W3C home > Mailing lists > Public > www-rdf-interest@w3.org > July 2000

Re: "Namespaces facility"

From: Ralph R. Swick <swick@w3.org>
Date: Wed, 05 Jul 2000 16:00:50 -0400
Message-Id: <200007052001.QAA26116@tux.w3.org>
To: Francois-Paul Servant <hypersol@micronet.fr>
Cc: www-rdf-interest@w3.org
At 09:01 PM 6/27/2000 +0000, Francois-Paul Servant wrote:
>As I understand M&S, only properties, not resources, have the 
>"Namespaces facility" :

In XML 1.0 plus namespaces only XML element names and attribute names
are defined to have [the possibility of] XML namespace qualifiers.

If I understand your question, you are wondering why the convenience of
namespace prefixes is not permitted within the attribute value of a property
element such as those typically found in a schema such as in Appendix A
of the RDF Schema specification:
http://www.w3.org/TR/2000/CR-rdf-schema-20000327/#xmlcore

 <rdf:Property ID="label"> 
   <rdf:type resource="http://www.w3.org/1999/02/22-rdf-syntax-ns#Property"/>
   ...

The answer is that until XML Schema came along there was no mechanism
to state in XML that an attribute value had a certain data type -- in particular,
that it was a QName or URI and is subject to namespace interpretation.

RDF 1.0 syntax could have defined the QName attribute value expansion itself,
and in fact the RDF Schema Working Group debated doing exactly this for quite
some time.  Ultimately, the RDF Schema WG was persuaded to defer this
design decision until after XML Schema was completed in order to be better
able to leverage whatever technology would be provided by XML Schema and --
perhaps more importantly -- to not define more application-layer syntax
that could diverge from the direction that XML Schema was to choose.

> if resources are sharing a same base URL 
>within an RDF file, this URL must be repeated in the file.
>
>This puzzles me. Nobody else ?

It is inconvenient to have to type the same [long] string over and over again,
I agree.  I will also caution that at an architectural level, namespaces are
much more than a short-hand for a common base URI and should not be
confused with URI abbreviation mechanisms.
Received on Wednesday, 5 July 2000 16:01:21 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:43 GMT