- From: Eric Stephan <ericphb@gmail.com>
- Date: Thu, 8 Oct 2015 08:20:48 -0700
- To: Deirdre Lee <deirdre@derilinx.com>
- Cc: Public DWBP WG <public-dwbp-wg@w3.org>, Richard Cyganiak <richard@cyganiak.de>
- Message-ID: <CAMFz4jhDsmfUKJPo_zO_NgJuSSohgVRDYHQ7Yh1csszXuGB0Sg@mail.gmail.com>
Deidre thanks for sharing Richard thoughts. The rationale for wanting to use the dcat namespace was to closely correlate new vocabularies that were viewed as extensions of the dcat:Dataset and dcat:Distribution. I was thinking after our F2F meetings how other vocabularies handled perhaps similar circumstances. Dublin core [1] accommodates two namespaces to not break the previously defined vocabulary and provide extended in the new vocabulary. I'm wondering if instead of using: * http://www.w3.org/ns/dcat# (dcat:) for the Dataset Usage Vocabulary * use something like http://www.w3.org/ns/dcat/usage# (dcatu:) This would achieve from the human perspective loose association with dcat without causing possible confusion from someone looking for usage terminology in the DCAT specification. Eric S References [1] http://wiki.dublincore.org/index.php/FAQ/DC_and_DCTERMS_Namespaces On Wed, Oct 7, 2015 at 5:36 AM, Deirdre Lee <deirdre@derilinx.com> wrote: > Thanks to Richard for providing feedback on the reuse of DCAT ns for DQV > and DUV: > -------------------------- > > Hi Dee, > > My view is that different vocabularies should have different namespaces. > Having some terms in a namespace governed by one document and then other > terms in the same namespace governed by a different document is confusing > and counter-intuitive. It makes future maintenance harder, as several > documents with complex interrelationships would be affected. > > The strongest argument for putting everything into one namespace is, I > suppose, convenience for data publishers. They wouldn’t have to remember > which term is in which namespace. But this ship has sailed a long time ago. > In RDF, we have to live with terms being scattered over different > namespaces. Even when just using plain DCAT, one has to use terms in the DC > and SKOS namespaces. > > That being said, it’s the WG’s decision and not Fadi’s or mine, and I > don’t believe there’s any rule against adding terms to a REC-defined > namespace using a NOTE. > > Please feel free to forward this message to the list or share its contents > in whatever way you see fit. > > All the best, > Richard > > > > >
Received on Thursday, 8 October 2015 15:21:22 UTC