E-mail List Archives
Thread: Heading levels
Number of posts in this thread: 16 (In chronological order)
From: Al Sparber
Date: Sun, Jun 25 2006 10:50AM
Subject: Heading levels
No previous message | Next message →
Can someone provide a plain language explanation of why non-sequential
headings might be considered inaccessible? For example, the following
scenario fails in an online automated checker.
<h1>Creating Multi-Level CSS Menus</h1>
Intro
<h2>Adding the markup</h2>
Intro
<h3>Adding the Root-Level Markup</h3>
Steps
<h3>Adding the Sub-Level Markup</h3>
Steps
<h2>Creating the CSS</h2>
Intro
<h3>Creating Rules for Root-Level Menu Items</h3>
Steps
<h3>Creating Rules for Sub-Level Menu Items</h3>
Steps
It seems perfectly logical to me, but apparently this is not correct?
--
Al Sparber
PVII
http://www.projectseven.com
"Designing with CSS is sometimes like barreling down a crumbling
mountain road at 90 miles per hour secure in the knowledge that
repairs are scheduled for next Tuesday".
From: Helen A
Date: Sun, Jun 25 2006 11:00AM
Subject: Re: Heading levels
← Previous message | Next message →
At 17:43 25/06/2006, Al Sparber wrote:
>Can someone provide a plain language explanation of why
>non-sequential headings might be considered inaccessible? For
>example, the following scenario fails in an online automated checker.
><h1>Creating Multi-Level CSS Menus</h1>
>Intro
><h2>Adding the markup</h2>
>Intro
><h3>Adding the Root-Level Markup</h3>
>Steps
><h3>Adding the Sub-Level Markup</h3>
>Steps
><h2>Creating the CSS</h2>
>Intro
><h3>Creating Rules for Root-Level Menu Items</h3>
>Steps
><h3>Creating Rules for Sub-Level Menu Items</h3>
>Steps
its because you have gone from <h3> back to <h2>. You can't
back-track up a level, so working down a page you can't have <h3> on
line N then on line Q put <h2>, it would have to be <h4> as 4 follows
3 numerically speaking.
Helen
From: Al Sparber
Date: Sun, Jun 25 2006 11:10AM
Subject: Re: Heading levels
← Previous message | Next message →
From: "Helen A" < = EMAIL ADDRESS REMOVED = >
To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
Sent: Sunday, June 25, 2006 12:55 PM
Subject: Re: [WebAIM] Heading levels
> At 17:43 25/06/2006, Al Sparber wrote:
>>Can someone provide a plain language explanation of why
>>non-sequential headings might be considered inaccessible? For
>>example, the following scenario fails in an online automated
>>checker.
>><h1>Creating Multi-Level CSS Menus</h1>
>>Intro
>><h2>Adding the markup</h2>
>>Intro
>><h3>Adding the Root-Level Markup</h3>
>>Steps
>><h3>Adding the Sub-Level Markup</h3>
>>Steps
>><h2>Creating the CSS</h2>
>>Intro
>><h3>Creating Rules for Root-Level Menu Items</h3>
>>Steps
>><h3>Creating Rules for Sub-Level Menu Items</h3>
>>Steps
>
> its because you have gone from <h3> back to <h2>. You can't
> back-track up a level, so working down a page you can't have <h3> on
> line N then on line Q put <h2>, it would have to be <h4> as 4
> follows 3 numerically speaking.
I understand the rule, my question concerns the logic. It doesn't make
sense. Is this, then, one of those areas where one follows the rule to
pass a test, throwing logic to the wind?
In my scenario, I would most certainly want to go back. Perhaps that's
why a lot of online tutorials are hard for me to follow :-)
--
Al
From: Al Sparber
Date: Sun, Jun 25 2006 11:20AM
Subject: Re: Heading levels
← Previous message | Next message →
> From: "Helen A" < = EMAIL ADDRESS REMOVED = >
>> its because you have gone from <h3> back to <h2>. You can't
>> back-track up a level, so working down a page you can't have <h3>
>> on line N then on line Q put <h2>, it would have to be <h4> as 4
>> follows 3 numerically speaking.
>
> I understand the rule, my question concerns the logic. It doesn't
> make sense. Is this, then, one of those areas where one follows the
> rule to pass a test, throwing logic to the wind?
>
> In my scenario, I would most certainly want to go back. Perhaps
> that's why a lot of online tutorials are hard for me to follow :-)
Further-
Let's go back to my original scenario and use your "explanation":
<h1>Creating Multi-Level CSS Menus</h1>
Intro
<h2>Adding the markup</h2>
Intro
<h3>Adding the Root-Level Markup</h3>
Steps
<h4>Adding the Sub-Level Markup</h3>
Steps
<h4>Creating the CSS</h2>
Intro
<h4>Creating Rules for Root-Level Menu Items</h3>
Steps
<h4>Creating Rules for Sub-Level Menu Items</h3>
Steps
Now that makes absolute no sense to me. Apparently, it makes little
sense to publishing software either, as that type of hierarchy would
not lend itself to any automated TOC generator. So again, what is the
logic behind this rule?
