The standard says "F9 Place Equal to: E53 Place"
The RDFS file [2] says "F9_Place subClassOf  E53_Place".
These are not the same thing.

You could state they are EquivalentClasses (or add subClassOf  in the opposite 
direction, which is the same thing).
However, I would advise against such proliferation of classes.

Already in ResearchSpace we observe that fully 40% of all triples come from 
rdf:type statements, due to CRM's deeply nested class hierarchy.
Out of 1.5B triples that's some 600M, which is a lot and has significant 
consequences. I hope to say a lot more about this at CRMEX [1]

Can you give an example of a FRBR F9_Place that is *not* a CRM E53_Place ?
Following Occam's Razor principle, classes shouldn’t be created without proven 
need.

EquivalentClasses should be used when two classes are created by independent 
authorities, have been used for years, there are data using these classes, and 
only later it’s discovered they represent the same entity.
But I don't think that's the case here.

I would strongly recommend that the FRBRoo ontology simply imports CRM and uses 
CRM classes, without introducing CRM-equivalent classes.

(In fact, I would recommend a similar thing for the FRBRoo standard: just refer 
to CRM classes and if needed amend their scope notes in CRM.
Don’t repeat in FRBRoo definitions of CRM-equivalent classes. This will cut 
down the volume of the standard significantly)

Cheers! V

--

[1] http://www.ontotext.com/CRMEX
Workshop "Practical Experiences with CIDOC CRM and its Extensions", 26 Sep 2013 
in Valetta, Malta 
Please submit!

[2] www.cidoc-crm.org/rdfs/FRBR2.0-draft-B.rdfs


Reply via email to