search

ICA-EGAD / RiC-O

ICA Records in Contexts-Ontology (ICA RiC-O) GitHub repository web pages
https://ica-egad.github.io/RiC-O/
47 stars 17 forks source link

Shouldn't hasOrHadRelation be related to agentHasWorkRelation through some property chain ? #39

Closed tfrancart closed 7 months ago

tfrancart commented 1 year ago

hasOrHadRelation subProperty of "agentHasWorkRelation + workRelationConnects"

But I see the problem : by doing so, since the relation is not oriented, this would create a hasOrHadRelation to the entity itself.

Maybe in this case add an informal documentation link between the 2 properties, and a small explanation ?

tfrancart commented 1 year ago

Also : I note that isAgentAssociatedWithAgent is defined with a formal property chain equivalence (agentConnectedToAgentRelation + agentRelationConnects), which will trigger the problem described above : the agent will be linked to itself

florenceclavaud commented 1 year ago

Well this is to be thought of obviously. Conceptually speaking (not formally) I am not sure it would be such a big problem that this very generic relation (isAgentAssociatedWithAgent) is inferred between the agent and itself. However of course we can (and should probably) remove it.

florenceclavaud commented 7 months ago

Actually, when I rolified the relations (see #67) I included the few n-ary quite high level Relation classes that have a symmetric relation as a shortcut. So the problem is stil there; and I am not sure this can be solved in RiC-O. Anyway IMHO again it is not really a problem, at least conceptually or semantically speaking.

florenceclavaud commented 7 months ago

I think I will soon close this issue, unless somebody there has an idea about this.

florenceclavaud commented 7 months ago

I am closing this issue. We can reopen it whenever it seems useful and possible to work on this.