If an assistive reader user sets his program to list headings, then my
original scenario still makes a lot more sense than your answer does.
In fact, it makes perfect sense to a lot of people (which is why I
posed the original question). Please try to explain further, if you
will. I would love to teach this concept to our customers and to
employ it on our own site - if someone can provide the logic and not
just recite the specification :-)
Thanks.
--
Al Sparber
PVII
http://www.projectseven.com
"Designing with CSS is sometimes like barreling down a crumbling
mountain road at 90 miles per hour secure in the knowledge that
repairs are scheduled for next Tuesday".
From: Christian Heilmann
Date: Sun, Jun 25 2006 12:30PM
Subject: Re: Heading levels
← Previous message | Next message →
> Can someone provide a plain language explanation of why non-sequential
> headings might be considered inaccessible? For example, the following
> scenario fails in an online automated checker.
which is a proof that online automated checkers are just not a valid
test for accessibility but can only be considered an aid for human
testing. As a human I consider your heading structure alright, for
example, and I guess a blind user using the heading navigation option
would agree.
No point discussing the value of automated testing as the only means
of testing _again_.
--
Chris Heilmann
Blog: http://www.wait-till-i.com
Writing: http://icant.co.uk/
Binaries: http://www.onlinetools.org/
From: zara
Date: Sun, Jun 25 2006 12:40PM
Subject: RE: Heading levels
← Previous message | Next message →
> its because you have gone from <h3> back to <h2>. You can't
> back-track up a level, so working down a page you can't have <h3> on
> line N then on line Q put <h2>, it would have to be <h4> as 4 follows
> 3 numerically speaking.
>
The techniques document for WCAG 1.0 indicates :
"Users should order heading elements properly. For example, in HTML, H2 elements should follow H1 elements, H3 elements should follow H2 elements, etc."[1]
However, I believe that this information can be taken too literally or was poorly explained (perhaps in attempt to prevent abuse). While it is important to preserve logic in how headings are ordered, it is also reasonable to consider the purpose of the content and use them accordingly (as long as it makes sense structurally).
Also, the W3C "The global structure of an HTML document" indicates :
"Some people consider skipping heading levels to be bad practice. They accept H1 H2 H1 while they do not accept H1 H3 H1 since the heading level H2 is skipped."
So there does not seem to be a clear disapproval in this document of what Mr. Sparber is proposing. The example he has provided seems quite logical in that it is clear that the H3s following each H2 are subsection titles relating to the section.
If this is truly forbidden by WCAG 1.0 or just the result of inadequate explanations, it should be fixed in any case. I have found nothing specific about this in information relating to headings in the current draft of WCAG 2.0 nor the complementary documents so it is hard to tell if WAI's position has evolved on this issue though I may have missed it as I find these documents somewhat difficult to consult.
Catherine
--
[1] http://www.w3.org/TR/WCAG10-HTML-TECHS/#document-headers
[2] http://www.w3.org/TR/REC-html40/struct/global.html#h-7.5.5
--
Catherine Roy, consultante
www.catherine-roy.net
From: Patrick H. Lauke
Date: Sun, Jun 25 2006 12:50PM
Subject: Re: Heading levels
← Previous message | Next message →
Christian Heilmann wrote:
> No point discussing the value of automated testing as the only means
> of testing _again_.
Spoilsport...
P
--
Patrick H. Lauke
___________
re
From: Al Sparber
Date: Sun, Jun 25 2006 1:00PM
Subject: Re: Heading levels
← Previous message | Next message →
From: "zara" < = EMAIL ADDRESS REMOVED = >
>> its because you have gone from <h3> back to <h2>. You can't
>> back-track up a level, so working down a page you can't have <h3>
>> on
>> line N then on line Q put <h2>, it would have to be <h4> as 4
>> follows
>> 3 numerically speaking.
> Also, the W3C "The global structure of an HTML document" indicates :
>
> "Some people consider skipping heading levels to be bad practice.
> They accept H1 H2 H1 while they do not accept H1 H3 H1 since the
> heading level H2 is skipped."
>
> So there does not seem to be a clear disapproval in this document of
> what Mr. Sparber is proposing. The example he has provided seems
> quite logical in that it is clear that the H3s following each H2 are
> subsection titles relating to the section.
Thank you Catherine.
From: Al Sparber
Date: Sun, Jun 25 2006 1:10PM
Subject: Re: Heading levels
← Previous message | Next message →
From: "Patrick H. Lauke" < = EMAIL ADDRESS REMOVED = >
> Christian Heilmann wrote:
>
>> No point discussing the value of automated testing as the only
>> means
>> of testing _again_.
>
> Spoilsport...
Sanity saver...
