xml-dev - Re: Feeler for SML (Simple Markup Language)

[David Megginson <david@megginson.com>]

"Michael Champion" <Mike.Champion@softwareag-usa.com> writes:

> Perhaps, but just as XML inself sprang from a concern that the SGML standard
> was far too complex for Web applications, I'm seeing increased concern that
> the XML standards are getting too complex, and the standards processes too
> slow moving,  to really meet the needs of e-business. 

Sure, I'm hearing the same thing, but the complaints aren't about the
XML 1.0 Syntax, they're about the related work in schemas, linking,
etc.  Creating a new, variant XML syntax won't help much with any of
that.

> > Even if the separate dialect is a pure subset, it will still split the
> > XML market for no good reason
> 
> I guess the reason why I'm intrigued by Don Park's proposal is that it seems
> to me (especially having made a career move from a text/publishing XML
> vendor to an enterprise commerce XML vendor)  like there *is* a good reason
> for considering whether the set of XML features needed by e-business
> applications is massively smaller than the set needed by text
> authoring/publishing/browsing applications.

But people writing that kind of application don't much care about the
fact that the XML parser they're using happens to support PIs or
notations -- sure, the parser's maybe 5% larger than it would be
without that stuff, but that's a lot better than software based on
some other specs and standards.

> > That said, it's certainly useful to define APIs that hide some of that
> > stuff -- applications should not have to worry about unparsed
> > entities, notations, etc. unless they want to
> 
> Right.  Maybe the "SML" idea would meet less resistance if it referred to
> XML processing tools and APIs that quietly ignored some well-defined set of
> legal XML constructs (attributes, comments, PIs, notations, entities, or
> whatever) in well-formed XML documents rather than defining a subset of XML
> itself in which these are illegal. I think that's well within the spirit of
> both Don's posting and the "less is more" perspective in the xml.com
> articles mentioned above.

But you can do that already.  With SAX, it's trivial: just don't call
Parser.setDTDHandler and implement
DocumentHandler.processingInstruction as an empty method:

  public void processingInstruction (String target, String data)
  {
    // I don't care.
  }

If you're inheriting from HandlerBase, you don't even have to do
that.

What we need (and perhaps what you and Don are getting at) is for
individual applications to make declarations like

  XHTML user agents will ignore processing instructions.

Sure, why not?  It won't hurt much, though there might be some
complications when we start mixing vocabularies.


All the best,


David

-- 
David Megginson                 david@megginson.com
           http://www.megginson.com/

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To unsubscribe, mailto:majordomo@ic.ac.uk the following message;
unsubscribe xml-dev
To subscribe to the digests, mailto:majordomo@ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@ic.ac.uk)