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

Re: The case for untidy literal semantics

From: Jeremy Carroll <jjc@hpl.hp.com>
Date: Tue, 24 Sep 2002 21:36:18 +0200
To: w3c-rdfcore-wg@w3.org
Message-Id: <200209242136.18592.jjc@hpl.hp.com>


While Brian's summary is helpful, I keep wondering whether we are comparing 
the worst that came out of the tidy camp, with the worst that came out of the 
untidy camp.

Personally I like the syntactic transform on input, which adds a bnode into 
the graph in the place of each untyped  literal, and then another arc. The 
resulkting graph is understood with tidy literal semantics. The RDF/XML doc 
now has untidy literal semantics.

It's pretty minimal work for the tool builders; there is a difficult 
transitional phase for RDF apps built on top of the tools. Tool builders may 
need to support transitional deprecated APIs that make the bnode-arc-literal 
combination just look like a literal.

Mandating the transform and then tidy semantics should make it relatively easy 
to ensure that the community does the same thing (rather than variations on a 
theme).

Jeremy
Received on Tuesday, 24 September 2002 15:37:51 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:51:02 EDT