[dsdl-discuss] Re: FW: DSDL: Physical validation

Hi,

On Tue, 2004-09-28 at 12:38, Martin Bryan wrote:
> The following message captures user requirements identified as a result of
> Alex Brown's presentation of DSDL at XML Open last week. These requirement
> should be reviewed by all Part editors to identify which ones they
> can/should be able to meet.

As far as Part 10 is concerned, the impact would be that if all these
constraints were to be enforced, Part 10 should mandate that the
mechanism used to convey the XML between the different processors and
validators is able to carry enough information so that they can be
checked.

Depending on how far we want to go in the validation of the "*physical*
structure", that could exclude the possibility to transfer information
between processors using SAX events or DOM trees and impose using actual
character streams or at least data or event models that are not the
common ones that everyone uses.

If we pushed the idea to the extreme, that might require that Part 10
use custom parsers.

I think that this would on one hand be very useful for some applications
but on the other hand, that could also make Part 10 so heavy to
implement that it could loose a lot of attractiveness for common
applications.

This kind of validation seems quite useful to have as a standalone tool,
but including it in the DSDL framework would have an impact on the
architecture (it's like including an oliphant in a sheepfold : that'll
have an impact on the architecture of the sheepfold that'll probably
become much less convenient for sheep).

Eric (having very mixed feelings about that proposal)

-- 
Lisez-moi sur XMLfr.
                       http://xmlfr.org/index/person/eric+van+der+vlist/
------------------------------------------------------------------------
Eric van der Vlist       http://xmlfr.org            http://dyomedea.com
(ISO) RELAX NG   ISBN:0-596-00421-4 http://oreilly.com/catalog/relax
(W3C) XML Schema ISBN:0-596-00252-1 http://oreilly.com/catalog/xmlschema
------------------------------------------------------------------------
--
DSDL members discussion list
To unsubscribe, please send a message with the
command  "unsubscribe" to dsdl-discuss-request@dsdl.org
(mailto:dsdl-discuss-request@dsdl.org?Subject=unsubscribe)