Identifying aggregations of period concepts about periodo-data HOT 3 CLOSED

With the merge of periodo/periodo-server#65 we now have the ability to create and identify arbitrary sets of period definitions. We have decided not to implement any “conflated” concepts in the dataset or via the server. Instead, one will create a “bag” of definitions via the server and obtain a stable ID for that bag, which can be passed to local functions for conflating them in whatever way is desired.

from periodo-data.

Via Lex Berman:

One followup on our conversation: when Linda Hill first described to me her use of "broader" and "narrower" relations in Alexandria Feature Type Thesarus, it was pretty clearly neither spatial nor temporal, but purely semantic in the sense of 'superordinate' and 'subordinate.'

That's why I think in PeriodO case, if you have 12 items that each define the "Bronze Age," and you want to CONFLATE only three of them into your own concept of the Bronze Age, then you have:

Bronze Age (new conceptual container)
conflates - Bronze Age (item1)
conflates - Bronze Age (item2)
conflates - Bronze Age (item3)

however it does not work in the thesaurus sense:

Bronze Age (new conceptual container)
NT - Bronze Age (item1)
NT - Bronze Age (item2)
NT - Bronze Age (item3)

All of those items are peers. We are conflating them into a conceptual container, but we are not able to use the temporal span of the subordinate items and say that the super-ordinate Bronze Age is a "broader term" because it's not semantically correct. It IS a conflationOf them, though, so I think you need to mint your own relationship definition, unless you can find one for this.

from periodo-data.

Summary of above: A narrower spatiotemporal footprint does not imply a subordinate concept. Whatever relation we end up using must not imply skos:narrower.

from periodo-data.