- From: Maxime Lefrançois <maxime.lefrancois@emse.fr>
- Date: Mon, 27 Mar 2017 07:48:05 +0000
- To: SDW WG Public List <public-sdw-wg@w3.org>
- Message-ID: <CALsPASUsaZ+8NUPBxEw8BGqpO2+pNvfg+JYVJW2Qgva-UOVXiA@mail.gmail.com>
Dear all, During last week's F2F we mentioned the qualified cardinality restrictions that SSN will add to sosa:Observation. Some of the local restrictions on sosa:Observation are: ObjectAllValuesFrom( sosa:featureOfInterest sosa:FeatureOfInterest ) ObjectExactCardinality( 1 sosa:featureOfInterest sosa:FeatureOfInterest ) ObjectAllValuesFrom( ssn:observedProperty ssn:Property ) ObjectExactCardinality( 1 ssn:observedProperty ssn:Property ) Now It would make sense to axiomatize SSN further and represent the fact that: "If an observation observes a property, then it also observes the feature of interest of this property." This would be modelled with SubObjectPropertyOf( ObjectPropertyChain( ssn:observedProperty ssn:isPropertyOf ) sosa:featureOfInterest ) On the other hand, one cannot use exact cardinality restriction of 1 on non-simple properties, and this axiom makes sosa:featureOfInterest become a non-simple property. So my question is: which one of these two axioms do you think are more useful in practice? ObjectExactCardinality( 1 sosa:featureOfInterest sosa:FeatureOfInterest ) Or: SubObjectPropertyOf( ObjectPropertyChain( ssn:observedProperty ssn:isPropertyOf ) sosa:featureOfInterest ) Best, Maxime
Received on Monday, 27 March 2017 07:48:48 UTC