RE: ANN: SAX Filters for Namespace Processing

["Simon St.Laurent" <simonstl@simonstl.com>]

On 31 Jul 2001 17:03:18 -0700, Michael Brennan wrote:
> Given this example, I'd say it is terribly dangerous to make the general
> assumption that an unqualified child element is equivalent to one in the
> containing namespace. You have to know the explicit intent of the author.

That's why the namespace filters allow developers to explicitly include
and exclude particular namespaces from such extension to their child
elements. For example, if you want to make sure SOAP envelopes don't
influence their child elements, it's simple:

<namespaces xmlns="http://simonstl.com/ns/namespaces/elements/">
<include nsURI="##any" />
<exclude nsURI="http://www.w3.org/2001/6/soap-envelope" />
</namespaces>

Exclusions always override inclusions.

Honestly, though, I'm finding less and less interest in (or sympathy
for) the intentions of the author.  It seems like these are really
choices for recipients, to be made with consideration for how much they
have to gain or lose by dealing with authorial intention.

Maybe I've just spent too long in publishing, where the reader is the
final arbiter.