[dsdl-discuss] Re: Part 10 Scenario

Hi Rick,

Le mercredi 23 février 2005 à 19:53 +1100, Rick Jelliffe a écrit :
> XPL has a lot in common with my Schemachine proposal
> http://www.topologi.com/resources/pdfs/Schemachine.pdf
>
> Both are pipelines, with input and output and top-level params.
> Both seem to work strictly XML in/XML out.

Yes. The current XPL implementation built into Orbeon PresentationServer
has extended that a little bit through a "binary
mode" (http://www.orbeon.com/ops/doc/processors-generators-url#d114290e290) which is very convenient to carry non XML stuff in pipelines, but that's basically base64 encoding within an XML root element.

> XPL provides for parameters using the same mechanism as inputs,
> Schemachine has a separate element. No big difference.
>
> Schemachine provides mechanisms for supporting passes, and
> phases. XPL does not seem to have a mechanism for allowing the
> failure of one stage to abort or cause the subsequent validation
> of other stages.
>
> XPL has a selection mechanism, for-each, which schemachine does
> not. If this were limited to a streaming untyped subset of Xpath,
> this would be a good addition.
>
> XPL allows inline documents/data. This seems a good innovation.
>
> So my take is that XPL has some worthwhile marginal improvements on
> Schemachine but is basically the same thing. However, it
> needs specific mechanisms to allow the results of one validation
> to influence subsequent steps (such as the grouping of phases
> and passes in Schemachine) to be as useful. The difference between
> a validation pipeline and a processing pipeline is that
> for validation pipelines are constructed so that a valid document
> is the *exception* case rather than the target case IYSWIM:
> locating problems then detecting their cause and diagnosing the
> fix is the purpose of the pipeline.

I agree with your technical analysis and I think that the main
difference is a difference of perspective.

When we've first made or proposals (your schemachine and my XVIF
followed by my "outie" proposals), we were in a logic where we wanted to
proposed something new focused on validation. These proposals have been
considered as "too procedural" and I the working group has asked me to
make a new proposal "more declarative".

Unfortunately, I haven't found the time that would have been needed to
carry on this task and the working group has decided to document how
existing pipeline mechanism could be used as our Validation Management
rather than to propose something from our own conception and, as far as
I understand the situation, that disqualifies schemachine as well as
that disqualifies my two proposals...

Now if we can, based on these proposals, made recommendations on how XPL
could be used or even define a subset of XPL that is well suited to
validation purposes, I think that this would be very useful.

Eric

>
> Cheers
> Rick Jelliffe
> --
> 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)
>
>

-- 
Don't you think all these XML schema languages should work together?
                                                         http://dsdl.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)