W3C home > Mailing lists > Public > www-html@w3.org > August 2003

RE: Bottom-up Sections

From: Jewett, Jim J <jim.jewett@eds.com>
Date: Fri, 8 Aug 2003 10:45:10 -0400
Message-ID: <B8CDFB11BB44D411B8E600508BDF076C15C322DB@usahm010.exmi01.exch.eds.com>
To: www-html@w3.org

Mikko Rantalainen

> Stefan Ram

> >   certain "small sections" [such as theorems]
> >   do not fit into the XHTML 2 section scheme. 
> >   [These are often rendered as sidebox; theorems 
> >   are an example]

> Do you really think that those "boxes" aren't 
> *logically* part of containing section.  Traditional 
> rendering might not hint that the "box" is really a 
> sub section, but is that really the intent? Or are those 
> "boxes" logically more analogous to diagrams and 
> pictures or something else which is logically out of 
> flow? 

Yes, they are part of the section.
Yes, they are out of flow, exactly like a diagram or picture.

I believe his concern is that "section" doesn't work because it 
counts nesting from the top.  Using this example:

<h>Chapter 1</h>
	Intro Text.
	<example id="ex1">here is the first sidebox</example>
	Here is more text.

	<h>Section 1.1</h>
		Here we talk about a subcase.
		<example id="ex2">here is the second sidebox</example>

If <example> is replaced by <section>, then ex1 will (wrongly)
appear to be more important than ex2.

I do think it would work properly if <example> were replaced by <div>.

There are still problems with headings though.  Logically, the
<h> elements should be children of the appropriate section 
rather than siblings.  The examples should have headings (or
titles or better yet captions) as well.

Stefan may also be suggesting that this is a special type of
div that should have its own name.  There, I disagree.  I
could support several types of div (diagram, sidenote, 
example, dialogue, definition) or I could support using
div for all of them and looking at the class.  

I don't see extra value in grouping a subset together just
to say that they count levels from the bottom (depth of
descendent tree).   I'm not completely sure that these 
constructs should even nest at all - but I really don't see
why the difference between 




is worth a new element.

Received on Friday, 8 August 2003 10:45:21 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:06:04 UTC