GeologicFeatureRelation in OWL

[smrgeoinfo]

v4 uses the abstract class pattern to decouple the basic and extended namespace in XML.
For OWL implementation, looking at v3.2 it looks to me like implementation by shapeChange from the UML is tricky. relationshipType is essentially a soft type, but GeologicRelation subclass tree has several hard types in the model with constraints on the source and target classes, but those association ends are not named in the UML. The geologicRelation base class has a sourceRoleTerm and targetRoleTerm, but I don't think that's what we want. The draft OWL implementation (2_after_manual_edit/gsmlb.ttl), defines relatedFeatureSource and relatedFeatureTarget object properties; this is what we need but its not in the UML. The range of these would be restricted to the appropriate classes for the hard type subtypes, and other relationshipTypes would be subclasses of the appropriate UML hard type