W3C home > Mailing lists > Public > public-qt-comments@w3.org > July 2002

Re: Type as a sequence of QNames

From: Jeni Tennison <jeni@jenitennison.com>
Date: Tue, 16 Jul 2002 09:07:37 +0100
Message-ID: <125851000183.20020716090737@jenitennison.com>
To: public-qt-comments@w3.org, "Ashok Malhotra" <ashokma@microsoft.com>
CC: "Kay, Michael" <Michael.Kay@softwareag.com>

Hi Ashok,

> As far as I know, RELAX-NG is a Schema language that does validation
> but does not provide type information. There is, for example,
> nothing like the PSVI. Of course, it could be made to provide type
> info but that's a lot of work and what is the motivation for doing
> this work?

The reason any schema language provides type information is that it
can be used in downstream processing of the XML document. Currently,
there's no standard way of using that information at all; XPath 2.0
provides one. So the motivation for creating RELAX NG validators that
make available type information would be the ability for people using
XPath 2.0 to use it.

It's certainly true that it is a lot of hard work to enable RELAX NG
validators to create the data model currently required for XPath 2.0.
That's why I wondered whether a slightly different XPath 2.0 data
model would make that easier/possible and therefore XPath 2.0 less
schema-language specific.



Jeni Tennison
Received on Tuesday, 16 July 2002 04:07:45 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 16:56:43 UTC