From: Tim Beadle
Date: Sun, Jun 25 2006 1:50PM
Subject: Re: Heading levels
← Previous message | Next message →
On 25/06/06, Helen A < = EMAIL ADDRESS REMOVED = > wrote:
> its because you have gone from <h3> back to <h2>. You can't
> back-track up a level, so working down a page you can't have <h3> on
> line N then on line Q put <h2>, it would have to be <h4> as 4 follows
> 3 numerically speaking.
If that's the case (which I do not believe for one second that it is)
then creating well-structured, hierarchical document outlines is going
to be *really, really difficult*.
Tim
From: Patrick H. Lauke
Date: Sun, Jun 25 2006 4:40PM
Subject: Re: Heading levels
← Previous message | Next message →
Helen A wrote:
> its because you have gone from <h3> back to <h2>. You can't back-track
> up a level, so working down a page you can't have <h3> on line N then on
> line Q put <h2>, it would have to be <h4> as 4 follows 3 numerically
> speaking.
Just to clarify, that's wrong. The issue here is whether or not it's ok
to skip a level going down (i.e. having an H1 followed by an H3, without
any H2 in between). Going back up one level is perfectly valid and logical.
P
--
Patrick H. Lauke
___________
re
From: Al Sparber
Date: Sun, Jun 25 2006 4:50PM
Subject: Re: Heading levels
← Previous message | Next message →
From: "Patrick H. Lauke" < = EMAIL ADDRESS REMOVED = >
> Helen A wrote:
>
>> its because you have gone from <h3> back to <h2>. You can't
>> back-track up a level, so working down a page you can't have <h3>
>> on line N then on line Q put <h2>, it would have to be <h4> as 4
>> follows 3 numerically speaking.
>
> Just to clarify, that's wrong. The issue here is whether or not it's
> ok to skip a level going down (i.e. having an H1 followed by an H3,
> without any H2 in between). Going back up one level is perfectly
> valid and logical.
Just to clarify my original question, it was that the following
scenario should be acceptable and correct:
<h1>
<h2>
<h3>
<h2>
<h3>
No skipping on my part - just repeating where it makes sense to.
--
Al Sparber
From: Patrick H. Lauke
Date: Sun, Jun 25 2006 5:10PM
Subject: Re: Heading levels
← Previous message | Next message →
Al Sparber wrote:
>> Just to clarify, that's wrong.
...
> Just to clarify my original question, it was that the following scenario
> should be acceptable and correct:
>
> <h1>
> <h2>
> <h3>
> <h2>
> <h3>
>
> No skipping on my part - just repeating where it makes sense to.
Just to clarify: next time I'll read the thread starter :)
In that case, the automated tool is *very* wrong if it flags it as an error.
P
--
Patrick H. Lauke
___________
re
From: Andrew Arch
Date: Sun, Jun 25 2006 10:50PM
Subject: RE: Heading levels
← Previous message | Next message →
Nothing wrong with your semantics - blame the tool :)
Basically, the rule that we apply is don't skip levels on the way down, you
can jump back up to any level (though most would argue that a page should
only have one H2).
Think about it like a text book table of contents - does it make sense then?
Are all the sub-sub-headings subservient to the right sub-headings? Etc. The
AIS Web Accessibility Toolbar [1] has a nice headings call out feature to
give you the relative levels of all headings on a page (in a new window to
separate them from the content for analytical purposes) - see
structure/heading-structure.
Andrew
[1] www.visionaustralia.org.au/ais/toolbar/
From: Andrew Arch
Date: Sun, Jun 25 2006 11:40PM
Subject: RE: Heading levels
← Previous message | Next message →
Oops - typo - I meant to say that many folk would suggest there should only
be one H1. There can be many H2s.
Sorry, Andrew
From: Jon Gunderson
Date: Mon, Jun 26 2006 9:00AM
Subject: RE: Heading levels
← Previous message | No next message
The functional accessibility evaluator [1] allows any heading
level before the last H1 element (Note: H1 are reserved for
titling pages and the last H1 is considered the sub page title
of a page [2]). After the last H1 the heading should use
hierarchical structure. H2 should also be used to label
navigation bars[3], so there maybe h2 before the first H1 element.
Jon
[1] Functional Accessibility Evaluator
http://fae.cita.uiuc.edu
[2] HTML Best practices: Unique titles
http://html.cita.uiuc.edu/nav/title.php
[3] HTML Best practices: Navigation bars and menus
http://html.cita.uiuc.edu/nav/menus.php
---- Original message ----
>Date: Mon, 26 Jun 2006 15:28:10 +1000
>From: "Andrew Arch" < = EMAIL ADDRESS REMOVED = >
>Subject: RE: [WebAIM] Heading levels
>To: "'WebAIM Discussion List'" < = EMAIL ADDRESS REMOVED = >
>
>Oops - typo - I meant to say that many folk would suggest
there should only
>be one H1. There can be many H2s.
>
>Sorry, Andrew
>
>