Re: [dxwg] Make dcat:Dataset a subclass of prov:Entity

@dr-shorthair
During the DCAT Call two or three weeks ago,  you pointed out that an agreement on this issue might depend on what kind of "reasoning"  are we expecting consuming the DCAT metadata, and also whether or not we are going to provide constraints with SHACL. I think you were right,  have we made any progress on the two points above?

Perhaps, another relevant point is to decide between

-  a "pay-as-you-go" use of provenance, where we support catalogs in gradually moving to a PROV enabled version when need, but we accept that datasets and catalogs might not always need a full use of the provenance;
 or 
- a "fully-fledged provenance anyway" approach, where we force the use of a fully-fledged provenience in every DCAT dataset and catalog.

I tend to prefer the pay-as-you-go approach, as  I am not sure that having the PROV qualified form is strictly necessary for all the DCAT catalog. For example,  I would prefer to recommend the materialization of DCT-PROV mapping only for certain kind of catalogs, e.g. the catalogs That harvest DCAT dataset from third-party catalogs. 


-- 
GitHub Notification of comment by riccardoAlbertoni
Please view or discuss this issue at https://github.com/w3c/dxwg/pull/94#issuecomment-365338163 using your GitHub account

Received on Tuesday, 13 February 2018 17:20:13 UTC