Argh intellectual purity versus political necessity. I fear that Bismarck trumps Aristotle.
Stephen Stead Tel +44 20 8668 3075 Mob +44 7802 755 013 E-mail [email protected] LinkedIn Profile http://uk.linkedin.com/in/steads -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Vladimir Alexiev Sent: 11 June 2013 10:11 To: [email protected] Cc: 'FRBR Group'; [email protected] Subject: Re: [Crm-sig] FRBRoo classes equivalent to CRM classes > The reason is simple: RDFS does not contain owl:same_as. > We (you) can make an owl version that solves the problem. owl:sameAs is a different construct from owl:equivalentClass. You don't have to use owl:equivalentClass though. You can stay in the rdfs vocabulary by using rdfs:subClassOf in both directions, as I wrote previously. See http://www.w3.org/TR/owl-ref/#equivalentClass-def : If we wanted to "upgrade" an axiom of the form "A subClassOf B" to "A equivalentClass B" (meaning that the class extension of A is not just any subset, but in fact the same set as the class extension of B), we could add a second subClassOf axiom of the form (B subClassOf A), which by definition makes the two class extensions equivalent (and thus has the same meaning as "A equivalentClass B"). Such subClassOf "cycles" are explicitly allowed. > The naming is more political. Some class names should look more FRBRish. The spirit of the sem web is to reuse existing ontologies as much as possible. CRM doesn't do that, e.g. it doesn't relate to previously defined common classes for places, times, events, concepts, etc. But I can live with that since CRM is a self-contained intellectual effort, and such semantic relations (subClass and equivalentClass) have wide-ranging consequences that are very nontrivial to analyze. AFAIK, Simon Reinhardt is the first to attempt this, see http://lists.ics.forth.gr/pipermail/crm-sig/2009-September/001393.html However, FRBRoo is explicitly an extension of CRM. For FRBRoo not to "clean up" its classes, and not to use CRM classes directly where appropriate, seems like intellectual sloppiness to me. And disregard for the practical implementers of these ontologies. Steve> this is a joint project between 2 communities IFLA and CRM SIG. Yes, and these groups work closely together, have identified the equivalent classes, and I see no impediment to merge the classes now. > Why you materialize all superclass links is not clear to me. No one > queries "all persistent items" or so. Not all triple stores do that. Could you give an example of *any* CRM system that does *any* useful reasoning but doesn't materialize the inferences mandated by CRM? Note: CLAROS doesn't have any reasoning, as you can check at http://data.clarosnet.org/sparql/ by comparing this query SELECT * WHERE { ?s a crm:E53_Place; crm:P87_is_identified_by ?o } LIMIT 50 to this one SELECT * WHERE { ?s a crm:E53_Place; crm:P1_is_identified_by ?o } LIMIT 50 _______________________________________________ Crm-sig mailing list [email protected] http://lists.ics.forth.gr/mailman/listinfo/crm-sig
