RE: Headings test

Hi Loretta,

The worry I have is that people will get confused if a heading isn't a
logical subsection of a previous heading. For example: a page with three
subsections and one list of related articles about the whole topic is marked
up using H1 - h2 - h2 - h4 where the H4 is the heading for the 'related
articles' section. From the headings, it follows that they are related
articles to the last subsection although if we go with your suggestion it
could be either the entire page (if people ignore our advise about
hierarchy) or the last section. 

I think headings are one of the most important elements to define the
structure of the document. If we let go of the hierarchy need, we are
letting a lot of people down. I even think the 1.3.1 SC requires hierarchy
because otherwise there is no structure but they're just a bunch of
headings. For example a sighted user will often spot (from the design) that
the 'related articles' list refers to the whole page and not just the last
subsection, a screenreader user should have the same benefits.

Yvette Hoitink
Heritas, Alphen aan den Rijn, the Netherlands
E-mail: y.p.hoitink@heritas.nl
WWW: http://www.heritas.nl  


> -----Original Message-----
> From: public-wcag-teamb-request@w3.org 
> [mailto:public-wcag-teamb-request@w3.org] On Behalf Of 
> Loretta Guarino Reid
> Sent: vrijdag 10 februari 2006 2:52
> To: Yvette Hoitink; public-wcag-teamb@w3.org
> Subject: RE: Headings test
> 
> 
> If I understand you correctly, you are defining a hierarchy, 
> just one in which we are looser with our definition of "next 
> level". I actually think this will be more confusing for 
> people than just requiring
> hierarchy.       
> 
> The sense that I got from the working group today is that 
> they don't want to introduce hierarchical requirements for 
> HTML under GL 1.3. We might be able to bring them in under 
> 2.4.6, although I think even there we were getting some pushback. 
> 
> Loretta Guarino Reid
> lguarino@adobe.com
> Adobe Systems, Acrobat Engineering 
> 
> > -----Original Message-----
> > From: public-wcag-teamb-request@w3.org [mailto:public- 
> > wcag-teamb-request@w3.org] On Behalf Of Yvette Hoitink
> > Sent: Thursday, February 09, 2006 1:31 PM
> > To: public-wcag-teamb@w3.org
> > Subject: Headings test
> > 
> > 
> > Hi group,
> > 
> > I will not try to write this in an understandable manner 
> yet but focus 
> > on mathimatical exactness (I've defined the heading level of
> > H1 as lower than
> > the heading level of H2):
> > 
> > For each header H, determine the heading number (X). For 
> each heading 
> > level lower than X, determine the last occurrence (L) of 
> that heading 
> > that preceedes X. H should be a logical child of L.
> > 
> > This algorithm should allow the following case:
> > H1 - H2 - H4 - H2 as long as both the H2s are childs of the 
> H1 and the 
> > H4 is a child of H2 and H1.
> > 
> > It does not allow H1 - H2 - H2 - H4 where H4 is the heading 
> for a set 
> > of related links that are about the whole page (h1) and not 
> the second 
> > subsection (h2).
> > 
> > Yvette Hoitink
> > Heritas, Alphen aan den Rijn, the Netherlands
> > E-mail: y.p.hoitink@heritas.nl
> > WWW: http://www.heritas.nl
> > 
> 
>

Received on Friday, 10 February 2006 08:53:48 UTC