words (RE: extensibility in XSchema?)

Peter Murray-Rust peter at ursus.demon.co.uk
Tue Jun 23 00:20:28 BST 1998


At 16:29 22/06/98 -0500, W. Eliot Kimber wrote:
>At 07:51 PM 6/22/98, Peter Murray-Rust wrote:
[...]

>You could try:
>
><?XML version="1.0"?>
><?IS10744 arch 
>  name="foo"
>  public-id="+//IDN me.com//NOTATION Foo Architecture//EN"
>  dtd-system-id="foo.dtd"
>  doc-elem-name="foo"
>?>
><foo bar="baz"/>
>
>Which at least tells you that the document is governed by the "Foo
>Architecture", which you could, given some appropriate infrastructure
>(which might be a mapping from architecture public IDs to MIME types if
>such a thing made sense--it's no more silly than mapping extensions to MIME
>types, and at least uses names that can be made universally unique),
>dereference to various definitional documents and formalisms.  The pointer
>to the architectural DTD lets you syntactically validate the document using
>that DTD using a parser like SP and apply normal XML processing to the
>architectural instance.

You and I are tackling different problems, I think. I *think* I understand
what AFs are for - it allows you to identify one (or more) elements and
their relation to one (or more) other elements. IOW when your document is
mapped onto a meta-DTD is says - FOO in doc maps to BAR in the meta-DTD.
Since we 'know' what BAR means, we have now associated its meaning with FOO.

What I am asking for is different. I 'know' what <FOO> is. I simply want to
map it onto a program. NOT a piece of prose. The program will define the
semantics/behaviour of the element. [Before you all say this is awful, it's
a tremendous advance.] Example. If I get:
	<MathML:EQN>really hairy eqn</MathML:EQN>
I do not want to be referred to the MathML spec. I want some software to
process this for me. I don't necessarily want it rendered. It might be
spoken, as in T.V.Raman's wonderful work. It might be differentiated... etc. 
	Suppose I want to send this equation to Mathematica - or Maple. How am I
going to do it? If I get an image/gif, it gets processed automatically. If
I get application/mathematica it gets processed automatically (if you have
configured your mailcap file). But if I get text/xml with <MathML> in the
middle I can't do anything, except read the prose of the MathML spec. 
 >
>Remember that, by definition, the external ID of a notation is intended to
>get you to the definition of that notation, in the same way that a MIME
>type gets you to the official definition of the type.

Understood - I think. 
>
>I see you as asking for two things:
>
>1. A way to unambiguously relate a document to a governing semantic
>definition.

Yes - though this isn't my main problem. I can accept that either AFs or
PIs or RDF can map a text document onto the NS URN.

>2. A way to get the parts of that definition in some predictable ways.

Don't understand this.

>
>Certain the PI shown above satisfies requirement 1. I'm not sure that
>requirement 2 can be satisfied in a completely general way, but certainly
>one could define domain-specific conventions or standards ala MIME
>processing (e.g., a registry of definition documents in some agreed upon
>form (which might be simply "prose")).

I don't need the registry, though it might help.

What we do in molecular science - and this is Henry's brainwave - is simply:

(a) identify the document as (say) chemical/x-pdb (there is NO registry for
this)
(b) set a mapping in mailcap as:
	chemical/x-pdb: /usr/bin/rasmol
(c) sit back and enjoy the hyperactive molecules.

This method is used tens of thousands of times a day and is extremely useful.

Its (b) I'm asking for with XML. e.g:

	<Namespace ns="urn:www.xml-cml.org">
	<Element language="Java" Type="Molecule">jumbo.cml.MOLNode</Element>
	</Namespace>
This maps an XML namespace onto a Java class. It works. I just need an
official syntax :-)

	P.



>
>Cheers,
>
>E.
>--
><Address HyTime=bibloc>
>W. Eliot Kimber, Senior Consulting SGML Engineer
>ISOGEN International Corp.
>2200 N. Lamar St., Suite 230, Dallas, TX 95202.  214.953.0004
>www.isogen.com
></Address>
>
>xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
>Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
>To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
>(un)subscribe xml-dev
>To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
>subscribe xml-dev-digest
>List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)
>
>
Peter Murray-Rust, Director Virtual School of Molecular Sciences, domestic
net connection
VSMS http://www.nottingham.ac.uk/vsms, Virtual Hyperglossary
http://www.venus.co.uk/vhg

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)




More information about the Xml-dev mailing list