RE: SSFS clarification: Datatype definitions and typed literals

Hi!

Talking for the RDF-Based Semantics: If we would allow for literals 
of defined datatypes, such as

  "42"^^ex:IntegersSmallerThan50

with 

  ex:IntegersSmallerThan50 := xsd:integer[xsd:maxExclusive 50]

AFAICT, the RDF-Based Semantics would then need to be extended to 
be aligned with the Direct Semantics (to make the correspondence
theorem valid). The reason is that the current RDF-Based semantics 
(actually the semantics of "D-entailment" in Section 5 of the 
RDF Semantics specification) only applies to datatypes that 
are included in the used datatype map. But defined datatypes 
are in general not part of the datatype map. So the above 
example literal, while being allowed to be used in an 
OWL 2 Full ontology (aka RDF graph), would not have the 
intended/expected meaning (it would be just some undetermined 
literal). 

I would not say that I am against such an extension per se (it 
seems even nice to have to me at a first look), but I guess 
that this comes pretty late in the process, and would be, 
IMO, an extension of the design of OWL 2 (both in DL and Full).

I therefore think that Mike's suggestion, to explicitly state
in the Structural Spec that such literals are disallowed in DL, 
is fine. This statement would be editorial, since it just 
clarifies things. In the RDF-Based Semantics, no action will 
then be needed.

Michael

>-----Original Message-----
>From: public-owl-wg-request@w3.org [mailto:public-owl-wg-request@w3.org]
>On Behalf Of Mike Smith
>Sent: Friday, June 12, 2009 7:45 PM
>To: W3C OWL Working Group
>Subject: SSFS clarification: Datatype definitions and typed literals
>
>I believe SSFS should be clarified w.r.t. whether datatypes defined in
>DatatypeDefinition axioms can be used with literals.
>
>E.g., given
>
>DatatypeDefinition( :gt18 DatatypeRestriction( xsd:integer
>xsd:minExclusive "18"^^xsd:integer ) )
>
>Is "19"^^:gt18 a valid literal?
>
>Section 9.4 says "A datatype definition DatatypeDefinition( DT DR )
>defines a new datatype DT as being semantically equivalent to the data
>range DR; the latter MUST be a unary data range. This axiom allows one
>to use the defined datatype DT as a synonym for DR — that is, in any
>expression in the ontology containing such an axiom, DT can be
>replaced with DR without affecting the meaning of the ontology. "
>
>And Section 5.7 says "Each literal consists of a lexical form, which
>is a string, and a datatype from the datatype map."
>
>Since section 9.4 says nothing about the new datatype being added to
>the datatype map, I assume that the final condition does not hold for
>the literal, and it is therefore invalid.  Further, substituting a
>datarange into a typed literal doesn't make sense.  If this
>interpretation is correct, I suggest adding a sentence to the second
>paragraph Section 9.4 that explicitly states this, e.g.,
>
>"The datatypes defined by datatype definition axioms are not added to
>the datatype map and MUST NOT be used in literals."
>
>Thanks,
>--
>Mike Smith
>
>Clark & Parsia

--
Dipl.-Inform. Michael Schneider
Research Scientist, Dept. Information Process Engineering (IPE)
Tel  : +49-721-9654-726
Fax  : +49-721-9654-727
Email: michael.schneider@fzi.de
WWW  : http://www.fzi.de/michael.schneider
=======================================================================
FZI Forschungszentrum Informatik an der Universität Karlsruhe
Haid-und-Neu-Str. 10-14, D-76131 Karlsruhe
Tel.: +49-721-9654-0, Fax: +49-721-9654-959
Stiftung des bürgerlichen Rechts, Az 14-0563.1, RP Karlsruhe
Vorstand: Prof. Dr.-Ing. Rüdiger Dillmann, Dipl. Wi.-Ing. Michael Flor,
Prof. Dr. Dr. h.c. Wolffried Stucky, Prof. Dr. Rudi Studer
Vorsitzender des Kuratoriums: Ministerialdirigent Günther Leßnerkraus
=======================================================================