Re: Disjuction of subPropertyOf chain? was Re: possible semantic bugs concerning domain and range

I wrote:
>
> In the presence of an rdfs:subPropertyOf chain, the interpretation of the
> literal would, I presume, be the disjunction of the conjunctions of the
> datatypes indicated by each of the ranges for each rdfs:subPropertyOf i.e.
>
> foo rdfs:range (Range(foo) union Range(bar)) ...
>
> foo rdfs:range _:x
> _:x owl:unionOf (xsd:[integer union string], xsd:[string union integer])
>
No, that can't be correct... it would have to be something like:

john foo "10" .
=>

john foo _:x

_:x rdf:type xsd:[integer union string]
_:x rdf:type xsd:[string union integer]

(or rewrite that to conform to the RDF datatypes version de jure , but
hopefully you know what I mean :-))

Jonathan