xml-dev - ANN: XSet an XML property set

[Jonathan Borden <jborden@mediaone.net>]

Tim Bray wrote, a few days ago:

>
> At 07:28 PM 25/07/00 -0400, Simon St.Laurent wrote:
> >At 05:46 PM 7/25/00 -0400, Jonathan Borden wrote:
> >>XML is a serialization of a logical document structure defined
> by the XML
> >>Infoset.
> >
> >WARNING: 'XML is a serialization' is only one way to look at it.
> >
> >XML is a syntax, and the 'logical document structure' is a particular
> >interpretation of that syntax is another.
>
> What Simon said.  XML took a lot of static in its early days because it
> was "just syntax" - there are certainly a lot of people who want to think
> only in terms of object models (groves, DOMs, whatever) and see the syntax
> as disposable fluff.  Me, I think syntax is crucial.  Because describing
> data structures in a straightforward, interoperable way is really hard to
> get right and very often fails.  At the end of the day, if you
> really want
> to interoperate, you have to describe the bits on the wire.  That's what
> XML does.

I make a distinction between XML the character stream (which I term a
serialization) and XML the character stream ***as interpreted by XML 1.0***.
A full fidelity XML property set, also defines the "bits on the wire" but in
a logical sense.

Specifically, it is the productions in http://www.w3.org/TR/REC-xml which
formally specify the abstraction of the XML document. To say this another
way, the XML "grove" is defined by the XML "property set" which consists of
these productions.

Now we all understand that from the abstractions defined in XML 1.0 we can
layer other abstractions and that is the whole point of what I am saying. I
am talking about a foundation from which to derive particular abstractions
of an XML 1.0 document, e.g. XML Infoset, Common XML, SAX, DOM etc.

>
> Not that SAX and the DOM and so on aren't good things.  But at the
> end of the day, de facto and de jure, XML is syntax. -Tim
>

I agree. To make my point, I am willing to put my money where my mouth is
and define an XML property set. I believe the XML 1.0 productions
(http://www.w3.org/TR/REC-xml) serve as a proper foundation to describe the
XML syntax. That's what this property set defines: syntax.

The URI http://www.openhealth.org/XSet/xml.xml

is a *very very* rough sketch of an XML property set describing the XML 1.O
production rules in RDF. This property set, when completed, will define the
XML grove in RDF.

Plan: 1) Complete the rest of the XML 1.0 productions
	2) Refine the production rule description language if necessary
	3) Incorporate XML namespace productions
	4) Define a language with which to subset XSet (current candidate XSLT)

Goals:
	1) To get it "right"
	2) to produce an subset (?XSLT "grove plan") which transforms XSet into the
XML Infoset WG Appendix D, RDF Schema (i.e. define XML Infoset as a proper
subset of XSet).
	3) To produce a subset defining SAX
	4) To produce subsets defining Common XML, SML

Help and comments are actively requested.

Jonathan Borden
The Open Healthcare Group
http://www.openhealth.org