search

SNAP-DRGN / Cookbook

The SNAP:DRGN Cookbook – recommendations for producing interchange Linked Open Data from historical person data
Apache License 2.0
5 stars 2 forks source link

WG2 Person types (from Linked Pasts 6) #2

Open gabrielbodard opened 2 years ago

gabrielbodard commented 2 years ago

Working group: Proposal to document new person types in the SNAP Cookbook

Members: Gabriel Bodard, Greta Hawes, Ulrike Peter, Scott Smith

Proposal

Recommend a range of classes (in lawd: namespace or elsewhere) for person-types/agent-types in SNAP Cookbook. This might allow distinction of historical from mythical figures (e.g. gods, monsters…), but should also permit fallback to higher-order entities (agent/group) for databases that don’t make that distinction internally

LAWD ontology of person-types, as recommended by SNAP:

A type drawn from this list:[1]

  • lawd:Person
  • lawd:Deity
  • lawd:Group
  • lawd:MythologicalCreature

Or, if you cannot distinguish between the above categories, use the catch-all:

  • lawd:Agent

The LAWD Ontology (see https://github.com/lawdi/LAWD#classes-1) doesn’t actually define MythologicalCreature, despite this recommendation.

The main objection to the current state of affairs is the hierarchical nature of the ontology: all of the person-types (person, sacred/divine, monster, animal, etc.) should be able to elegantly degrade to either "agent" or "group", suggesting perhaps that all of these classes should be seen as stackable, rather than a hierarchy—with only a top level class enabling compatibility between more and less granular datasets.

gabrielbodard commented 2 years ago

Previous discussion on this WG took place in the comments section of this Googledoc.