Discussion: COB Workshop 2020-09-02

[jamesaoverton]

Let's use this issue for general feedback on my presentation to the COB Workshop. Specific issues should get their own tickets.

Presentation Goals

1. present the problem: shared modelling of quantitative values
2. present existing alternatives, and a classification scheme
3. present a draft proposal

Feedback

I would appreciate feedback on:

• desiderata: what do we want out of qualitative modelling?
• coverage: did I miss an important alternative?
• specific examples: did I make mistakes?
• general approach: too much detail? should we do a similar analysis for other important topics?

[jamesaoverton]

Notes from the call:

• desiderata
  • simplicity: break this down; simplicity for people, machines, Protege?
• alternatives: look at BCO
• other: can we use the same patterns from proposal-1 to say things about classes?
• Units Ontology
  • they're happy to update to ODK, Kai is happy to work on it
  • local names for units: consider using human-readable IRIs for units (e.g. UO:cm), rather than numeric IDs (e.g. UO:1234567)
  • UO is already punning classes as named instances
  • if we're using them as datatypes, that's one more pun; why not try to resolve this properly