi Jeanne!— but that doesn’t matter here.
- We didn’t have any meetings with any of the downstream government data providers.
- We didn’t have any meetings with the RPI people who did much of the data conversion work to RDF; we also know many of them—Hi Jim, Peter, Deb, Jiao, Jie, etc—but, again, shrug...
- We didn’t have any meetings with the Virtuoso people (Yo, Kingsley!) who’ve put up a SPARQL endpoint for much of the data, etc.
Not only did we not have any of those meetings, we didn’t have to exchange emails, design docs, APIs, EAV diagrams, software bundles, data dictionaries, blah, blah, with any of those people or organizations.
In other words, the whole affair from our perspective was entirely uncoordinated, which is another way (to a first approximation) of saying that it was cheap and easy.
Now we’ve put up an enrichment of the datagov.clarkparsia.com app that relates government datasets to books that are about the same subject matter (via Library of Congress Subject Headings). So when you’re looking for a data.gov dataset by faceted browsing, you can also see published books that are related by subject matter. Pretty cool: contextualized knowledge for the win!
Some examples chosen at random:
- Books about the Darfur conflict related to Department of State dataset about Darfur
- Books about Global Warming related to a NASA global warming dataset
- Books about bank lending laws related to a Federal Reserve dataset
And, again, we did this rather quickly—it took about 2.5 hours this time. But, again, we didn’t have any meetings with Library of Congress people, Google Books people, or anyone else.
Painless, efficient, and mostly meeting-free information integration is another reason the Semantic Web kicks ass.
UPDATE: We fixed an irritating bug that was preventing the bookmark links from working correctly; and we also switched from Google Books to WorldCat for related books. Both work much better now.
