OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

 


 

   Re: XML Extensibility / XML Schemas

[ Lists Home | Date Index | Thread Index ]
  • From: AndrewWatt2000@aol.com
  • To: xml-dev@lists.xml.org
  • Date: Thu, 24 Aug 2000 17:10:01 -0400 (EDT)

In a message dated 24/08/00 14:19:57 GMT Daylight Time, sbpxhtml@yahoo.com 
writes:

Hi Sean,

You raise a number of points which interest me.

<< Have you noticed that when you give XML a DTD, it
stops becoming extensible?
 >>

Well, interestingly enough, you could argue that XML is never extensible. If 
it can have an infinite number of different elements then it is not possible 
to extend it at all. It is only possible to constrain it.

<< We seem to forget that XML isn't really a language -
it's a meta-language; a means of setting up (sometimes
proprietary) new languages. >>

I think it is important that we don't forget that XML should really be called 
a meta-language and that XMML is a more accurate (less inaccurate?) acronym 
than XML.

It is important not least because if the fact that XML/XMML is a metalanguage 
is put up front it makes newcomers realise pretty fast that they are dealing 
with something out of the orbit of their likely past experience. That 
awareness would hopefully breed a healthy alertness to some of the subtleties 
of XML.

Very few people have previous familiarity with a "language without words", 
such as XMML really is.

<< 
Which all seems to favour some kind of xml+ MIME type
for XML applications; to show that these new languages
are XML based, but have a DTD.

Of course, the saving grace may be XML Schemas; these
will allow XML to be 'valid' documents, but ones that
are still extensible.

Problem: I wonder if it's possible to transform both
XSLT & XML Schemas with XSLT? >>

Well, I think that in principle both are possible. It isn't so immediately 
obvious what they would be used for - in terms of some "killer app". At least 
it isn't to me.

I raised the idea of using XSLT to e.g. create new XSLT transformation sheets 
in a discussion recently but the feedback I had at that time was that it was 
a gimmick. To be honest, I still don't see a "killer app" arising from that, 
but perhaps in a couple of years we will look at n-tier XSLT transformations 
with as much equanimity as Java server people view such things in J2EE.

By some bizarre coincidence, the potential of using XSLT to manipulate XML 
Schemas was something I was in discussion about earlier today. Until XML 
Schema is stable then it is likely that usage will be limited to 
experimentation but it certainly is an intriguing possibility. I am not sure 
if my brain is elastic enough to accomodate all the implications though. :)


<< - Probably yes because they are valid XML, but can you
imagine the carnage/innovation this could cause...
 >>

Need there be more "carnage" than with n-tier web application servers? Sure, 
there needs to be an adjustment of perspective on what the Extensible Markup 
MetaLanguage is, but isn't that what life-long learning is about? <grin/>

Andrew Watt




 

News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS