RE: Comments on DCAT draft

Ah - So you clicked through and found the dcat-schema.ttl RDF file which had not yet been referred to explicitly from the document. Very enterprising. 

We probably need to say 'alignments to other vocabularies, some of which are normative and some merely recommendations' since we probably want to leave the door open to either on a case-by-case basis. That's what we did with SSN - see for example https://www.w3.org/TR/vocab-ssn/#OM_Alignment which does not have the 'non-normative' flag that most of the other alignments have in this section. 

Of course this strategy somewhat begs a question about conformance. On the one hand we are packaging the alignments into separate graphs so users can take them or leave them. On the other hand we do not segregate the conformance section to enable this, so there is an implication that conformance should be to *all* normative material found in the specification. Practically I doubt if there would be many consequences, but in principle there is a tension here. 


(OGC and ISO resolved this by having several separate conformance classes in a standard. These typically build on each other to satisfy more and more requirements, but can also segment a standard into parallel implementations from a common base - other patterns as well.) 


Simon 

-----Original Message-----
From: Karen Coyle [mailto:kcoyle@kcoyle.net] 
Sent: Thursday, 20 September, 2018 17:02
To: Cox, Simon (L&W, Clayton) <Simon.Cox@csiro.au>; public-dxwg-wg@w3.org
Subject: Re: Comments on DCAT draft

Thanks, Simon. What tripped me up, reading from top to bottom, was when I reached 6.1, which mentions the "alignments":

The (revised) DCAT vocabulary is available in RDF. The primary artefact dcat.ttl is a serialization of the core DCAT vocabulary. Alongside it are a set of other RDF files that provide additional information, including:

    alignments to other vocabularies
    additional axioms, which can be useful in some contexts

Perhaps there it may be useful to say "non-normative alignments..."
since the schema.org section is much further down in the document.

Also note that I chatted with Tom Baker of DCMI and he indicated an upcoming effort between schema.org and DCMI to create mutually agreed equivalency statements between those vocabularies. I'll report back to this group any progress on that. There is a preliminary list but it hasn't been made public.

kc


On 9/20/18 3:14 AM, Simon.Cox@csiro.au wrote:
> Should have dropped in this link related to #2 
> https://rawgit.com/w3c/dxwg/dcat-alignments-schema/dcat/index.html#dca

> t-sdo The note below the heading says "This section is non-normative" 
> and also there is a note three paragraphs down that says "This alignment of DCAT with schema.org is provisional and non-normative. "
> 
> So I think we are in the clear. 
> 
> Simon
> -----Original Message-----
> From: Simon.Cox@csiro.au [mailto:Simon.Cox@csiro.au]
> Sent: Thursday, 20 September, 2018 07:39
> To: kcoyle@kcoyle.net; public-dxwg-wg@w3.org
> Subject: [ExternalEmail] RE: Comments on DCAT draft
> 
> #1 - perhaps the header is wrong - should be "DXWG Family of Documents" ?
> #2 - the only mention of equivalence relationships between external vocabularies and schema: elements is in the Alignments chapter. This is clearly marked 'Non-normative' (actually nothing in the ED as I write this email, though a PR is pending which does show these). 
> 
> -----Original Message-----
> From: Karen Coyle [mailto:kcoyle@kcoyle.net]
> Sent: Thursday, 20 September, 2018 00:23
> To: public-dxwg-wg@w3.org
> Subject: Comments on DCAT draft
> 
> I had promised to do a read of the DCAT draft, mainly to catch any areas that weren't clear to me from a "new reader" point of view. Most of the comments below are editorial, and thus possibly a matter of taste - use or not, as desired.
> 
> I have two comments that might be substantive:
> 
> #1
> DCAT Family of Documents
> - I think that the profile-related documents are perhaps a family, but the intention was that those not be specific to DCAT. Does DCAT fit the definition of "family member" in this case?
> 
> #2
> 6.2 says:
> 
> "The definitions (including domain and range) of terms outside the DCAT namespace are provided here only for convenience and MUST NOT be considered normative. The authoritative definitions of these terms are in the corresponding specifications:"
> 
> What about where statements define an equivalence or relationship between a class or property in DCAT and one in another namespace, e.g.:
> 
> dcat:Catalog
> owl:equivalentClass schema:DataCatalog ;
> 
> or in the case of statements in which both subject and object are outside of DCAT:
> 
> dc:title
> owl:equivalentProperty schema:name ;
> 
> Is the definition about "normativity" clear for the interpretation of these types of examples?
> 
> ## Editorial comments ## (my suggestions after the "-")
> 
> Introduction:
> "It also makes it possible to have a decentralized approach to publishing data catalogs and makes federated search for datasets across catalogs in multiple sites possible using the same query mechanism and structure. Aggregated DCAT metadata can serve as a manifest file as part of the digital preservation process."
> - It also makes possible a decentralized approach to data catalog publication and facilitates federated search for datasets in catalogs across sites using a single query mechanism and structure. Aggregated DCAT metadata can serve as a manifest file for digital preservation.
> 
> 3. Namespaces
> However, it can be noted that DCAT...
> - However, note that DCAT
> 
> DCAT itself defines a minimal set of classes and properties of its own.
> - DCAT defines a minimal...
> 
> 4. Conformance
> 
> A data catalog conforms to DCAT if:
> 
>     Access to data is organized into datasets, distributions, and data-services.
>     An RDF description of the catalog itself and its datasets, 
> distributions, and data-services is available
>  
> - Wasn't there discussion of datasets not in catalogs?
> https://github.com/w3c/dxwg/issues/62 Is this still an open question? If so, should it be noted here?
> 
> 5. Vocabulary
> dcat:Resource represents an item in a catalog.
> - dcat:Resource represents a dataset or service in a catalog.
> - dcat:Resource represents an entry in a catalog.
> 
> That's all - very impressive work. Thanks, everyone.
> 
> kc
> --
> Karen Coyle
> kcoyle@kcoyle.net http://kcoyle.net

> m: 1-510-435-8234 (Signal)
> skype: kcoylenet/+1-510-984-3600
> 

--
Karen Coyle
kcoyle@kcoyle.net http://kcoyle.net

m: 1-510-435-8234 (Signal)
skype: kcoylenet/+1-510-984-3600

Received on Thursday, 20 September 2018 07:24:16 UTC