Open nmamanager opened 6 years ago
This is a data change and will be looked at in Phase Two. So out of scope for this phase of the project.
I think this is something that Trove themselves could do in their harvest setup. We could suggest to them which additionalType
values correspond to books and maps, but otherwise I don't think there's anything we'd need to add to the API.
I used SPARQL to make a list of all the types (there were 2419 distinct values) and then I scrolled through the list and assigned each one to the appropriate Trove category. Almost all the types belong to the "Images, Maps & Artefacts" category, but there were also dozens of types which belong to the other categories (mostly various kinds of books, business records, and music).
I supplied the mapping to Trove via email, and since I'd made a stylesheet to review the mapping, I also passed that on since they may be able to reuse it.
Currently all of our objects in Trove will display as 'objects' in the "Pictures, photos, objects" zone. However some also fit into other zones such as 'Books' and 'Maps'.
We do not currently have a way to easily identify these for the API, but will work on this for a future release.
Having additional type values would also be useful for other reasons.