Difference between revisions of "OntologyRelationsMay19MeetingNotes"
| Line 20: | Line 20: | ||
Development relations: Questions about "simple" binary versions, versus refoccurent (one per occurrent) versus ternary (more than one refoccurrent) | Development relations: Questions about "simple" binary versions, versus refoccurent (one per occurrent) versus ternary (more than one refoccurrent) | ||
| + | |||
| + | AI: Larry write up oocyte maturation/follicle generation case motivating more than one refoccurent | ||
Revision as of 09:43, 19 May 2008
What is the scope of "All" needs to be addressed taking in to account some ontologies are canonical and others are outside normal. But "Normal" is contentious.
Sort out the case of cell membrane. Historical part of cell is OK. Is it still a cell membrane?
Two buckets more complex "relation" like expressions that are good for "user interface" versus simpler smaller set of core relations from which we can construct them. No assumption that everything that looks like a relation, is a relation in the sense of RO
On_the_surface_of lake, cell membrane - great if they were the same relation.
The scope of RO is both type level relations and instance level relations. But we make clear which are which.
Same or different identifiers for class/instance relations such as part_of?
Numeric ids for RO terms?
RO commits to BFO as upper level ontology
RO will have separate identifiers for class level and instance level relations. (And different class level ids where the quantification is different).
We will use numerical ids for RO terms.
Development relations: Questions about "simple" binary versions, versus refoccurent (one per occurrent) versus ternary (more than one refoccurrent)
AI: Larry write up oocyte maturation/follicle generation case motivating more than one refoccurent