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

Re: OWL Syntax

From: Jeremy Carroll <jjc@hpl.hp.com>
Date: Thu, 13 Feb 2003 05:40:10 +0100
To: www-webont-wg@w3.org
Message-Id: <200302130540.10185.jjc@hpl.hp.com>

> >The abstract syntax undermines the I18N requirements agreed by the WG.

> This has been mentioned in a number of emails, but it doesn't seem
> to correspond to anything that I could find in the Requirements or 
> Issues documents.  Can someone explain how to dereference the I18N 
> identifier?

Sorry for speaking in code:
118N = I 18letters N = internationalization

The specific requirements are the last three in the requirements doc:
i.e.
R16. User-displayable labels
R17. Supporting a character model
R18. Supporting a uniqueness of Unicode strings

There is also the goal

http://www.w3.org/TR/webont-req/#goal-internationalization

The specific features of RDF that support these requirements and the overall 
goal are:
+ the use of xml:lang, which for historical reasons becomes embedded within 
literals in RDF
+ the use of rdf:parseType="Literal" to permit embedded xhtml which in turn 
supports a richer variety of languages than simple strings (e.g. 
bidirectional text, accessability support in labels, ruby annotation for 
japanese labels)
+ the use of Unicode Normal Form C in both literals and URI references

AS&S appears to deliberately avoid these fundamental building blocks from RDF 
- not supporting xml:lang, 
- prohibiting rdf:XMLLiteral (a.k.a. rdf:parseType="Literal"), 
- treating unicode strings as wholly unproblematic (ignoring the extensive 
work by the W3C in this area), 
and hence "undermines the I18N requirements agreed by the WG"

Jeremy
Received on Wednesday, 12 February 2003 23:39:28 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:57:57 GMT