Reframing "Time" as "Event-like features" or "Feature succession"

[sgillies]

I think a reframing helps discussion and documentation. Yes? No?

Actions would be rewrites of some of the markup files in this repo.

[kgeographer]

Agree adding "when" is adding support for "event-like features" and that is more descriptive than simply 'adding time.' It's also opening worm cans (a.k.a. re-framing opportunities) as predicted (by @sgillies) but worthwhile tackling this imho, even if the answer turns out to be to steer clear. [xref https://github.com/geojson/geojson-ld/issues/16 ]

Three questions arising as I encode Çatalhöyük data, where typically several time statements are associated with an object (excavation, deposition, etc.):

• do we care what a when object refers to, e.g. the class of temporal statement
• can the value of a when object be a URI pointing to a time period defined elsewhere? e.g. future period thesauri like PeriodO or Pleides periods
• shouldn't WhenCollection join Instant and Interval as types, similarly to geometry?

I confess to not being adept at LD yet, so I have more questions than answers. Making some stab at this for Çatalhöyük is high priority over next 3 weeks though.

[p3dr0]

@kgeographer

• what when is ? I guess it depends of what the event is (excavation, deposit, etc) all other time values could in fact be properties of the event and "explained" by the LD value
• good point, would it be worth to open a different issue focusing on linking ?
• agree, and in this case would they be restricted to a type (IntervalCollection, InstantCollection)?

[sgillies]

Reframing has been done in time.md.