Sorry my last message was cut off! It's freezing here today (in mid-summer!) and my numb fingers slipped on the keyboard and somehow sent it off :-(
I agree with Wendell, and I suggest there's something to be said for using both conventions.
What I was saying is that you can use the numbered convention for the convenience of encoders, and still translate it to use the un-numbered div convention for the convenience of later processing (converting to HTML etc). The transformation is trivial.
(Something like this). To convert from numbered to un-numbered:
The other way:
> -----Original Message-----
> From: Wendell Piez [mailto:[log in to unmask]]
> Sent: Saturday, 14 February 2004 9:41 a.m.
> To: [log in to unmask]
> Subject: Re: <div> versus <divN> again
> Hi Mark,
> Could you explain what you mean by "nesting order" in the below?
> Your argument seems to suggest that the encoder has a choice
> of what level
> of numbered div to place inside another, whereas the DTD mandates that
> inside a div1 can only be a div2, inside a div2 can only be a
> div3 etc.
> Given this constraint, an XPath //div1 (in a document where
> there is no
> div0) is nothing more than //div[count(ancestor-or-self::div)=1] in a
> "plain div" model. In other words, given the constraints of
> the DTD, there
> is no semantic content whatsoever in the X in "divX" that
> cannot be equally
> well derived from inspecting the nesting in the instance.
> In other words, it's nothing but syntax sugar. To back this
> claim up, I'd
> be willing to take a shot at transforming any valid TEI numbered-div
> structure you want to throw at me into an otherwise-identical
> TEI with an
> unnumbered-div structure *and back again*, using plain old
> XSLT 1.0 and
> *not* a "level" flag (following David's not unreasonable
> suggestion). You
> only need to tell me whether you want the result of the final
> document to
> start with a div0 or a div1 (another TEI distinction without
> a semantic
> Please explain to me what I'm missing! or show me a
> numbered-div document
> where the numbering in the div names provides a semantic otherwise
> impossible (or even difficult) to derive. I'm willing to be
> proven wrong.
> Heck, I think I'll write those stylesheets right now....
> At 11:20 AM 2/13/2004, you wrote:
> >Numbered divs are an easy way for the *encoder* to provide the
> >desired object nesting for indexing/searching, contextualization,
> >and object display independent of the content of a type attribute.
> >This is particularly useful if you are implementing a collection,
> >or any group of collections, that can be made up of any number of
> >different kinds of divs that require distinct nesting orders.
> >Unnumbered divs leave it to the implementor to either guess, on
> >a case by case basis, the best nesting order, or simply to assume
> >that all divs are equal. IMHO, having the encoder explicitly
> >identify the nesting order that s/he considers most appropriate
> >for a particular document outweighs the implementation problems
> >that numbered divs might cause.
> Wendell Piez mailto:[log in to unmask]
> Mulberry Technologies, Inc. http://www.mulberrytech.com
> 17 West Jefferson Street Direct Phone: 301/315-9635
> Suite 207 Phone: 301/315-9631
> Rockville, MD 20850 Fax: 301/315-8285
> Mulberry Technologies: A Consultancy Specializing in SGML and XML