Closed maximelefrancois86 closed 8 months ago
Related - in OMS we also have OMS:Sampling.featureOfInterest
Will we be keeping the O&M alignment in this version? See #121 Since OMS = O&Mv3.0 the OMS alignment supersedes the O&M alignment.
I propose to retire the O&Mv2.0 alignment from this edition.
The O&Mv2 alignment has been retired.
The OMS alignment is summarized here: https://w3c.github.io/sdw-sosa-ssn/ssn/#OMS-Alignment-Property . There are several 1:n mappings. Each of these possibly implies a missing 'super-property' in OMS.
However, there is no sub-property capability in UML. And it is a UML pattern that scopes property names to classes, even when the useful semantics don't differ.
Nevertheless, it would be possible to define sub-properties in the sosa-oms: namespace, with the sub-property axioms living in the ssn-oms.ttl graph. e.g.
# in the SOSA-OMS graph
sosa-oms:observationHasFeatureOfInterest a owl:ObjectProperty .
sosa-oms:samplingHasFeatureOfInterest a owl:ObjectProperty .
sosa-oms:samplingHasFeatureOfInterest a owl:ObjectProperty .
# in the SSN-OMS graph
sosa-oms:observationHasFeatureOfInterest rdfs:subPropertyOf sosa:hasFeatureOfInterest .
sosa-oms:samplingHasFeatureOfInterest rdfs:subPropertyOf sosa:hasFeatureOfInterest .
sosa-oms:actuationHasFeatureOfInterest rdfs:subPropertyOf sosa:hasFeatureOfInterest .
and then use these in the alignment table. This is arguably more precise semantically, however it moves the SOSA-OMS alignment one step further away
What do you prefer?
O&Mv2 alignment has been retired
This is a very minor issue:
In section 6.3.5 Property Alignments, alignment to OM, it is stated that:
iso19156-om:OM_Observation.featureOfInterest equivalent property sosa:hasFeatureOfInterest
Actually sosa:hasFeatureOfInterest can also be applied to Actuations. And depending on the resolution of issue #111 , there may be even more classes that it will apply to.
I suggest to change this alignment as follows:
iso19156-om:OM_Observation.featureOfInterest sub property of sosa:hasFeatureOfInterest
And see if the same reasoning applies to other alignments.