Re: Use of XSD namespace in RDF recommendations

On Sep 4, 2012, at 12:36 PM, Richard Cyganiak <richard@cyganiak.de> wrote:

> Ivan, Antoine, Dan, Richard, Jeremy,
> 
> I guess I like the idea of informatively linking to both the 2006 SWBP Note on datatypes [1] and to the OWL 2 datatype definition mechanism [2], stating that both XML Schema and OWL 2 provide facilities for formally defining RDF datatypes, but that support for neither mechanism is required for RDF.

Perhaps I'm missing something, but it seems that RDF Concepts does have a normative relationship to XSD, as literals with no datatype IRI or language tag get the datatype xsd:string.

Also, in Turtle, native number representations are associated with xsd:integer, xsd:decimal and xsd:double.

true/false values are represented as literals with xsd:boolean.

We considered this in JSON-LD; JSON numbers are translated to xsd:integer or xsd:double, and true/false to xsd:boolean when transforming to RDF. When going from RDF, strings are used unless an option is specified do use xsd types.

Gregg

> Jeremy: Is [1] still considered up-to-date, or should we avoid drawing attention to it?
> 
> Antoine, Ivan, Dan: Perhaps one of you wants to take a stab at drafting a sentence that could be inserted into the Datatypes section [3] as another Note?
> 
> Cheers,
> Richard
> 
> [1] http://www.w3.org/TR/swbp-xsch-datatypes/
> [2] http://www.w3.org/TR/owl2-syntax/#Datatype_Definitions
> [3] http://dvcs.w3.org/hg/rdf/raw-file/default/rdf-concepts/index.html#section-Datatypes
> 
> 
> On 4 Sep 2012, at 18:38, Ivan Herman wrote:
> 
>> Dan,
>> 
>> the question is of course justified, but I should also add that, in fact, very very few people use xmls schema datatype definitions together with RDF, too. In both cases the difficulty is identical: to understand and process those datatypes an external 'tool' has to be brought in: either an xml schema or an owl processor... Mainly in a non-XML RDF world (ie, as Richard said, with a diminishing usage of RDF/XML) the chance of using XML schema based derived datatypes is getting smaller and smaller in my view.
>> 
>> I find the OWL 2 datatype definition possibilities one of the most interesting and potentially important part of OWL 2. I actually wish the relevant part of the specification was also made more known and possibly used in isolation; at present it is burried in the overall OWL 2 spec, which is of course not an easy read...
>> 
>> (Maybe it is worth some extra blog/note)
>> 
>> Ivan
>> 
>> ---
>> Ivan Herman
>> Tel:+31 641044153
>> http://www.ivan-herman.net
>> 
>> (Written on mobile, sorry for brevity and misspellings...)
>> 
>> 
>> 
>> On 4 Sep 2012, at 15:37, Dan Brickley <danbri@danbri.org> wrote:
>> 
>>> On 4 September 2012 21:11, Antoine Zimmermann
>>> <antoine.zimmermann@emse.fr> wrote:
>>>> FWIW, OWL 2 has a feature to define custom datatypes that can be written
>>>> completely in RDF, without using XML Schema.
>>>> 
>>>> Your example for Chapman codes can be written as follows, in Turtle syntax:
>>>> 
>>>> @prefix geo: <http://www.example.com/geo#>
>>>> @prefix xsd: <http://www.w3.org/2001/XMLSchema#>
>>>> @prefix owl: <http://www.w3.org/2002/07/owl#>
>>>> @prefix rdfs:<http://www.w3.org/2000/01/rdf-schema#>
>>>> 
>>>> geo:chapman-code  a  rdfs:Datatype;
>>>>  owl:equivalentClass  [
>>>>      a  rdfs:Datatype;
>>>>      owl:onDatatype  xsd:string;
>>>>      owl:withRestriction ( [xsd:pattern "[a-zA-Z]{3}"] )
>>>>  ] .
>>> 
>>> Interesting! Are many of these showing up "in the wild" yet?
>>> 
>>> Dan
>>> 
>> 
> 
>

Received on Tuesday, 4 September 2012 21:19:07 UTC