Re: 3.8.10. The address element

Hi Ben and Smylers,

On Aug 7, 2007, at 10:28 AM, Smylers wrote:

>
> Ben Boyle writes:
>
>> On 8/8/07, Simon Pieters <simonp@opera.com> wrote:
>>
>>> <address> is not a sectioning element.
>>
>> That's good news for HTML; bad news for the spec. The spec reads  
>> as if
>> address elements are sections and definitely requires clarification.
>
> Which bit reads like that to you?  Contrast, for example, <aside>  
> which
> is described as:
>
>   Sectioning block-level element
>
> with <address> as:
>
>   Block-level element


I  had the same problem reading this as Ben, so I definitely see the  
ambiguity. I think part of the problem is that the chapter "Sections"  
includes several elements that are not "sectioning elements": in  
particular ADDRESS and H! – H6. That makes it easy to think all of  
the elements in the chapter are sectioning element. It might be  
better to rename the chapter to something like: "Sections and their  
metadata". Then it has the HEADER element and the FOOTER element that  
are ambiguous because they both signify metadata containers and  
suggest a section (as in a site HEADER and FOOTER). Finally  
BLOCKQUOTE is thrown in here (and then excluded from the outlining  
algorithm) for no particular reason.

>
> That seems unambiguous that the spec doesn't think <address> is
> sectioning, something re-inforced by the introduction to the Sections
> section, where the paragraph immediately after the definition of
> 'sectioning elements' shows that <address> elements are something that
> appear inside section elements:
>
>   Some elements are scoped to their nearest ancestor sectioning  
> element.
>   For example, address elements apply just to their section.
>
> Is there somewhere else I missed which gives a contradictory  
> impression?

Again, it's the exposition of all of this as a whole. There is not  
enough care taken to avoid confusion for readers.

>> "Sectioning elements" should explicitly state the element names.
>> http://dev.w3.org/html5/spec/Overview.html#sectioning
>
> That's a good idea.

This is needed for each element kind. Providing lists of the elements  
take the burden off the prose defining the element kind.

Take care,
Rob