Namespaces and XML validation

[W. E. Perry]

Lisa Rein wrote:

> peter murray rust wrote:
>
> >
> > As I have often
> > (probably boringly) said, I think my community is far more interested in
> > semantic than syntactic validity. (Actually they probably don't care about
> > either much...)
>
> How can you possibly have one without the other?  If your syntax is
> bogus -- you won't get far with semantics.  How can you?
>
> Am I wrong?  Somebody please set me straight.
>
> lisa rein

I think that we have here the fundamental question of XML and, with it, the
likeliest source of schism among us. XML is routinely introduced as
        a) the (infinitely) extensible markup language, based on the mechanical
concept of well-formedness, on five hard-wired entities, and the three reserved
characters ordered 'xml';
    and also as
        b) a formal subset of ISO-standard SGML.
The hope expressed with both of these formulations is that XML will be used to
mark up meaning rather than presentation. Without terribly much extension,
'presentation' quickly comes to include syntactic forms, and there is a
reasonable argument that the minimal definition in (a) is both as far as we
should go in excluding presentation and as far as we can go while still retaining
some substance to call XML.

Charles Frankston's recent comments show us our choice:  we can draw the line in
defending the DTD as defined by XML 1.0 or we can admit that syntax is an aspect
of presentation and that we are as likely to have as many different syntactic
structures as style sheets to apply to the marked semantics of a given document.
As this list, and the use of XML, have expanded to developers from a wider range
of disciplines, new arrivals have brought from their particular experience
different, and, finally, mutually exclusive, opinions of what are normal,
desirable, or even permissible syntactic premises. Sticking only to text as the
stuff of XML documents, we have already seen how differently the database people
and the DOM people conceive the appropriate syntax for equivalent semantics.

This schism was inherent in the seditious nature of XML as originally defined. By
separating validity from WFness, XML implicitly offered the option of ignoring an
author's DTD. It was always possible in XML, unlike SGML, to take a document 'on
its own terms', even where that conflicted with the author's intent delineated in
a DTD.It has from the first been only a matter of time until other schemas and
processing paradigms routinely substituted their syntactic rules in the
consumption of a document for those which might have constrained the author in
its creation.


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)