Rename era:hasAbstraction object property

[DMPatru]

According to the relations between entities, I don't find era:hasAbstraction as a proper name for the object property with range era:InfrastructureObject and domain era:NetElement. As it is the inverse of era:hasImplementation, my proposal would be to rename it to rename it to era:isImplementationOf, or era:hasTopology, depending on which point of view the network is seen:

• if the network elements are described beginning with their topological representation, then I would choose era:isImplementationOf to say "era:InfrastructureObject era:isImplementationOf era:NetElement"
• but if the technical representation is more important and therefore the implementation layer "comes first", then the topology can be seen as extra information, meaning that the triples of form "era:InfrastructureObject era:hasTopology era:NetElement" make much more sense.

[Airy59]

Currently, era:hasAbstraction has domain era:InfrastructureObject and range era:NetElement (not the other way round). The problem is not so much the name but the class that is used as a range: in the overwhelming majority of cases, infra elements are "abstracted" or "matched", NOT by a full net element, but but a piece of a net element or a combination of net element pieces. These are "Locations" (spot, linear, area...).

In SemanticRSM, or RSM, or RTM, element parts are only there (if needed) to provide correspondence between network levels. Since it appears that RINF will modelled at MICRO level, maybe RINF can eschew these, and only use XxxLocation classes.

In another RINF version shown today, the range is the class TopologicalObject, rather than NetElement, which is better. TopologicalObject being a superclass of NetElement, this leaves room for using a NetElement is lieu of a "TopologicalObject" but, as said before, 1) this will nearly never happen; 2) it conflates the meaning of location (referring to topology) and topology itself.

I'd like to recommend the stricter separation Topology <=> Localisation <=> infra elements. You will spare one class (TopologicalObject) that will only have, I am afraid, anonymous individuals (instances) just to point to location. Better point to locations directly.

In such case, the property "hasAbstraction" and its inverse would disappear, being replaced by the explicit "locatedAt"(domain: Infra object, range: BaseLocation, maybe to be renamed Location, partitioned into disjoint classes Spot-, Linear- and AreaLocation) and an inverse property that you could forget about because it is not going to be useful ("locates" would be an adequate name).

[gatemezing]

New updates of the model diagram here https://linkedvocabs.org/data/era-ontology/3.1.0/doc/resources/era-ontology-3.1.0.jpg. You can also see the Turtle file in the repository.