Re: SPARQL 1.1 Entailment Regimes

On Fri, Jun 10, 2011 at 7:38 AM, Richard Cyganiak <richard@cyganiak.de>wrote:

>
> On 9 Jun 2011, at 16:17, Alex Hall wrote:
> > For those interested, I pulled all this together on a wiki page:
> >
> > http://www.w3.org/2011/rdf-wg/wiki/XSD_Datatypes
>
> For completeness, I added those XSD 1.0 types that are explicitly noted as
> being excluded from RDF.
>
> Are there any new types in XSD 1.1 that are not listed in the table?
>

The only other one I can find is xsd:precisionDecimal, which I've added to
the table. I believe this has been noted in the past as an at-risk feature.

There is also a new special built-in type, xsd:anyAtomicType, which is the
base type from which all atomic-valued types are derived. It's convenient
because it explicitly excludes the sequence-valued types which don't fit in
with the RDF datatype model, and might be worth mentioning in an informative
note ("Only those datatypes derived from xsd:anyAtomicType should be used in
RDF..."). Then again, it might just confuse the matter.


>
> > A couple of questions arise from this:
> > 1. Should RDF's XSD datatype map be expanded to include the new XSD 1.1
> xsd:dateTimeStamp (referenced by OWL2)?
>
> +1.
>
> And perhaps the various forms of durations too. (xsd:duration not being
> included in RDF is awkward, as one comes across ISO-8601 durations
> reasonably often in non-RDF formats.)
>
> > 2. Should we reference the new XSD1.1 spec for RDF?
>
> +1.
>
> Are these XSD 1.1 questions an ISSUE in the tracker yet?
>

No -- should they be?  I don't know the process for raising issues.

-Alex



>
> Best,
> Richard