W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > May 2003

RE: Languageless Typed Literals

From: <Patrick.Stickler@nokia.com>
Date: Mon, 5 May 2003 13:51:25 +0300
Message-ID: <A03E60B17132A84F9B4BB5EEDE57957B5FBBB4@trebe006.europe.nokia.com>
To: <jjc@hplb.hpl.hp.com>, <w3c-rdfcore-wg@w3.org>



> -----Original Message-----
> From: ext Jeremy Carroll [mailto:jjc@hplb.hpl.hp.com]
> Sent: 05 May, 2003 13:20
> To: Stickler Patrick (NMP/Tampere); jjc@hplb.hpl.hp.com;
> w3c-rdfcore-wg@w3.org
> Subject: RE: Languageless Typed Literals
> 
> 
> Jeremy:
> > Option 1:
> > XMLLiteral ceases to be a typed literal but we revert to the old
> > treatment  where it was simply a special.
> 
> 
> Patrick:
> > My strong preference is then for option 1, reverting (in a sense)
> > XML literals to the M&S definition.
> >
> > This has the additional benefit that lexical forms can be left
> > as-is in the graph per the RDF/XML serialization and only need be
> > canonicalized when testing for equality.
> >
> > Thus, plain and XML literals both may take lang tags and neither
> > are typed literals nor fall within the scope of RDF datatyping.
> >
> > Typed literals do not take lang tags, period.
> >
> > This avoids all the headaches relating to the bizzare datatype
> > rdf:XMLLiteral.
> >
> > Patrick
> >
> >
> 
> The old treatment was in say:
> http://www.w3.org/TR/2002/WD-rdf-concepts-20020829/

That looks right. Along with reverting the NTriples to the
XML"..."@nn syntax.

> I think that the reagle issue resolutions would in the main 
> stay, and the
> canonicalization would still be specified in the syntax, but with the
> implementation note that makes it clear that they "only **need** be
> canonicalized when testing for equality."

Right.

> I have three concerns about this option:
> 
> a) we had comments
> http://lists.w3.org/Archives/Public/www-rdf-comments/2002JulSep/0092.html
> linking to
> http://www.w3.org/2002/07/29-rdfcadm-tbl.html#xtocid103643

The key argument here is that RDF applications shouldn't have to have
XML infoset processors in order to compare XML literals. I don't see
how the proposed change affects that. If one is going to compare XML
literal values, one must canonicalize them. That is true whether the
values are treated as datatype values or built in XML literal values.

So the requirements on RDF processors appear to be equivalent whether
we treat XML literals as typed literals or M&S like XML literals.

> and
> 
> http://lists.w3.org/Archives/Public/www-rdf-comments/2002JulSep/0165.html

If I understand what is being said here (and I'm not 100% sure I do), the
key concern relevant to option 1 is about why we would have a special
type of literal to handle XML literals rather than just a built-in datatype.

The answer (now, presuming we nuke lang tags in typed literals) is that
per M&S, XML literals can be qualified by lang tag just as can plain 
literals and the lang tag is significant to equality tests, so XML literals
can't be addressed by a datatype, because typed literals don't take lang
tags.

And our charter (probably) precludes making lang tags irrelevant for
XML literals.

Thus, both plain and XML literals remain pretty much as defined in M&S,
with necessarily clarifications regarding canonicalization, etc.

> both of which would need resurrecting, since we have followed up saying that
> we have changed in the way they sort of wanted.

Right, but I think we know what our responses would be, if we
went with option 1.

The question is whether the persons submitting the comments would be
agreeable to the revision.

> b) how difficult would it be for Pat to go back and rework

Good question. Pat?

> c) impact on OWL support for XML Literals - webont are generally negative
> about them, the more work they have to do, the less support there will be in
> OWL for these.

In a sense, we're not really changing the bulk of what is said about
XMLLiterals:

1. the RDF/XML syntax remains the same
2. the NTriples syntax is analogous to the the present representation 
   (just moving a datatype URI 'rdf:XMLLiteral' at the end to an 'XML'
    flag at the start)
3. canonicalization and equality tests remain the same
4. the semantics essentially remain the same in that one has a pair
   of lexical form and (optionally absent) lang tag mapping to an
   XML literal value

It's simply the disassociation of XML Literals with the RDF Datatyping
machinery that needs to be done -- and doing so notably simplifies
RDF Datatyping (which I would presume OWL would be happy about).

Patrick
Received on Monday, 5 May 2003 06:51:31 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:57:26 EDT