Reduce the number of connectors and associated classes by using attributes

Why? To make the model more lightweight and condense it visually

SEMIC PwC Comment: In the original model it is often the case that the type of a class is expressed by a connector towards a second class. A concrete example of this would be that of Action.hasStatus.Status, where Action is the first concept, hasStatus is the connector towards the second concept Status. Status is then populated with a code lists. The recommendation is to add Status as an attribute on Action and give it range skos:Concept.

University of Bologna’s Comment: This reduces the expressiveness of the ontology to have reification and reasoning. If the specification of domains and ranges is prevented it does not mean that the ontology is more interoperable. It only means that the ontology has little or no semantics. Accordingly, this prevents correct, coherent and meaningful semantic interoperability, which, instead, should be the ultimate goal of any data schema or ontology. It is not clear the rationale behind this suggestion. All the reductions are proposed for paths that links entities of specific type with their classification (e.g. Action.hasStatus.Status - in which hasStatus in an owl:ObjectProperty).

The suggestion makes confusion between the RDF data model and object-oriented data modelling, which might look similar although inherently different. In fact, the proposed solution is to represent this knowledge by means of attributes (e.g. Action.status) having value being individuals of the class skos:Concept. Nevertheless, the right representation for this is still of the form Action.status.Concept, resulting in no path reduction in RDF. In this case the only path reduction can be achieved by using datatype properties instead of object properties, causing information loss.