Raw

Andrew Newman links to a paper called Implementing OWL Lite in rule-based systems and recursion-enabled relational DBs, also referring to two other papers : Description Logic Programs: Combining Logic Programs with Description Logic and Bubo - Implementing OWL in rule-based systems.

By a smallish-world coincidence I read the latter two just last night. I'll read the first tonight - thanks!

The idea in the ones I read is quite nifty. If you map what you can of logic programming (Prolog etc) and description logics (RDFS, OWL etc) to first-order logic, there's an overlap. The overlap is a stack of inference techniques that work in both LP and DLs. Which means e.g. you can use Prolog to work with RDF ( yes, you can) Interesting to see how the theory works (though I still need to look up Herbrand).

The LP rules can mostly be translated directly into relational database queries, so the papers also point towards how DL stuff (like RDF/OWL) may be represented in RDBMS.

See also : KAON