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

Semantics review: XMLLiteral

From: Jeremy Carroll <jjc@hpl.hp.com>
Date: Wed, 18 Jun 2003 17:57:14 +0300
To: w3c-rdfcore-wg@w3.org
Message-Id: <200306181757.14596.jjc@hpl.hp.com>

This is a review of:


(no apparant unique-id)

My review is in a number of parts.
This part is restricted to XMLLiteral.

I start with a question: your treatment of the relationship between URIs and 
datatypes is somewhat more thorough than that inc concepts.

Are the following two phrases from concepts OK or should they be changed:


"A datatype is identified by a URI."

concerning XMLLiteral

"The datatype URI
is http://www.w3.org/1999/02/22-rdf-syntax-ns#XMLLiteral."

possible change to a) is e.g.
"A datatype may be identified by one or more URI references"

possible change to b) is e.g.

"A datatype URI reference
is http://www.w3.org/1999/02/22-rdf-syntax-ns#XMLLiteral."
"The preferred URI reference 
is ..."


The value space of XMLLiteral is not XML documents but exclusive Canonical XML 
(see latest concepts editors draft) ... you've done pretty well given the 
lack of clarity I have given you.

Suggest global substitute 
  s/XML document/XML data/
(problem with actual making this change is if we are going to go back and 
accept I18N desire to keep the lang-tag - the easiest way to do that now 
would be in RDF/XML Syntax, which would then make it go back from XML data to 
XML document - I hope we can close this on Friday).

Also the reference

Canonical XML, Boyer, J. W3C Recommendation, March 2001.

should be replaced by

Exclusive XML Canonicalization Version 1.0, J. Boyer, D.E. Eastlake 3rd, J. 
Reagle, Authors/Editors. W3C Recommendation. World Wide Web Consortium, 18 
July 2002. This version of Exclusive XML Canonicalization is 
http://www.w3.org/TR/2002/REC-xml-exc-c14n-20020718. The latest version of 
Canonical XML is at http://www.w3.org/TR/xml-exc-c14n. 


I will detail this global substitution, and other points in order in the file, 
searching for XML, also refering to my paper copy with scribbles:

a) in change log last line, suggest s/XML documents/XML data/
b) 0.2 first para
s/XML document corresponding to/XML data corresponding to/
and change link from #section-XMLLiteral to
(if you agree, I will add at that definition -
  "(This is referred to as the *XML data corresponding to* the string)."
c) end of 1.2 @@ text
[[which is considered to denote exclusive canonical XML 
[XML-XC14N] (hereafter "XML data")
corresponding to the string in the typed literal;]]

d) section 3. third para s/XML document/XML data/ (three times)

Note: following the links to exclusive canonical XML makes it clear that 
exclusive canonical XML is octet sequences, not sequences of character codes, 
so we could decide to make a call concerning this agnosticism and go for them 
being different. I prefer not to.

e) section 3 in green box slightly lower down s/XML document/XML data/

f) suggest
"mis-typed XML literals" should be "ill-typed" (section 3 para after purple 

g) section 5 para above purple box

 "contain <rdf:XMLliteral, x> where x is the built-in XML datatype"

suggest "built-in XML Literal datatype"

h) Appendix C - note I have not read this yet

s/XML document/XML data/ twice

i) section 5 first para
"It also includes a single built-in datatype XMLLiteral."


Received on Wednesday, 18 June 2003 11:57:21 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:54:06 UTC