Open smrgeoinfo opened 3 years ago
You could do it this way:
thors:component rdfs:subPropertyOf [ owl:inverseOf skos:inScheme ] .
However, component
comes from ISO 19108, and is intended to point to the top members of the hierarchy. I think the original axiom is consistent with that. But it does not also say that nested eras are not allowed. Not sure if that can be expressed.
currently have thors:component rdfs:subPropertyOf skos:hasTopConcept ; and thors:referencePoint rdfs:subPropertyOf skos:hasTopConcept ;
I think I see the logic-- skos:hasTopConcept is the only skos property with domain skos:ConceptScheme and range skos:Concept, but semantically this doesn't make sense. Every thors:Era that is part of a thors:ReferenceSystem is not a top concept in that scheme.
we need a thors:isComponentOf relation (inverse of thors:component) from thors:Era to thors:ReferenceSystem to indicate the time scale instances a given era is a member of. Also thors:isReferencePointFor inverse of thors:referencePoint. These would be subProperties of skos:inScheme.