Dynamic documents revisited

Back in July, when Tim Bray pointed to a best-practices document that the W3C TAG (Technical Architecture Group) gathers periodically in exotic locations to discuss, I was exploring a way of dynamically building views of XML documents in the browser. In a couple of postings (and with some help from Bob Clary) I was able to post a Web page that turned the TAG's Web page into a dynamic outline that can selectively display elements styled as:

In addition, it can selectively display:

Here's a new version of that dynamic viewer for the latest revision of the TAG document, which Tim wrote about today. As I review the principles and practices, which my handy-dandy viewer dynamically assembles into tidy lists for me, I'm still not sure which of them accounts for my ability to do this neat trick. I supppose Good practice: Thoughtful URI creation comes closest, except that in this case, it's really more like Thoughtful content creation. Maybe that's covered by the separate note formattingProperties-19: Reuse existing formatting properties/names, coordinate new ones?

I'm also still not quite sure of XHTML's place in the world. Section 4.6.7., Media Types for XML, advises:

Good practice: In general, server managers SHOULD NOT assign Internet Media Types beginning with text/ to XML representations.
The TAG document has a doctype of XHTML 1.0 Strict, and comes over the wire as text/html, but I'm treating it as XML so I can query it with XPath and transform it. I'm not losing sleep over it, but is this a sneaky exploit or a principled technique?

Former URL: http://weblog.infoworld.com/udell/2003/11/12.html#a845