[
Lists Home |
Date Index |
Thread Index
]
- To: "Michael Champion" <michael.champion@hotmail.com>, <xml-dev@lists.xml.org>
- Subject: RE: [xml-dev] Two Questions - on XML Schema
- From: "Nathan Young -X \(natyoung - Artizen at Cisco\)" <natyoung@cisco.com>
- Date: Fri, 10 Mar 2006 12:19:10 -0800
- Thread-index: AcZDr3XeHRV9RNgQS8OnFfj0KVxdIwAz1TMg
- Thread-topic: [xml-dev] Two Questions - on XML Schema
Hi.
> > Oh dear. If the data binding folks are going to be stakeholders
> > in this, we can throw out all hopes of a clean design. They'll
> > mess it up even worse than the relational people did.
>
> I think this is an argument for letting Schematron do what it
> does well, and letting XSD do what it does .... uhh ...
> sortof tolerably after a couple of beers.
LOL!!!
> That is, whatever
> databinding contract you get from a schema, get from XSD; if
> you need path constraints in your actual data contract, do so
> with Schematron, and if you think you want both you are
> basically S.O.L. Vendors and standards orgs should focus on
> getting them to work together cleanly, not trying to get one
> to subsume the other.
I heard someone mention (and I can't remember where or when... but it's
not my original idea) the idea of having standards for a validation
warning/error format. When validation was done on an instance, it would
result in a document snippet that represented the results of that pass
of validation. Snippets could be concatenated from several validation
steps, and the result would be a coherent report on the validity of the
instance.
Is something along those lines what you mean when you say "getting them
[schematron and XSD] to work together cleanly"?
-------------->Nathan
>
>
>
>
> ________________________________
>
>
|