W3C home > Mailing lists > Public > public-swd-wg@w3.org > September 2008

RE: Draft CURIE review

From: Jeremy Carroll <jeremy@topquadrant.com>
Date: Thu, 11 Sep 2008 17:59:59 -0700
To: "'Jeremy Carroll'" <jeremy@topquadrant.com>, <public-swd-wg@w3.org>
Message-ID: <004e01c91472$e2fe1760$a8fa4620$@com>

A further aspect of the advantages of CURIEs over qnames came up today.
So I could enhance the personal comment as follows:
> 3. Personal Comment
> Our reviewer, Jeremy Carroll from TopQuadrant adds:
> Within the TopBraid Suite we put great emphasis on having human readable
> identifiers.
> We encourage the use of rdfs:label, but use QNames as a fall-back mechanism.
> As this new CURIE specification rolls out and possibly impacts languages still
> in development such as N3 and Turtle, this will allow us to improve the
> readability of the labels produced by our fall-back mechanism.

I suggest adding at this point:
A colleague has been working on "XML with Semantic Extenstions" and to avoid the
socially problematic sexml prefix was trying to use the prefix xmlse.
This was being silently corrected by Jena, which acts as a low level library to the 
TopBraid Suite, to j.0, since XML reserves all namespace prefixes starting in xml.
Since this gensym was not what was wanted, this has been, from the point of view
of the ontologist an annoying an inexplicable bug with TopBraid Composer.
In fact, all the tools were working correctly, except for the lack of an explanatory
message. The underlying problem is a mismatch between QNames as an XML syntactic device
and their use as abbreviated URIs in a tool such as the TopBraid Suite. This mismatch
will be resolved by uptake of this CURIE specification.

Received on Friday, 12 September 2008 01:01:39 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:07:54 UTC