Use contactPoint, not person

[jpmckinney]

A group discussion nixed changing person to contactPoint. If you think this was wrong, please re-submit a new issue just for this item for broader discussion.

@MarinaMartin Could you share more information about the group's reasons?

DCAT uses contactPoint. What will this project use in its RDF serialization? Every RDF consumer would expect it to use dcat:contactPoint given that it adopts DCAT. Will the documentation explicitly describe this difference between the RDF serialization (which should use contactPoint) and other serializations (which will use person according to the above decision)? It seems simpler to have consistent terms across all serializations.

[MarinaNitze]

Actually, very fair point, encompassing change now in #44. The group was erring on the side of reducing the number of changes to the metadata (we are all aware that changing it at this point is not the ideal), but let's do it as right as we can this time. Thanks.

It looks like DCAT expects a vCard for contactPoint, though? We have been prescribing just a name, which was previously "Last, First". Issues with just leaving it a free-form string?

[jpmckinney]

True, it's an issue to not respect the range for the property. I had flagged using vCard as a potential issue to the working group responsible for DCAT, but we had no evidence of there being a problem at the time. To respect the range, the RDF (in Turtle syntax) would look like:

ex:dataset dcat:contactPoint [ a vcard:VCard; vcard:fn "Last, First" ] .

It may make sense to mint a new property in order to avoid this range issue. However, if the above RDF is considered acceptable, then dcat:contactPoint can be used throughout.

[MarinaNitze]

I think tying what we want to be a person's name to a vCard is adding an unnecessary level of complexity. I would personally prefer leaving it a free text string altogether and not having "Last, First" since the goal is simply to have a responsible party. What does a good compromise look like?

[jpmckinney]

There is an Open Organizations ontology with an oo:contact property that has no range according to the docs. contact may be more semantic than person in any case. That ontology also has properties for corrections, etc. It may be a good alternative.

I also prefer free text. I don't especially understand the reasons for the "Last, First" format (which is normally used when you need to sort a list by last name), plus it has poor multilingual/multicultural support ("family name" and "given name" are better terms, since in some cultures the first name is the family name).

[jpmckinney]

Cool, looks like contactPoint was used in #44