rdf technology is coming on strong in late 2011; the tools are getting better fast -- we will be able to represent the social graph in more detail, and correlate it with the 'semantic graph' that shared human experience is coded in.
yes, RDF has taken a while to mature, but look at the time lag between Codd's paper and the commercial release of Oracle. 1000 flowers have been blooming for years and we know a lot about what works and what doesn't.
the trouble with foaf and other 'distributed social graph' is that muggles like Facebook the way it is... they don't care about data portability or privacy; unless a distributed system can provide a better user experience, it's got no hope.
I believe Maciej is saying that it's not just an issue with RDF or a technical problem really. He says: "Personally, I think finding an adequate data model for the totality of interpersonal connections is an AI-hard problem. But even if you disagree, it's clear that a plain old graph is not going to cut it."
That depends on your definition of a "plain old graph". If you allow arbitrary numbers of edges between nodes and allow edges to carry data, everything will be ok, right? (...right?)
In math... and we're talking math if we're talking RDF & graphs, right?... "arbitrary" is perilously close to a synonym for "infinite" and "unbounded".
That's true, but "infinite" and "unbounded" often also mean "there's no exact or precise solution, but at least for certain circumstances there's a useful approximation that works".
I can calculate pi to "enough" precision, whether I'm making a hula hoop, or a piston ring, or sending a probe to Saturn - I suspect a lot of what I'd like out of a functional "social graph" probably doesn't need much more precision than "three and a bit", 3.1416+-0.0004, or 3.14159265+-0.000000005.
rdf technology is coming on strong in late 2011; the tools are getting better fast
Can you expand on this? I'm very interested in RDF but have been frustrated by how little interest in Semantic Web tech there seems to be in the US. I'm interested in highly specified networks rather than slippery ones from the consumer marketplace, and in visualization/manipulation tools rather than the underlying DB technology. If you have any suggestions I'd be grateful, I feel like I've been plowing a lonely row with this stuff.
Many things can be usefully represented by graph structure, and RDF is a way to do that (a verbose and clunky way, but better than no standard at all).
Whether social relationships are one of those things is a separate question.
new entry into the market, Stardog from Clark and Parsia, extremely fast and embeddable / scalable RDF database with OWL reasoning built in (they created the Pellet OWL reasoner)
I agree - 2011 going into 2012 is the rise of the semantic web
yes, RDF has taken a while to mature, but look at the time lag between Codd's paper and the commercial release of Oracle. 1000 flowers have been blooming for years and we know a lot about what works and what doesn't.
the trouble with foaf and other 'distributed social graph' is that muggles like Facebook the way it is... they don't care about data portability or privacy; unless a distributed system can provide a better user experience, it's got no hope.