RE: [xml-dev] Has XML run its course?

["Champion, Mike" <Mike.Champion@SoftwareAG-USA.com>]

> -----Original Message-----
> From: Pete Kirkham [mailto:pete@cafemosaic.co.uk]
> Sent: Sunday, September 30, 2001 5:19 PM
> To: Didier PH Martin
> Cc: xml-dev@lists.xml.org
> Subject: Re: [xml-dev] Has XML run its course?
> 
> 
> Didier PH Martin wrote:
> > ... This is what is happening. There is no more XML per se, but a 
> > plethora of languages based on it. So, the next stage is that we 
> > hear about applications created with XML based languages. ... 
> 
> What did you see XML as?

> Certain of the application languages, such as XSL, XSchema, DocBook, are
> understood to be general accepted standards.  Others are generated for
> single use cases.

I think the question is "how far can you go with some chunk of XML data with
generic XML tools and knowledge?"  Text formats (DocBook, NewsML, etc.) are
probably at one extreme:  if I know XML, know the DTD/schema, have a text
editor and parser/validator, I can do pretty much whatever can logically be
done with an instance.  

SOAP might be in the middle: I need all of the above, but also need to read
the SOAP spec to know that I can't put a DTD internal subset or PIs in the
instance; I also need to know the details of not only the syntax but also
the specific "calling sequence" the service on the other end expects.  Maybe
with WSDL and UDDI an XML program could figure all this out, but it would be
a SOAP/UDDI/WSDL application, not a generic XML application (at least until
the Semantic Web comes into fruition and all this stuff that now requires
human intelligence can expressed in RDF ... but don't hold your breath!).

SVG is pretty much at the other extreme; without detailed knowledge of the
principles of computer graphics and the semantics of the SVG schema, an XML
expert with all the generic tools in the world wouldn't be able to do much
that is interesting or useful with an SVG document.  

> I was under the impression XML is a syntactic system for creating
> application specific languages 

We can imagine something even more extreme, e.g. a (hypothetical) future
version of the MS Word format that uses XML syntax but obfuscates the tags
so that only those who are privy to the application design can do anything
useful with the data.  I'm not sure why a vendor would bother using XML for
such a format ...(perhaps because they have a hyper-optimized XML parser
that is faster than a proprietary format parser would be?  Because they want
to market it as "XML"?  To make it SOAP-friendly, or XML-database friendly,
or whatever???).  

Anyway, in the "even more extreme" scenario, XML is nothing more than a
syntactic system, and generic XML tools and expertise would be no more
useful than generic BNF tools are today.  But that WOULD be a different
world than we live in now, but Simon and Didier seem to be saying that world
is changing fast.