"Namespaces in XML" idea?

[Tyler Baker]

David Megginson wrote:

> Dan Brickley writes:
>
>  > > By the way, I was a little surprised to read that expat does not
>  > > pass xmlns attributes to the application.  I had always assumed
>  > > that attributes were part of the document's data and therefore
>  > > the processor was required to pass them to the application.
>  > > [...]
>  >
>  > A pragmatic reason for passing on the xmlns:xyz information is that some
>  > applications might want to go beyond what 'Namespaces in XML' itself
>  > provides and use similarly abbreviated values _inside_ data attributes
>  > instead of full URIs (eg. for data types), and make use of the XMLNS
>  > info when doing so.
>
> The easiest way to resolve this question would be to decide that
>
> 1. Namespace declarations do not appear as attributes in the
>    Namespaces view; and
>
> 2. some applications may require information from both the XML 1.0 and
>    the Namespaces view concurrently.

This is one of the chief reasons why maybe splitting XML 1.0 and XML 1.0 with "Namespaces in
XML" into two separate beasts may not be such a bad idea.  IMHO XML 1.0 without namespaces and
XML 1.0 with namespaces are two totally different beasts.  How you actually deal with
processing XML 1.0 and XML 1.0 with "Namespaces in XML" will be a lot different at the parser
level and application level (not to mention the parser interfaces you use) than I think a lot
of people would like to believe.  So as not to confuse end-users with having to dynamically
configure the right parser interfaces to handle both plain old XML 1.0 and XML 1.0 with
"Namespaces in XML".  I would like to think of "Namespaces in XML" as an extension to XML 1.0
and not a "change" to XML 1.0.  This way we don't have to pollute the XML 1.0 parser
interfaces and application frameworks with "Namespaces in XML" (something I favor), as well as
the fact that people can use clean parser interfaces and application frameworks to handle the
"Namespaces in XML" flavor of XML.  For the DOM iteself, the current Level 1 Recommendation
would not have to be changed, rather you would just create an NDOM whose processing model is
based completely on "Namespaces in XML" being the current data environment.

Most XML Parser packages provide both a validating parser and a non-validating parser.  What
would be so bad with a non-validating namespaces parser to be added where validation would not
be done by DTD's but a more powerful schema language like DDML.

Tyler


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/ and on CD-ROM/ISBN 981-02-3594-1
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)