W3C home > Mailing lists > Public > www-webont-wg@w3.org > November 2001

RDF and datatypes

From: Jeremy Carroll <jjc@hplb.hpl.hp.com>
Date: Mon, 19 Nov 2001 16:42:26 -0000
To: <www-webont-wg@w3.org>
Cc: <bwm@hplb.hpl.hp.com>
Message-ID: <JAEBJCLMIFLKLOJGMELDEEIBCCAA.jjc@hplb.hpl.hp.com>


Summary: can WOW give feedback on the 'S' datatyping proposal being made in
RDF Core WG?

RDF Core WG & Datatypes

The RDF Core WG has been considering how to incorporate XML Schema datatypes
[4] into RDF.

This is the one *change* to RDF that is in-charter for the current WG.
(Slightly oversimplifying charter issues).

At the last RDF Core telecon [1], I took an action of explaining to WOW the
current favourite proposal and getting feedback.

Please can other joint members of RDF Core WG and WOW *not* start a second
round of discussions here; primarily I am trying to get feedback from WOW
members who are *not* in RDF Core.

Note daml already has a way of using XML schema datatypes; the mechanism
proposed here is *different*.

RDF Core WG is expecting to have a round of "implementators feedback" before
making any decisions. Hence while timely feedback is helpful, nothing is
even beginning to set at this stage.

The S Proposal

So here goes, the following is known as the 'S' datatyping proposal. The
original proposal was made in [2], and the overview of all the proposals is
found in [3]. 'S' is currently the clear favourite within the RDF Core WG.
No decisions have been made.

As in XML Schema [4] a datatype is a mapping from a lexical space to a value

In 'S' a typed datum is represented as an arc in the graph. One node is a
blank-node that represents the datum in the value space; the other end of
the arc is a node labelled with the string that is a lexical representation
of that value. The edge is labelled with the datatype name.

So in a N-Triple-like notation:

<aaa> <eg:prop> _:x .
_:x <xsd:integer> "10" .

equivalently in RDF/XML:

<rdf:Description rdf:about="aaa">
   <eg:prop xsd:integer="10" />

shows that <aaa> has a property eg:prop whose value is found in the value
space of xsd:integer and has lexical form being the String "10".

Thus, datatype information is expressed locally adjacent to each instance.
Also datatype information is a property rather than a class. This draws a
distinction between datatype and class.


In contrast DAML says:

Datatype values are written in a manner that is valid RDF syntax, but which
is given a special semantics in DAML+OIL. The preferred method is to give a
lexical representation of the value as a string, along with an XML Schema
datatype that is used to provide the type of the value as well as the
parsing mechanism to go from the string to the value itself. The XML Schema
datatype is the rdf:type of the value, and the lexical representation is the
rdf:value of the value. So the decimal 10.5 could be input as <xsd:decimal
rdf:value="10.5"> provided that xsd was defined as the URI of the XML Schema
Datatype specification.

As a nod to backward compatability, literals that occur outside this sort of
construction are interpreted as any of the XML Schema Datatype values with
this lexical representation. These values are mostly unusable unless some
typing information is available, such as a range for a property.

The question of whether any XML Schema datatype can be used in such
constructions, or whether only certain XML Schema dataypes can be so used
(such as only the predefined datatypes), remains open.


Is there support for 'S' amongst WOW?
Do WOW members find it hard to live with 'S'?
If RDF Core adopt 'S' would WOW continue to go with DAML datatyping or adopt
an RDF recommended datatyping? [OK I know that that is too hard at this


HP Labs
HP Rep to RDF Core and WebOnt

[1] Graham Klyne "Minutes 2001-11-16"
[2] Sergey Melnik "suggestions for datatyping (long)"
[3] Pat Hayes, "DATATYPES: mental dump."
[4] W3C "XML Schema Part 2: Datatypes"
[5] Frank van Harlem, Peter F. Patel-Schneider and Ian Horrocks, "Reference
description of the
DAML+OIL (March 2001) ontology markup"
Received on Monday, 19 November 2001 11:45:45 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:56:40 UTC