ckan / ckanext-dcat

CKAN ♥ DCAT
163 stars 142 forks source link

Food for thought: optionally support PROV standard and ontology #105

Open ted-strauss-K1 opened 6 years ago

ted-strauss-K1 commented 6 years ago

Was reading through this post: https://github.com/ckan/ideas-and-roadmap/issues/154

And learned about the W3C PROV data model for provenance: https://www.w3.org/TR/prov-primer/

As I understand it PROV provides a more detailed and nuanced representation of provenance than would be captured by DCAT, and yet the two standards should be fully compatible and interoperable.

An optional PROV feature added to this plugin would attach PROV vocabulary and schema to the dataset object, and expose those in all the usual formats. This might mean duplicating some metadata fields with different attribute labels; so there would be both the current DCAT tags and new PROV metadata.

It seems like a relatively simple proposal for an already useful and stable plugin, so I thought I would post it here as food for thought.

Cheers!

ref: https://www.w3.org/TR/vocab-dcat/

dr-shorthair commented 6 years ago

DCAT is currently in revision here https://www.w3.org/2017/dxwg/wiki/Main_Page and https://github.com/w3c/dxwg One of the big items on the agenda is richer provenance traces, likely using the PROV ontology It hasn't really got going yet, but some members of the team are very experienced with PROV so expect to see some worked examples and patterns appearing here soon.

ted-strauss-K1 commented 6 years ago

That's a very exciting set of features in the pipeline, that would go a very long way to making CKAN a stronger offering for scientific data publishing.