W3C home > Mailing lists > Public > public-webont-comments@w3.org > February 2004

Re: OWL DL, rdfs:Literal as range value, Reference vs. S&AS

From: Guus Schreiber <schreiber@cs.vu.nl>
Date: Tue, 10 Feb 2004 14:07:26 +0100
Message-ID: <4028D78E.8060005@cs.vu.nl>
To: Benjamin Nowack <bnowack@appmosphere.com>
Cc: webont-comments <public-webont-comments@w3.org>

Benjamin Nowack wrote:

> 
> Hi,
> 
> working on an OWL editor/species serializer I'm a little
> bit confused about using rdfs:Literal for DatatypeProps. The
> S&AS doc [1] mentions
> 
> [[
> dataRange ::= datatypeID | 'rdfs:Literal'
> ]]
> 
> but in the reference doc [2] the rdfs:Literal option was removed
> from the WD in the august 2003 CR version (After comments from 
> RDF Core IIRC). The "Parsing OWL in RDF/XML" note [3] seems to 
> ignore rdfs:Literal as well, but I'm not sure on that.

Benjamin,

Thanks for your comment.

There is a distinction in OWL between a data range and a datatype. As 
RDFCore pointed out in their review of Reference, rdfs:Literal is not a 
datatype. So, it was removed from the list of datatypes but added to the 
list of valid data-range specifications, see start of Sec. 6:

[[
  OWL allows three types of data range specifications:

     * A RDF datatype specification.
     * The RDFS class rdfs:Literal.
     * An enumerated datatype, using the owl:oneOf construct.
]]

This is in line with the S&AS grammar rule that you quote (which is the 
OWL Lite one, so it leaves out the enumerated option)..

> 
> 
> Furthermore, the validators from
> Manchester [4] and Pellet [5] both complain:
> 
> Mindswap, Pellet:
> "Non OWL-DL features used
> http://www.w3.org/2000/01/rdf-schema#Literal is an unknown datatype"
> 
> Manchester/Karlsruhe, OWL Ontology Validator
> "Undefined Datatype: http://www.w3.org/2000/01/rdf-schema#Literal"
> 
> The vOWLidator at http://owl.bbn.com/validator/ is fine but I don't
> know if it does species checking.
> 
> I guess it's too late for doc changes but thought this issue
> might be of interest. (will send a copy to www-rdf-logic as well).
> Maybe this is not an issue at all (a rdfs:Literal range for a 
> datatype property doesn't really add information..), but I wonder
> how to "translate" an rdf:Property with rdfs:range=rdfs:Literal to
> an owl:DatatypeProperty with rdfs:range=??. Simply drop the range?
> Or can I keep it because the validators are just incomplete and 
> the rdf/xml is DL-ok?

You do not have to drop the range, as rdfs:Literal is a valid value for 
a range specification of an OWL property.

Hope this helps,

Guus
> 
> 
> thanx in advance,
> benjamin
> 
> --
> Benjamin Nowack
> 
> Kruppstr. 82-100
> 45145 Essen, Germany
> 
> 
> [1] http://www.w3.org/TR/owl-semantics/syntax.html
> [2] http://www.w3.org/TR/owl-ref/#Datatype
> [3] http://www.w3.org/TR/2004/NOTE-owl-parsing-20040121/
> [4] http://phoebus.cs.man.ac.uk:9999/OWL/Validator
> [5] http://www.mindswap.org/2003/pellet/demo.shtml
> 

-- 
Free University Amsterdam, Computer Science
De Boelelaan 1081a, 1081 HV Amsterdam, The Netherlands
Tel: +31 20 444 7739/7718
E-mail: schreiber@cs.vu.nl
Home page: http://www.cs.vu.nl/~guus/
Received on Tuesday, 10 February 2004 08:09:37 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:16:01 UTC