Re: Can Shapes always be Classes?

On 11/20/2014 1:05, Karen Coyle wrote:
>
>
> On 11/4/14 9:16 PM, Holger Knublauch wrote:
>> I believe there is a fundamental difference in how the various proposals
>> treat the relationship between resources and their shapes:
>>
>> - In OWL and SPIN, constraints are attached to classes. rdf:type triples
>> are used to determine which constraints need to be evaluated for a given
>> instance.
>>
>
> With little effort, I was able to find data sets from the Linked Open 
> Vocabularies that either 1) do not use classes at all or 2) do not use 
> rdf:type triples. These are just a sample (each is a single example):
>
> http://eprints.soton.ac.uk/cgi/export/eprint/272587/RDFXML/eps-eprint-272587.rdf 
>
> http://datahub.io/dataset/sweto-dblp.rdf
> http://dati.camera.it/ocd/data/persona.rdf/p305757?output=xml
> http://www.dbpedialite.org/things/52780.rdf

FWIW all these examples are full of rdf:type triples. And I guess the 
instances that do not have a type can infer an rdf:type triple using 
rdfs:domain and ranges.

But for the sake of a theoretical discussion let's assume you have a 
document without any rdf:type triples. What would you do with it? How 
would a constraint engine know which constraints to run? I guess someone 
could define a property that links a resource with a Shape. But that's 
basically what rdf:type does in SPIN. I'd rather use an already 
well-established mechanism than coming up with a new property for a very 
similar task.

>
> Also note that Europeana, one of the larger cultural heritage 
> datasets, uses the class designation form:
>
> <edm:WebResource rdf:about="http://content.staatsbibliothek......
>
> for all classes.

I am not sure what point you are trying to make here. The above means 
that the resource has rdf:type edm:WebResource.

Regards,
Holger

Received on Wednesday, 19 November 2014 23:21:23 UTC