On 1/6/15 4:41 AM, Gordon Dunsire wrote:
> I therefore think it is a benefit that the
> BIBFRAME Initiative (BFI), IFLA, and the JSC for RDA are providing
> constrained RDF element sets for BF, FRBR, ISBD, and RDA. I also think the
> provision of unconstrained element sets is a good thing, together with
> mappings from constrained to unconstrained properties.
Gordon, I heartily recommend that we not use the term "constrained" for
things that are typed in the RDF class sense. There is no constraining
taking place. I realize that the first edition of OWL used the term
"constraint" but the developers have realized that was a mis-nomer and
now use the term "axiom". RDF provides axioms for inferencing but does
not constrain any statements or graphs. You can create "bad data" all
you want, and RDF/OWL allows you to determine if data forms a graph
consistent with an ontology.
Because there is no constraining in the open world, W3C is now working
on closed world rules for data validation. Those rules will constrain,
but only in a closed world.
RDA and BIBFRAME properties are typed, not constrained.
[log in to unmask] http://kcoyle.net