Publisher/source metadata at a record level

In the Technical Committee meeting today we discussed the potential need for record level publisher metadata (building from Proposal for 3.2 - Extra metadata with API feeds which provides feed level publisher data)

This would create potentially 2 levels of “publisher” the original source of each record and the collator of a group of records which are published as a feed.

In BODS at the statement level when looking at beneficial ownership declarations we distinguish these as the “publisher” (e.g. Companies House publishing a dataset) and the “source” (e.g. the business owner submitting their information)

For HSDS this discussion is concerned wiith the “source” level of data. This is potentially something we could incorporate into the metadata object or this could be a new “publisher” or “source” object.

What information would be useful here? name, contact info (including where to raise corrections)

copying over @bloom comments about possible field names

Yeah i mean these should be subject to further consideration, but i think there’s publisher of the aggregated dataset, steward of individual records, and then there should be a term for a representative of the organization that the record is about (maybe that’s source tho in other contexts i use registrant or representative )

In the BODS schema we have something a bit like what @bloom is describing (although each of these are held at the record level rather than any being aggregated to a higher level).

  • publisher (as above)
  • source - this is an object which is kind of a 2 in 1 for what we could describe here as a steward or representative - it includes source.type for whether it’s a self declaration, third party information, or from an official register. source.assertedBy is the name of the agent making the assertion. So this object could be used for a person declaring their own interests or an official register declaring their interests. It doesn’t include contact information beyond just a URL though.

I don’t think this approach would transfer directly to OR but might be useful to have as a reference