Convert additional RBMS XML vocabularies to RDF

[rjyounes]

All vocabularies are now in /tools/rbms_vocabs/data. We will want to put these in namespaces like https://w3id.org/arm/core/vocabularies/rbms_genre/0.1/> etc. and the RDF output in /core/vocabularies/rbms_genre/0.1/ etc.

@timathom Would you be willing to run your XSL on the additional vocabularies? I think it just requires a change of the namespace, so someone else could do it too if you don't have the time. (I don't think we should create different XSLs for each namespace, but maybe just add a note about modifying it as needed.) I believe we will not be using any of the other vocabularies for VitroLib -- @jak473 please correct me if I'm wrong -- so this is not top priority, but it would be nice to have it done.

I'm assigning this to myself just so I remember to keep track of it.

[timathom]

Yes, I can work on this. Should I add the "type" suffix to any of the namespaces? Or do we need this suffix at all?

[rjyounes]

The "type" suffix was just for the binding type vocabulary, since there are several other RBMS vocabs for bindings. The namespaces should all have "rbms_" as a prefix, so that it's clear they are RBMS vocabs and not ours.

Thanks a lot!

[rjyounes]

@timathom Do you think we should manually add an rdf:Description of the entire vocabulary at the top of each file, to identify the source and date of the export? We could use type void:Dataset, or not.

[timathom]

Yes, I was thinking of generating a skos:ConceptScheme resource for each vocab (e.g., http://id.loc.gov/authorities/genreForms.skos.rdf), but could use void:Dataset, depending on preference.

[rjyounes]

I forgot about skos:ConceptScheme. That would be preferable.

[rjyounes]

@timathom Your question about the "_type" affix got me thinking, and after looking at the original XML file I realized the binding namespace should just be https://w3id.org/arm/core/vocabularies/rbms_binding/0.1# without "_type." Sorry about the confusion.

I've made the change in the xsl file, the test xspec file, and the output rdf file.

[timathom]

Okay, that makes sense. Also, I had been using / instead of # on the namespace. I guess # is preferred for SKOS?

[rjyounes]

Can you send me a reference for # being preferred for skos vocabularies?

In general I prefer / because with # the entire document is returned for any requested term. That could be problematic on large vocabularies.

[timathom]

Oh, I thought that you had switched to # in your last update, and I remembered seeing that in other SKOS vocabs, so I was going on that. :) But maybe I was the one who introduced the # by accident. Anyway, I'm fine with using /.

[rjyounes]

I think it must have been you and I didn't notice it in your files - I've used / in my SHACL file.

In this variable assignment you have neither slash nor hash:

<x:param name="base-uri"

https://w3id.org/arm/core/vocabularies/rbms_binding/0.1</x:param> Is that correct?

[timathom]

Yes, I was adding the slash later, but I will update so that everything is consistent.

[rjyounes]

OK, I'll leave it to you to make all the namespace updates so that we aren't stepping on one another's toes. Thanks.