Re: [dxwg] definitions clarifications of conforms to (#1130)

Perhaps the root of the problem that the subject of the triples in dcat:Resource is a resource in the world; the subject of the triples in dcat:CatalogRecord is a set of triples that describe a resource, and the subject of triples in dcat:Distribution is a representation of a resource documented in a dcat:Resource instance that links to the dcat:Distribution instance. Apparently this is not clear in the documentation to some readers.  This might be partly because most existing metadata schemes don't clearly distinguish metadata about the metadata record itself, metadata about the resource (at the FRBR 'work' level), and metadata about distributions (representations, expressions/manifestations) of the resource.  

The UML representation obscures the relationships because the identity of the catalogRecord, and the identity of distributions of the resource are not explicit in the diagram; the identifier for the resource is dcat:Resource.dct:identifier, which ideally would be the subject of the triples  that implement the properties in dcat:Resource, but the UML model doesn't make that clear.  

It appears to me that the various comments above all recognize that there is metadata about the metadata record, metadata about a resource (including catalog, which is a resource), and metadata about distributions of a resource.  Practically speaking, because there are often 1:1 relationships between the metadata record, the resource, and its distribution (if there is only one distribution), the simple implementation of the conceptual model can be flattened into one record with one identifier. This works fine for many use cases, but it's critical to bear in mind that it's a simplification that doesn't work for more complex kinds of situations. 

GitHub Notification of comment by smrgeoinfo
Please view or discuss this issue at using your GitHub account

Received on Monday, 4 November 2019 18:24:49 UTC