- From: Frank Manola <fmanola@acm.org>
- Date: Sat, 5 Jan 2008 15:55:00 -0500
- To: Frank Manola <fmanola@acm.org>
- Cc: Garret Wilson <garret@globalmentor.com>, SWIG <semantic-web@w3.org>
Garret-- On further thought, I think my comments about the "triple" relation may have gotten us off on a tangent. The "triple" relation is a notation intended to convey certain relational semantics, namely those of binary relations (according to the translation rule I cited from RDF Semantics), but including the predicate name in the triple, rather than "outside": e.g., in RDF you imagine a relation that looks like triple(#john ex:name "Smith") rather than ex:name(#john, "Smith") the latter being the "purer" relational (or logic) version. I don't see that anything terribly complicated is going on here. In the actual RDF, all you have in hand is #john ex:name "Smith" (a triple), and the RDF semantics talks about the "triple" relation as being "notional". In a graph of triples, there are as many relations as far as *semantics* are concerned as there are distinct predicates included in those triples (not one per resource, as you had it earlier). Does this help? --Frank
Received on Saturday, 5 January 2008 20:55:34 UTC