RE: [xml-dev] Rick Jelliffe quotable quote on the purpose of schemas

["Rick Jelliffe" <rjelliffe@allette.com.au>]

Klotz, Leigh said:
> This is much the approach we took with XForms.

Here's an example from the XForms spec, to be more concrete:

    <xforms:bind nodeset="/my:payment/my:number"
		 relevant="/my:payment/@method = 'cc'"
		 required="true()"
		 type="my:ccnumber"/>

> An initial prototype instance provides the structure, though it is
> application-specific and not generic to all possible classes of
> documents. We augment this system with the XSD type system and
> constraints; XPath location paths bind the constraints and XSD types to
> the nodes. Additionally we [1] profiled XML Schema, using only
> simpleTypes and no structures, and called the resulting system "XForms
> Basic" [2], so an XSD can define type libraries to be associated with
> nodes via XPath location paths.

So, XForms ditched the grammars from XSD, used paths instead, and kept the
datatypes. So I'm *not* the only one who sees the practicality of this :-)

I think the fact that path APIs are ubiquitous, easily just as widely
available as XSD APIs...probably more available is another attractive
point here: the grammar layer can be swapped out for a path-based layer in
many cases (certainly not saying all...swings and roundabouts) without
disruption.

> With the addition of full structural validation for XML Schema (i.e.,
> support for Part 1 Structures), it's full XForms 1.0.

I don't see any reference to Part 1 in
http://www.w3.org/TR/xforms/slice12.html#conform-levels-full
Any pointers to clarify what you mean?

Cheers
Rick