W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > March 2002

RDF & XML Literals: Intro

From: Jeremy Carroll <jjc@hplb.hpl.hp.com>
Date: Mon, 4 Mar 2002 12:43:10 -0000
To: <w3c-rdfcore-wg@w3.org>, <w3c-ietf-xmldsig@w3.org>
Message-ID: <JAEBJCLMIFLKLOJGMELDGECKCDAA.jjc@hplb.hpl.hp.com>

The RDF specification shows how labelled graphs can be encoded in XML.

e.g. the graph

http://lists.w3.org/Archives/Public/www-archive/2001Jun/att-0021/00-part#38

is encoded as

http://lists.w3.org/Archives/Public/www-archive/2001Jun/att-0021/00-part#65




It permits the use of well-balanced fragments of XML as one sort of label,
and refers to these as "XML Literals".

A typical example is:

<rdf:Description
  xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
  xmlns:dc="http://purl.org/metadata/dublin_core#"
  xmlns="http://www.w3.org/1999/xhtml"
  rdf:about="http://example.org/papers/paper1">
  <dc:Title rdf:parseType="Literal"><!-- Relevant text start. -->
    Foo<em>bar</em>
  <!-- Relevant text end. --></dc:Title>
</rdf:Description>

Many users may omit the default xmlns declaration.

It is unclear precisely what the label in the RDF graph is.
There is agreement that the relevant text in the input file lies between the
two comments.

Some RDF APIs provide a string, others provide a means of accessing the SAX
events.
This is the content of the issue
http://www.w3.org/2000/03/rdf-tracking/#rdfms-literal-is-xml-structure

As will be familiar to the C14N people, there are specific issues to do with
namespace processing.

These are identified in:
http://www.w3.org/2000/03/rdf-tracking/#rdfms-xml-literal-namespaces
and also present in an example in the spec.
http://lists.w3.org/Archives/Public/www-archive/2001Jun/att-0021/00-part#257

Other problems relate to inconsistent treatment of character references etc.
which, for instance, the HP team found problematic when upgrading Jena to
work with ARP instead of RDFFilter. The two parsers had taken different
choices with respect to the exact representation of xml literals and the
regression tests then failed.

A large part of RDF Core WG's work has been making RDF more formal.
Particularly the RDF test cases and RDF model theory depend upon a well
defined equality over node labels.

These goals are in conflict with the current spec. that explicitly ducks the
issue of equality.

http://lists.w3.org/Archives/Public/www-archive/2001Jun/att-0021/00-part#220
[[[
This specification does not state a mechanism for determining equivalence
between literals that contain markup, nor whether such a mechanism is
guaranteed to exist
]]]
RDF Model Theory
http://www.w3.org/TR/rdf-mt/
RDF Test Cases
http://www.w3.org/TR/2001/WD-rdf-testcases-20010912/


Hence, a resolution of these issues will give:
- a set of representation for XML Literals,
- a means for turning a specific well balanced XML fragment into such a
representation
- a means for comparing two such representations for equality

The RDF Core is currently agreed that C14N will be the basis of addressing
these issues.
Received on Monday, 4 March 2002 07:43:22 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:46:13 EDT