ltc:Role - new class to be added

[essvee]

See meeting notes: 2nd March 23

Adding a separate, bare-bones, role class was recommended by the reviewer team after working through the PersonRole/Person classes, partially because it allows us to be able to tag a role (separate to the peron or persons performing it) with an identifier, which could be handy if you wanted a way to refer to an external controlled vocab/ontology.

• [x] Add new class-level term ltc:Role (we don't ahve a suggested def/any mappigns to external terms for this afaik)
• [ ] Add Role properties:
  • [x] Role.roleName (string, not mandatory, not repeatable)
  • [x] Role.hasIdentifier (array, not mandatory, repeatable)
  • [ ] ~Role.hasFunction (string, not mandatory, not repeatable)~
• [x] Delete old PersonRole.role property
• [x] Add PersonRole.hasRole property (array, not mandatory, repeatable)
• [x] Edit PersonRole class description/notes to recommend that, if you use this class, you should have at least hasPerson or hasRole populated.

[mswoodburn]

As the role is being broken out into a separate class, it might be clearer to rename PersonRole to e.g. PersonAttribution? Would also align better with the terminology for linking between agent and entity in PROV (https://www.w3.org/TR/prov-o/#wasAttributedTo).

[jbstatgen]

Mmh, there is also the relationship between a Person and an Activity: wasAssociatedWith (https://www.w3.org/TR/prov-o/#wasAssociatedWith). Will it be sensible to distinguish these two types?

[mswoodburn]

Changes made, apart from the hasFunction property - not sure what that was meant to be, and don't think featured when we reviewed the issue last week?

[essvee]

Changes made, apart from the hasFunction property - not sure what that was meant to be, and don't think featured when we reviewed the issue last week?

Yeahhh so apparently I added that in during the last meeting and I've no idea why - it shouldn't be a 'has...' for a string anyway. I reckon ignore and if it comes up again we can change it.