[dsdl-discuss] Re: Part 10 Scenario

Eric

The decision made at the Washington meeting, bearing in mind that we had
received no input on Part 10 for over a year, was that we should abandon the
idea of doing our own standard and instead do a technical report that would
show how a number of existing pipelining languages could be used for
Validation Management. Which ones to use was not determined, though DPML and
JAXP were mentioned and are ones that I know who to ask for information
regarding. I would very much welcome an example in XPL if you think that can
handle the scenario. As far as I am concerned the more examples we can show
in the TR the better.

My proposed scenario could be misleading, as it could, I expect, be handled
directly in NVDL (but I need to have that proved). What I was trying to do
with the scenario was to work out how any of these languages can deal with
the case of different namespaces needing different processing and then
recombination in the original order. At present the only way I can see of
doong this is to split things up using NVDL and then develop some other
technique for determining how the parts recombine. I can't see myself how
what we have at present, or what has been proposed to date, can cope with
this scenario, which is very close to a "real-life" scenario I have (though
deliberately differentiated from it). Hence my setting it as a "test" for
Validation Management.

Martin

----- Original Message -----
From: "Eric van der Vlist" <vdv@dyomedea.com>
To: <dsdl-discuss@dsdl.org>
Sent: Tuesday, January 04, 2005 7:11 PM
Subject: [dsdl-discuss] Re: Part 10 Scenario

Hi Martin,

Le mardi 04 janvier 2005 à 18:59 +0000, Martin Bryan a écrit :
> I've now managed to put together a "scenario" that I think will test any
> pipelining language that we want to "adopt" for Part 10. I've asked a few
> people to suggest how things like Cocoon, DPML and JAXP would handle this
> scenario. Can you provide examples of other pipelining (or even Make type
> programs such as Ant) that could be used to manage data using this
scenario.
> Better still, can you think of a more demanding test for any candidate
> pipelining language that we could put into Part 10?

I am very confused by this scenario that goes against most of what I had
understood from previous meetings and decisions (and against the
arguments that had ruled out the previous proposal that I had called
"outie").

In particular, I thought we had decided that part 10 should not be a
pipeline language because pipeline languages were too procedural and
that part 10 shouldn't support splitting documents into parallel
validation since that was done the goal of NVDL.

Have you taken other orientations during the last meetings (that I am
shameful to have missed)?

If that's the case, rather than Cocoon or Ant, I'd suggest that we have
a look at Orbeon's XPL
(http://www.orbeon.com/ois/doc/reference-xpl-pipelines) that seems
perfectly able to do all what you're asking for.

Eric

>
> Martin

-- 
Have you ever thought about unit testing XSLT templates?
                                                     http://xsltunit.org
------------------------------------------------------------------------
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)
--
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)