W3C home > Mailing lists > Public > www-webont-wg@w3.org > July 2003

Fwd: RDF XMLLiterals reagle-01,-02

From: Jeremy Carroll <jjc@hpl.hp.com>
Date: Wed, 2 Jul 2003 10:58:00 +0300
To: www-webont-wg@w3.org
Message-Id: <200307021058.00943.jjc@hpl.hp.com>

In another role, I've just sent this message to Jim, concerning XMLLiterals in 
RDF, and the webont last call comment.

I believe this is now ready for WebOnt to respond.

Personal comment:
There is current conflict between RDF Core and I18N about whether XMLLiterals 
should or should not have language tags from the outer scope e.g.

<rdf:Description xml:lang="it">
  <eg:prop rdf:parseType="Literal">
 ho detto, <em>mammamia</em>

RDF Core's recent decisions do not include the xml:lang="it" on the 
xmlliteral, so there is no indication that ho detto mamamia is italian.  The 
desired effect should be written more like:

  <eg:prop rdf:parseType="Literal">
<span xml:lang="it"> ho detto, <em>mammamia</em>

It would, at least IMO, be helpful if webont specifically commented on this 
point of contention. (If we have an opinion)


----------  Forwarded Message  ----------

Subject: RDF XMLLiterals reagle-01,-02
Date: Wed, 2 Jul 2003 10:50:11 +0300
From: Jeremy Carroll <jjc@hpl.hp.com>
To: www-rdf-comments@w3.org, eric@w3.org, reagle@w3.org, hendler@cs.umd.edu

You earlier commented on the RDF Core Last Call working drafts, particularly
on the design of XMLLiterals.

This message concerns those comments, listed as:
"Confusion about wrapping of XMLLiteral"
which did not get an issue number assigned.

We have modifed the decisions reflected in our earlier reply:

While contemplating various last call issues RDF Core WG has simplified the
design of typed literals, particularly XMLLiteral, by removing the language

This permits addressing:
"Confusion about wrapping of XMLLiteral" by simply removing the wrapper

The resolution to modify the design was decided on 09th May 2003
and reads:
RESOLVED: Typed literals option 4 from msg 0086

which refers to

> Option 4:
> Language tag is simply dropped from all typed literals including
> rdf:XMLLiteral

  Concepts is changed to say that a literal can have either a datatype or a
language tag and not both.
  rdf:XMLLiteral datatype is changed to have the identity as its lexical
value mapping (no wrapping), with consequential change to the value space of
  Other editors to make consequential changes.
we further clarified (20th June) that "the identity" was over-simplistic, and
the value space remains a set of exclusively canonicalized XML,

The new text defining XML Literal can be found in the editors draft at:

Please note that the I18N-WG have made a comment criticizing this decision:
which the RDF Core has not officially yet agreed a response (although we have
spent more than an hour of telecon time on it, with still a majority opposing
further changes).

I will update you on any further changes (well I hope not!).

Please let us know, copying www-rdf-comments@w3.org if:
- these changes make the previously confusing decisions acceptable
- if you find these changes address
"Confusion about wrapping of XMLLiteral"
in an acceptable fashion

Eric, WebOnt:
- if we have acceptably addressed your comments:


Received on Wednesday, 2 July 2003 04:58:14 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:56:54 UTC