I'm going to stick with ORCID and ROR as the two sources to work with. Other identifier schemes exist in DataCite, but ORCID and ROR are open and resolvable. In the processing of a set of DataCite records, I will assemble data objects with simple key/value pairs (#1). Where I encounter ORCID or ROR identifiers for these entities, I'll resolve those from the source and pull in additional key/value content that I can exploit on the associated list.html pages for the labels as taxonomy terms.
I'm going to stick with ORCID and ROR as the two sources to work with. Other identifier schemes exist in DataCite, but ORCID and ROR are open and resolvable. In the processing of a set of DataCite records, I will assemble data objects with simple key/value pairs (#1). Where I encounter ORCID or ROR identifiers for these entities, I'll resolve those from the source and pull in additional key/value content that I can exploit on the associated list.html pages for the labels as taxonomy terms.