issue-72: language for ssn core module, focused on inverseOf in particular

Here is a short summary of this issue as raised and discussed but unresolved in the ssn  meeting  today. There may have been more  arguments that  I missed from this summary.
Please comment!!!!!

inverseOf is an OWL property that, in some circumstances requires reasoning to use effectively. Therefore it is not clear that it should be used in any ssn core module, that is intended for "should stand on its own as a lightweight vocabulary, but also be common core"[1]

Why inverseOf for some properties is needed:

P1: Web devs intuitively understand and expect it to be there
P2: Useful for "backlinks" where parts of the instance data is defined in different  repositories
P3: Triplestores can specify the necessary entailments if they want to
P4: can help web devs formulate the right queries
P5: Aidan Hogan's comment here [4]
P6: off-list discussion here [5]
P7: allows data to be compressed

Why inverseOf should not be  in the core:

c3: SPARQL does not need it - can just reverse the args of the property in question
c4: Requires an owl reasoner (beyond RDFS)  to be used in order to behave as users would expect. Users may not be aware whether or not a reasoner is being used.
c5: If defined in vocab and used with SPARQL but without a reasoner then queries are complex
c6: Is an owl:  feature and that means  is not lightweight
c7: Is not even in OWL-RL (Correction. It is not in OWL-EL but is in OWL-RL , OWL-QL and OWL-DL[2]  - Kerry)
c8: could be avoided by documenting what the properties would be called if they *were* in the core, but not putting them there (see prov-o [3] for precedent)
c9: could be avoided by including at higher vertical layers instead


[1] https://www.w3.org/2016/08/09-sdwssn-minutes
[2] https://www.w3.org/TR/2012/REC-owl2-profiles-20121211/
[3] https://www.w3.org/TR/prov-o/#inverse-names
[4] http://richard.cyganiak.de/blog/2006/06/an-rdf-design-pattern-inverse-property-labels/
[5] https://github.com/w3c/sdw/issues/334