metadata model V2
how to make (faster) progress on defining what should come back from an ark?info inflection request
- KH: wishlist is helpful
looking for something simple to specify, eg, kernel
but it can wait, so that we finish the spec.
regarding n2t, it could be a reference or exemplar implementation
- Mark: unless you're creating a place to hang the core metadata,
here's where you put the core elements, but don't think we should
be in the business of defining those cores
point to community mechanisms for defining the cores,
but don't define themInfo? inflection data: what should be returned?
- Figuring out exactly what should be in the core spec vs extension is not easy.
- New a couple of pages added to clarify the intent.
- need use case approach.
TC: what machines will be requesting this?
JK: I'd say same as the LOD users
adopt-a-thing (use case)
- natural history specimens
- genealogical records
- archival record/object
- vocabulary terms
- archaeological artifact
- water sample
- various assignments made
- right now just trying to understand what metadata is of most value in a domain
TC: what I'd like my users to be able to look at a thing with an ARK and to get
a formal citation to it and to its parent;
would be willing to sign up for this genealogical citation use case,
also possibly for persons
KH: I could do the journal article use case
MP: I'll look into archival record use case
RM: here's an example authority record. is it a person
JK: if it could be a person or org, maybe it's an Agent
KH: how to put structure around, mechanism to bridge between core and kernel
KH: how about pulling out the DataCite examples and applying this to it?
MP: reminds me of what IIIF does, eg, here's where pairs go
CM: will get advice from Brian about helping with use case