Example JSON has facet type from the Getty Art & Architecture Thesaurus, we don't have this information

[dannymandel]

In this example JSON https://github.com/isamplesorg/metadata/blob/develop/examples/OpenContext/test1.0Valid/ark-28722-k28d0b21r-v1.json, we have the following: Getty metadata keywords:

        {
            "keyword": "pottery",
            "keyword_uri": "http://vocab.getty.edu/page/aat/300010666",
            "scheme_name": "Object Facet: Getty Art & Architecture Thesaurus"
        }
        {
            "keyword": "Early Bronze Age",
            "keyword_uri": "http://vocab.getty.edu/page/aat/300019276",
            "scheme_name": "Styles and Periods Facet: Getty Art & Architecture Thesaurus"
        }

However, the OC API returns things like this:

 "inorganic material [getty-aat-300010360]": ["glass (material)"], 
 "inorganic material [getty-aat-300010360] [URI]": ["https://vocab.getty.edu/aat/300010797"]

so we end up with

{
  'keyword': 'glass (material)', 
  'keyword_uri': 'https://vocab.getty.edu/aat/300010797', 
  'scheme_name': 'Getty Art & Architecture Thesaurus'
}

which is almost what the example has, but we don't have the facet value, so we can't include it in the scheme name.

@smrgeoinfo @ekansa @datadavev please advise.

[dannymandel]

Eric has a way of including this, but it is involved. We shouldn't refetch until the work is done on the OC side.

[dannymandel]

This isn't a showstopper.

[ekansa]

The scheme name (vocabulary) can be added, since we keep those data internally. But it would involve additional database lookups, so it would involve a performance hit.

Also, it would take some extra work on my end to update concepts in all the various Getty AAT "facets" (we currently just track everything in the AAT as belonging to the AAT, not a more specific facet of the AAT).

So, should this be a "nice to have" but not really needed update? I do like the idea of better organizing our use of the Getty AAT. It's big and complicated. Organizing the AAT concepts we reference according to the AAT facets (in effect, having multiple "sub vocabularies") seems handy, but that's a big job for me.

[smrgeoinfo]

If we have the Getty AAT URI's we're already miles ahead in the linked data world. Scheme names can be 'nice to have'...