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

RE: N-Triples VS RDF/XML bNode identifiers

From: Jeremy Carroll <jjc@hpl.hp.com>
Date: Wed, 29 Oct 2003 09:58:25 +0100
To: w3c-rdfcore-wg@w3.org
Message-Id: <200310290958.25111.jjc@hpl.hp.com>

On Tue, 28 Oct 2003, pat hayes wrote:

> Well, OK, but guys, he has a point. Why not just tweak Ntriples to
> allow all of ASCII? That wouldn't break any of our stuff and it would
> be generally good-citizen-ish.

I find the argument underwhelming.

The current situation is that the possible identifiers in Ntriple are a proper 
subset of those in RDF/XML.

The proposed new situation is that the possible identifiers in Ntriple are a 
proper subset of those in RDF/XML.

If you are generating identifiers and wish them to be interoperable between 
RDF/XML and NTriple it can be done in either case (generate Ntriple ids)

If you are reading in RDF/XML and wisht to use the identifiers in Ntriple it 
can't be done in either case. (You need to do some work).

We should only change things that are broken - I am not even convinced this is 
an improvement.

Jeremy
Received on Wednesday, 29 October 2003 03:58:52 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 29 October 2003 03:58:55 EST