RE: [xml-dev] My report on experiments with unused namespaces

[Richard Salz <rsalz@us.ibm.com>]

"Costello, Roger L." <costello@mitre.org> wrote on 09/22/2010 04:58:34 AM:
> ... unused namespaces can contain 
> large amounts of information which is unchecked by XML applications 
> ...
> This is a significant concern for me and others.

Really?  A *significant* concern?  I am skeptical.  As others have pointed 
out, there are many ways to have covert channels in an XML document.

> Goal: identify and eliminate unused namespaces.

I don't believe you can do this unless you have a schema (XSD, or maybe 
Relax) that completely describes the content. Without that you have no way 
of knowing where QName's can appear in content.  And even then, you have 
to be able to parse all content -- how many QName's in this element:
    "<analogy>a:b :: c:d</analogy>"

        /r$

--
STSM, WebSphere Appliance Architect
https://www.ibm.com/developerworks/mydeveloperworks/blogs/soma/