Shirky: The Semantic Web, Syllogism, and Worldview
Clay Shirky tears up the Semantic Web as appropriate only for an imaginary world where syllogistic reasoning matters.
Syllogisms sound stilted in part because they traffic in absurd absolutes. Consider this gem from Dodgson:
– No interesting poems are unpopular among people of real taste
– No modern poetry is free from affectation
– All your poems are on the subject of soap-bubbles
– No affected poetry is popular among people of real taste
– No ancient poetry is on the subject of soap-bubblesThis, of course, allows you to conclude that all your poems are bad.
This 5-line syllogism is the best critique of the Semantic Web ever published, as its illustrates the kind of world we would have to live in for this form of reasoning to work, a world where language is merely math done with words. Actual human expression must take into account the ambiguities of the real world, where people, even those with real taste, disagree about what is interesting or affected, and where no poets, even the most uninteresting, write all their poems about soap bubbles.
I agree with him about logic and the real world, but I’d like to hear from, oh, James Stewart if he’s reading, who is a fan of RDF and suchlike semantic things, to see what he thinks of Shirky’s overall critique.
You there James? :)
I am indeed here :)
I like a lot of what Clay Shirky has to say, but he seems to have a tendency to overstatement. That comes through very clearly in that particular essay. I’d generally agree with him on logic too, but (noting that essay is a couple of years old) I think he leaves himself open to some pretty simple rebuttals about the opportunities offered by a web richer in metadata.
Appreciation of poetry is never going to be something we can do with SemWeb reasoning, but we could use triples to state that Wilfred Owen is the author of “Dulce et decorum est.” We can state that my house has particular geographical co-ordinates relative to some agreed reference scheme, and that the nearest bus stop is at the intersection of Fulton St. and Fuller Ave. And if that information is specified in some sensible serialization then we can build agents that can perform much smarter queries than current search engines.
We don’t need a complete data set, we just need a few agreed reference schemes and some way of assembling those snippets of information that we can assert. URIs work pretty well as identifiers, though a vocabulary like FOAF (which I like for sharing biographical information and group membership, but which wisely steers clear of trying to define interpersonal relationships) also uses hashed email addresses (considerably more unique than Shirky’s human name example, though also sometimes needing disambiguation).
Authority and reliability remain issues in the way of the Semantic Web vision. If we can’t figure out ways to resolve conflicts or decide who counts as authoritative then it becomes dangerous to just suck up any metadata we find out there. But that’s an issue for all sorts of projects well beyond the semantic web, including Wikipedia which I seem to remember Shirky being rather more fond of, and isn’t necessarily a showstopper for more specific/limited implementations of those technologies. See for example the use by Opera of SPARQL for querying a lot of their online resources:
http://www.oreillynet.com/pub/wlg/8609