Re: nesting header elements (3.5.1)

Also, from a practical viewpoint, it will depend on how existing browsers 
handle headers.  By "browser" I mean a standalone speech enabled browser or 
a combination of standard browser and screenreader.

  For example, if a browser has a function  "Jump to the next highest level 
heading" which will jump through the H3's if those are the highest, then 
having top level H3's is OK.

However, if the browser only has a function "Jump to the next H1" or "Jump 
to the next H2"  Jump to the next H3, etc. then the user has to figure out 
what the top level heading is, say H3, and use that particular command.  In 
that case, having top level headings of H3 is not good.

Does anyone know offhand what the navigation situation is?

Len


At 03:41 PM 3/9/00 -0500, Charles McCathieNevile wrote:
>This doesn't work for HTML. H1 etc are not container elements, so a document
>will not end with the same level header as it started with (except in
>exceptional circumstances)
>
>I agree with the general idea I think - that a page which has
>
>H1  H2  H3  H3  H2  H3  H3
>
>would be ok, as would a page that had
>
>H3  H4  H5  H6  H5  H5  H6  H4  H5  H4  H4
>
>but not a page that went
>
>H2 H1 H3 H4 H3 H4
>
>(unless specified by the author. There are legitimate use cases for having H2
>H1 H2 I suspect, although they may be presentational.)
>
>Charles McCN
>
>On Thu, 9 Mar 2000, Harvey Bingham wrote:
>
>   At 2000-02-18 17:06-0500, Wendy wrote:
>   >Hello,
>   >
>   >Len Kasday raised the following issue regarding technique 3.5.1 in the 21
>   >December draft [http://www.w3.org/WAI/ER/IG/ert/ert-19991221.html
>   >]:
>   ><blockquote>
>   >| 1. The first header element in the document must be H1
>   >| 2. There must be only one H1 element in the document
>
>   Since a document may be part of a higher-level document, the requirement
>   to have any, or even only one H1 element in a document, is too 
> constraining.
>   My preference is that a document should complete a level, in other words,
>   if it starts with <Hn>, it should end with </Hn> and not contain any
>   <Hm>...</Hm> where m, n in 1, 2, ... 6; and m < n. That would allow
>   a reasonable way to make the set of pages with textual content nest
>   consistently. In the set, the "top-level" document could have just one
>   <H1>...</H1>.
>
>   The concept of a document manifest, containing a correctly nested set of
>   page references, is part of the open e-book work. In it, presumable
>   the document's pages are ordered, and their contents may have nested
>   levels of detail, referenced at appropriate places, either in the manifest
>   or in pages.
>
>   In most pages today, many references are to pages that are not strictly
>   part of the logic of the current document. For these, the expectation
>   that they have <H1>...</H1> is certainly acceptable.
>   ...
>
>   >Technique 3.5.1 [priority 2] Check document for header nesting
>   >Discussion Status:
>   >awaiting discussion
>   >Evaluation:
>   >Header elements (H1-H6) should be checked to ensure they are nested
>   >according to the following rules
>   >Header levels must not increase by more than 1 level. Example: H2
>   >following H1 is good. H3 following H1 is bad.
>
>   Would have to check any internal references to other docs to see if they
>   contribute correctly nested material. Many referenced pages are "out of the
>   hierarchy", so somehow they would need to be exempted.
>
>   Regards/Harvey Bingham
>
>
>--
>Charles McCathieNevile    mailto:charles@w3.org    phone: +61 (0) 409 134 136
>W3C Web Accessibility Initiative                      http://www.w3.org/WAI
>Location: I-cubed, 110 Victoria Street, Carlton VIC 3053
>Postal: GPO Box 2476V, Melbourne 3001,  Australia

-------
Leonard R. Kasday, Ph.D.
Institute on Disabilities/UAP, and
Department of Electrical Engineering
Temple University
423 Ritter Annex, Philadelphia, PA 19122

kasday@acm.org
http://astro.temple.edu/~kasday

(215) 204-2247 (voice)
(800) 750-7428 (TTY)

Received on Thursday, 9 March 2000 16:08:46 UTC