- From: Frank Manola <fmanola@mitre.org>
- Date: Wed, 25 Apr 2001 11:43:15 -0400
- CC: w3c-rdfcore-wg@w3.org
This is interesting. However, when I wrote: > how much do existing implementations use reification, and > exactly how do they use it? I meant to refer to implementations that *use* RDF reification, rather than implementations that *implement* RDF reification. In other words, how do people who use RDF for creating metadata for Web pages, site descriptions, etc. actually use reification (as opposed to how to RDF tools handle reification). NB: A particular issue that I'm interested in here is exactly how it is intended that reification be used to make a comment about some RDF statement that is located elsewhere (at someone else's site). For instance, Web site foo contains (using the "standard example") "Ora Lassila is the creator of http://www.w3.org/Home/Lassila", and Web site bar wants to say "that statement was made by Ralph Swick". I can use RDF reification as defined to reify the original statement at Web site bar, but I've lost all the context in which the original statement was made. I can certainly duplicate that context (like that the statement was made at site foo) by adding additional information at site bar. I'd certainly want to do that if, e.g., I wanted to point out conflicting statements that Ralph may have made at different times or locations. But the amount of context I might have to duplicate at the site where I was making the statement could get arbitrarily large. What I'd like to do is just point to the original statement where it "lives" (that seems like the "Web" way to do things, after all). Some of these approaches that involve imagining that every triple has an additional "column" containing an identifier seem to effectively be steps in this direction. --Frank jos.deroo.jd@belgium.agfa.com wrote: > > Jena and SiRPAC seem to be glad with > > <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" > xmlns:rdfc="http://www.w3.org/2001/rdf-core#" > xmlns="file:/daml/reification/n3.n3#"> > > <rdf:Description> > <rdfc:is rdf:parseType="Quote"> > <rdf:Description rdf:about="#subject"> > <predicate1 rdf:resource="#object1"/> > </rdf:Description> > </rdfc:is> > <predicate2 rdf:resource="#object2"/> > </rdf:Description> > <rdf:Description> > <rdfc:is rdf:parseType="Quote"> > <rdf:Description rdf:about="#subject"> > <predicate3 rdf:resource="#object3"/> > </rdf:Description> > </rdfc:is> > <predicate2 rdf:resource="#object2"/> > </rdf:Description> > <rdf:Description rdf:about="#subject"> > <predicate1 rdf:resource="#object1"/> > </rdf:Description> > </rdf:RDF> > > and an rdf:parseType="Quote" aware processor could build > the appropriate contexts. Isn't that nice? > -- Frank Manola The MITRE Corporation 202 Burlington Road, MS A345 Bedford, MA 01730-1420 mailto:fmanola@mitre.org voice: 781-271-8147 FAX: 781-271-8752
Received on Wednesday, 25 April 2001 12:22:17 UTC