Re: is FRBR relevant?

On Thu, Aug 12, 2010 at 12:33:23PM -0400, Young,Jeff (OR) wrote:
> ex:nomen1 a frsad:Nomen ; # could be inferred from skosxl:Label
> 	a skosxl:Label ;
> 	frsad:soundLabel ex:genericresource1 ; # content-negotiable for
> audio/* media-types
> 	skosxl:literalForm "Fire Alarm" .
> 
> frsad:soundLabel a owl:ObjectProperty ;
> 	rdfs:domain frsad:Nomen ;
> 	rdfs:range owl:Thing .
> 
> Should we start hoping that SKOS B.2.4 can be relaxed so frsad:Nomen
> doesn't become a specialized niche?

Hang on... - I just made up the hypothetical "ex:soundLabel"
for the purpose of the example.  Are you saying that there
already exists a "frsad:soundLabel" (along with an frsad:Nomen)
in an already existing frsad: namespace?

Also, I thought frsad:Nomen was analogous to a SKOS label,
not a concept (i.e., the range of frsad:soundLabel, not
the domain).  Do I have it backwards?

To be clear, I was picturing:

    [instance of Thema]          ex:soundLabel       [instance of Nomen]
    [instance of Nomen]          ex:soundForm        (serialization of sound)

analogously to:

    [instance of skos:Concept]   skosxl:prefLabel    [instance of skosxl:Label]
    [instance of skosxl:Label]   skosxl:literalForm  (literal)

Tom

> > Skos:prefLabel is a sub-property of rdfs:label, and the
> > rdfs:range of rdfs:label is rdfs:Literal [2] -- but that only
> > applies to the label properties, not to the skosxl:Label
> > class itself.  I don't see any obvious arguments against
> > coining a convention to the effect that the property chain
> > "ex:soundLabel, ex:soundForm" expresses the "sonic label"
> > of a SKOS concept, with skosxl:Label as the rdfs:range of
> > ex:soundLabel. Or something to that effect...

-- 
Thomas Baker <tbaker@tbaker.de>

Received on Thursday, 12 August 2010 18:41:11 UTC