- From: pat hayes <phayes@ai.uwf.edu>
- Date: Mon, 5 Feb 2001 23:05:19 -0600
- To: Peter Crowther <Peter.Crowther@melandra.com>
- Cc: www-rdf-logic@w3.org
> > From: Dan Connolly [mailto:connolly@w3.org] > > Sent: 02 February 2001 23:18 >[...] > > Triples are an idiom that show up all over the place, > > in my experience. They look like a pretty important > > and useful modelling primitive. > >You can model a directed graph using a set of triples; you can model an >arbitrarily complex data structure with a directed graph. As primitives, >they are sufficient to model any other structure. That depends on what you mean by 'model' (and 'structure'). They are sufficient to *implement* any finite directed graph, indeed, in common with a host of alternative datastructuring primitives (eg LISP s-expressions based on dotted pairs). They are not so suitable for efficiently representing numerically indexed structures such as arrays and tables, and they are only suitable for representing linguistic structures such as structured code or logical expressions if supplemented with some conventions for binding label names and indicating scope boundaries. (All of this is well-known, almost classical, in both the logic and programming communities and has been for about 40 years or more: I learned it at grad school and it was considered ancient lore way back then.) Triplet structures have their uses, but the almost irrational passion revealed for them by members of the RDF community remind me of what happened when I once questioned Christian doctrine on a fundamentalist email list. Pat Hayes --------------------------------------------------------------------- IHMC (850)434 8903 home 40 South Alcaniz St. (850)202 4416 office Pensacola, FL 32501 (850)202 4440 fax phayes@ai.uwf.edu http://www.coginst.uwf.edu/~phayes
Received on Tuesday, 6 February 2001 00:02:37 UTC