Re: Changes to make S&AS consistent with RDF Semantics document

Jeremy,

>
>Herman:
>> I would like to point out that there is no inconsistency between RDF 
>Semantics
>> and OWL S&AS.  There exists an inconsistency between S&AS and OWL Test.
>> As you noted, tests 201 to 205 would need reconsideration.
>
>I might be being a bit slow here ....
>
>OWL S&AS says that:
>  OWL DL datatype theories must contain xsd:string and xsd:integer and 
may 
>contain other builtin datatypes (including rdf:XMLLiteral)
>
>  OWL Full datatype theories must contain xsd:integer and xsd:string
>[[
>Definition: Let D be a datatype map that includes datatypes for 
xsd:integer 
>and xsd:string. An OWL interpretation, I = < RI, PI, EXTI, SI, LI, LVI >, 
of 
>a vocabulary V, where V includes the RDF and RDFS vocabularies and the 
OWL 
>vocabulary, is a D-interpretation of V that satisfies all the constraints 
in 
>this section. 
>]]
>
>The latter can be read with an understood "in addition to rdf:XMLLiteral" 

>which is required from RDF semantics

Exactly.

>
>If we take this reading of OWL S&AS, which I (as an individual - not as 
editor 
>nor as HP rep) could go along with 

This is the only reading that one can make of the current
version of S&AS.  The inclusion of semantic conditions from RDF Semantics
implies that rdf:XMLLiteral should be understood to be included here as 
well.

>then this suggests that the following part 
>of OWL test is misleading in the same way as the quoted para from OWL 
S&AS 
>vis:
>current text
>[[
>The datatype theory of an OWL consistency checker MUST minimally support 
at 
>least xsd:integer, xsd:string from [XML Schema Datatypes]. 
>]]
>could read
>[[
>The datatype map of an OWL consistency checker MUST minimally support at 
least 
>xsd:integer, xsd:string from [XML Schema Datatypes], and for an OWL Full 
>consistency checker, also rdf:XMLLiteral.
>]]

The current text that you quote here is indeed misleading.
I would add here, in addition to OWL Full as you do, also OWL DL, in view 
of what I remarked today about 'solution 1' [1].

>
>Of the tests I believe that only misc-205 is an issue.
>This would need to be changed from being valid in Lite and Full (and 
hence 
>also DL), to being a Lite (and implicitly DL) but not Full test.
>
>If we were wanting to be totally minimalist then we could say that Herman 

>appears to be proposing a change to miscellaneous-205 of 
>s/Lite Full/Lite/
>and corresponding change to the Manifest file.

Tests 201 and 202 and 204 (already in the medium version of Test)
speak of the case that rdf:XMLLiteral is *not* supported.
It seems to me that this text (and thereby these tests) should also 
be adapted, to become consistent with what I describe as 
'solution 4' in [1], which appears to be the only viable solution
which makes S&AS and OWL Test consistent.

>
>Jeremy
>


Herman

[1] http://lists.w3.org/Archives/Public/www-webont-wg/2003Dec/0058.html

Received on Monday, 15 December 2003 12:55:49 UTC