Closed bertvannuffelen closed 2 years ago
I think it would be convenient at this point to explain the distributions again. The EDP has a good definition:
If a dataset contains more than one distribution, all distributions are identical in content, they differ only in the representation of the data. For example, a dataset can have two distributions, one offering the data as PDF and the other offering the identical data as machine-readable RDF/XML.
A user may indeed expect to find further guidance on how to interpret 'Distribution' in such a user guide. However, it seems difficult to avoid the interoperability versus flexibility discussion as touched upon in these related GitHub issues: https://github.com/w3c/dxwg/pull/789, https://github.com/SEMICeu/DCAT-AP/issues/155. https://github.com/opendata-swiss/dcat_ap_ch/issues/112 among others...
A new paragraph on the content of distributions has been added.
A discussion on a specific case at W3C: https://github.com/w3c/dxwg/issues/1390
A short usage guide for datasets, distributions and data services as been proposed in https://github.com/SEMICeu/DCAT-AP/blob/2.1.0-draft/releases/2.1.0/usageguide-dataset-distribution-dataservice.md.
Comments and remarks w.r.t. this texts are welcome. This text is the proposal to resolve the issues labeled with topic:cataloged-resources