poe icon indicating copy to clipboard operation
poe copied to clipboard

odrl:partOf / dcterms:hasPart

Open joshcornejo opened this issue 6 months ago • 2 comments

I think it misses or should have owl:equivalentProperty

Current definition:

#odrl
:partOf
	a rdf:Property , owl:ObjectProperty, skos:Concept ;
	rdfs:isDefinedBy odrl: ;
	rdfs:label "Part Of"@en ;
	skos:definition "Identifies an Asset/PartyCollection that the Asset/Party is a member of."@en ;
	rdfs:domain [
		a owl:Class ;
		owl:unionOf ( :Asset :Party ) ;
	] ;
	rdfs:range [
		a owl:Class ;
		owl:unionOf ( :AssetCollection :PartyCollection ) ;
	] .

Proposed change:

:partOf
	a rdf:Property , owl:ObjectProperty, skos:Concept ;
	owl:equivalentProperty dct:hasPart ;                        # <-----
	rdfs:isDefinedBy odrl: ;
	rdfs:label "Part Of"@en ;
	skos:definition "Identifies an Asset/PartyCollection that the Asset/Party is a member of."@en ;
	rdfs:domain [
		a owl:Class ;
		owl:unionOf ( :Asset :Party ) ;
	] ;
	rdfs:range [
		a owl:Class ;
		owl:unionOf ( :AssetCollection :PartyCollection ) ;
	] 

dcterms:hasPart

That would connect seamlessly the different properties in DCAT/DPROD and ODRL for traversal.

Too early (in the day) to know if rdfs:subPropertyOf would work better.

joshcornejo avatar Jul 08 '25 06:07 joshcornejo

It is probably not equivalent due to the different definitions: DC: "A related resource that is included either physically or logically in the described resource." ODRL: "Identifies an Asset/PartyCollection that the Asset/Party is a member of"

subPropertyOf would suffice.

But is it an "errata" ?

riannella avatar Aug 01 '25 01:08 riannella

OWL, RDF ... Inclusion ... member ... perhaps too finicky.

IMHO, every "member" (of a Collection) is "included" physically or logically in the described resource.

The property rdfs:subPropertyOf is an instance of rdf:Property that is used to state that all resources related by one property are also related by another.

"related" ... inclusion ... member ... it is just ether.

Either way, to work seamlessly with the likes of DCAT's "included references that are also members" for refinements, an easy mechanism to infer relationships should be available - otherwise there will be duplication and unnecessary overheads.

joshcornejo avatar Aug 01 '25 06:08 joshcornejo