Re: status of xsd:duration in RDF

Hi Peter!

Am 09.05.2012 17:09, schrieb Peter F. Patel-Schneider:
 > Regardless of whether xsd:duration can make it into OWL 2, there is a
 > question of whether anyone in this group knows of any reason why
 > xsd:duration should not be added to RDF by the current RDF WG, aside
 > from a desire to have the datatypes in OWL match those in RDF.
 >
 > Any comments?

Two questions (with some implicit answers to your question):

1) What other datatypes are under consideration by the RDF WG to be 
"added to RDF"? There are a lot of datatypes being mentioned in the 
original RDF standard, which were not taken into account by the OWL 2 
and RIF specs, such as xsd:gYear, xsd:gMonthDay, xsd:gDay, xsd:gMonth. 
So, one could argue, if these datatypes are going to be added to RDF 
(again), why should we (the OWL WG) then care about xsd:duration? It'll 
be then just yet another datatype that is in RDF but not in OWL 2 and RIF.

2) What does "adding to RDF" precisely mean? Mentioning the datatype 
somewhere in the spec by its name? Or making it a normative part of the 
RDF semantics? In RDF 1, the definitions of RDF-, RDFS- and 
D-interpretations only contained rdf:XMLLiteral as a normative part. 
There was one specific datatype map being mentioned in the chapter on 
D-entailment, called the "XSD datatype map" (the one with the non-OWL2 
datatypes mentioned above), but it was not an official part of the 
semantics of any D-interpretation (as many people seem to believe), but 
was just a well-known datatype map, an example for how a "D" could look 
like (perhaps a "primus-inter-pares D", but not more). If "adding to 
RDF" is meant that way, then I would *not* care much, whether 
xsd:duration is included in the new version of the XSD datatype map or 
not. If, however, it is meant to become (together with its friends from 
the XSD datatype map) a normative part of any D-interpretation (which 
would be a big step from RDF 1), then I would care *much*, because any 
OWL 2 RDF-Based interpretation is a D-interpretation, and if all 
D-interpretations would include all the XSD datatypes from the RDF spec 
(plus, maybe, xsd:duration), this would mean that future OWL-N Full 
reasoners would have to support them all, which is quite a bit more than 
what OWL 2 Full reasoners are expected today (with possible 
implementation trouble here and there)! (To be mentioned: Most probably, 
this would then only be an issue for OWL N Full, not OWL N DL, as for 
current OWL 2 DL, the semantics and, in particular, the set of 
to-be-supported datatypes is specified independently from the RDF 
semantics specification).

So far from my slightly RDF-Based/OWL Full-centric point of view.

Best,
Michael

-- 
.........................................................
Dipl.-Inform. Michael Schneider
Research Scientist, IPE / WIM

FZI Forschungszentrum Informatik
Haid-und-Neu-Str. 10–14
76131 Karlsruhe, Germany
Tel.: +49 721 9654-726
Fax: +49 721 9654-727

michael.schneider@fzi.de
www.fzi.de

.........................................................
Forschungszentrum Informatik (FZI) an der Universität Karlsruhe
Stiftung des bürgerlichen Rechts
Stiftung Az: 14-0563.1 Regierungspräsidium Karlsruhe
Vorstand: Dipl. Wi.-Ing. Michael Flor, Prof. Dr. Ralf Reussner,
Prof. Dr. Rudi Studer, Prof. Dr.-Ing. J. Marius Zöllner
Vorsitzender des Kuratoriums: Ministerialdirigent Günther Leßnerkraus
.........................................................

Received on Wednesday, 9 May 2012 18:49:45 UTC