Policy for adding domain and range classes to RO

[wdduncan]

Can we develop a consistent policy for including classes mentioned in definitions? For example, the definition for participates in is:

a relation between a continuant and a process, in which the continuant is somehow involved in the process

However, the range is occurrent even though RO does have the class process (why is process not the domain?).

Similarly, the definition for derives from refers to material entities:

a relation between two distinct material entities, the new entity and the old entity, in which the new entity begins to exist when the old entity ceases to exist, and the new entity inherits the significant portion of the matter of the old entity

But, neither the domain nor range is specified as material entity.

[cmungall]

Agreed, The policy should be that we are consistent.

Copying questions from agenda to keep things in one place:

Bill D: We need to be more consistent about when we do or do not import the classes mentioned as domains and ranges in definitions.

Agreed

I think we should treat cases where the text definition don't agree with the logical axioms as errors.

Messing with domains and ranges might break things. Making it more restrictive may affect users.

Yes. In theory even relaxing can change entailments but in practice this is less of an issue, as no one should relying on ranges for classification

New terms to have well specified domains and ranges.

where possible. E.g. if there is no agreement on COB

For currently existing terms: change the textual definitions to match?

Let's make a distinct issue for each such case, and make decisions as to how we resolve the conflict on that issue.

Mostly terms in RO-core?

possibly - let's make issues as we encounter them