Thread Subject: Bypassing content.

Return to this mailing list's archives

From: Sean Hayes
Date: Fri, Jun 08 2007 9:55 AM
Subject: Bypassing content.

I have a slight issue on the testability of : "A mechanism shall be available to bypass blocks of content that are repeated on multiple Web pages."

What exactly constitutes a 'block of content'?

Also if the repeated content is the last thing on the page, is a bypass still necessary? If so, where should it bypass too?

Sean Hayes
Standards and Policy Team
Corporate Accessibility Group
Microsoft
Phone:
mob +44 7977 455002
office +44 117 9719730

From: Hoffman, Allen
Date: Fri, Jun 08 2007 10:50 AM
Subject: Re: Bypassing content.

This one is intended to allow people to basically use things like
headers to jump from each block to the next quickly. I'm not convinced
this language is exactly right, or better than the original
CFR1194.22(o) either. it is testable, but certainly is not easy to test
in an automated fashion.

I have been trying to work out testing algorithm to count the number of
links from start of page to each link on a page, take the average
link-count, and then by re-analyzing the links in variations of
sequence, determine if shorter paths indeed exist, and where to set
upper and lower boundaries on the tests to get the answer, is there a
way to shorten the link path. if anyone can put that into a provision
we would have a testable check on lowering "link" or "tab" path. The
reason "block of content" is good is that it doesn't really matter what
stuff one has to skip over to get past with the voice, the important
factors is to be able to skip it quickly when needed.






Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sean
Hayes
Sent: Friday, June 08, 2007 11:49 AM
To: TEITAC Web/Software Subcommittee
Subject: [teitac-websoftware] Bypassing content.

I have a slight issue on the testability of : "A mechanism shall be
available to bypass blocks of content that are repeated on multiple Web
pages."

What exactly constitutes a 'block of content'?

Also if the repeated content is the last thing on the page, is a bypass
still necessary? If so, where should it bypass too?

Sean Hayes
Standards and Policy Team
Corporate Accessibility Group
Microsoft
Phone:
mob +44 7977 455002
office +44 117 9719730

From: Peter Wallack
Date: Fri, Jun 08 2007 11:15 AM
Subject: Re: Bypassing content.

Allen --
I like your explanation of the intent of this standard far better than
the current wording. I'm not at all keen on the phrase 'repeated on
multiple web pages' as that seems more directed toward a situation where
the user is taking a linear path through a site, which is often not the
case. How about simply:

"A mechanism shall be available to bypass blocks of content"

leaving us only to haggle over the best wording for 'block of content',
which is a concept that I think we all intrinsically understand but we'd
like something a bit more scientific.

As for testability, this probably is no more testable than having a tab
sequence that is 'logical'. This just takes it to one level higher:
having a logical division of blocks of content that you can sequence
through.

Peter Wallack
Accessibility Program Director
Oracle Corporation



Hoffman, Allen wrote:
> This one is intended to allow people to basically use things like
> headers to jump from each block to the next quickly. I'm not convinced
> this language is exactly right, or better than the original
> CFR1194.22(o) either. it is testable, but certainly is not easy to test
> in an automated fashion.
>
> I have been trying to work out testing algorithm to count the number of
> links from start of page to each link on a page, take the average
> link-count, and then by re-analyzing the links in variations of
> sequence, determine if shorter paths indeed exist, and where to set
> upper and lower boundaries on the tests to get the answer, is there a
> way to shorten the link path. if anyone can put that into a provision
> we would have a testable check on lowering "link" or "tab" path. The
> reason "block of content" is good is that it doesn't really matter what
> stuff one has to skip over to get past with the voice, the important
> factors is to be able to skip it quickly when needed.
>
>
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sean
> Hayes
> Sent: Friday, June 08, 2007 11:49 AM
> To: TEITAC Web/Software Subcommittee
> Subject: [teitac-websoftware] Bypassing content.
>
> I have a slight issue on the testability of : "A mechanism shall be
> available to bypass blocks of content that are repeated on multiple Web
> pages."
>
> What exactly constitutes a 'block of content'?
>
> Also if the repeated content is the last thing on the page, is a bypass
> still necessary? If so, where should it bypass too?
>
> Sean Hayes
> Standards and Policy Team
> Corporate Accessibility Group
> Microsoft
> Phone:
> mob +44 7977 455002
> office +44 117 9719730
>

From: Peter Wallack
Date: Fri, Jun 08 2007 11:20 AM
Subject: Re: Bypassing content.

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Some proposed alternatives to 'block of content', stealing terms used
by our User Experience group:<br>
<ul>
<li>functional group (my preference)</li>
<li>grouped elements</li>
<li>grouped controls</li>
</ul>
<pre class="moz-signature" cols="72">Peter Wallack
Accessibility Program Director
Oracle Corporation</pre>
<br>
<br>
Hoffman, Allen wrote:
<blockquote
cite="mid: = EMAIL ADDRESS REMOVED = "
type="cite">
<pre wrap="">This one is intended to allow people to basically use things like
headers to jump from each block to the next quickly. I'm not convinced
this language is exactly right, or better than the original
CFR1194.22(o) either. it is testable, but certainly is not easy to test
in an automated fashion.

I have been trying to work out testing algorithm to count the number of
links from start of page to each link on a page, take the average
link-count, and then by re-analyzing the links in variations of
sequence, determine if shorter paths indeed exist, and where to set
upper and lower boundaries on the tests to get the answer, is there a
way to shorten the link path. if anyone can put that into a provision
we would have a testable check on lowering "link" or "tab" path. The
reason "block of content" is good is that it doesn't really matter what
stuff one has to skip over to get past with the voice, the important
factors is to be able to skip it quickly when needed.






Allen Hoffman -- <a class="moz-txt-link-abbreviated" href="mailto: = EMAIL ADDRESS REMOVED = "> = EMAIL ADDRESS REMOVED = </a>; v: 202-447-0303

-----Original Message-----
From: <a class="moz-txt-link-abbreviated" href="mailto: = EMAIL ADDRESS REMOVED = "> = EMAIL ADDRESS REMOVED = </a>
[<a class="moz-txt-link-freetext" href="mailto: = EMAIL ADDRESS REMOVED = ">mailto: = EMAIL ADDRESS REMOVED = </a>] On Behalf Of Sean
Hayes
Sent: Friday, June 08, 2007 11:49 AM
To: TEITAC Web/Software Subcommittee
Subject: [teitac-websoftware] Bypassing content.

I have a slight issue on the testability of : "A mechanism shall be
available to bypass blocks of content that are repeated on multiple Web
pages."

What exactly constitutes a 'block of content'?

Also if the repeated content is the last thing on the page, is a bypass
still necessary? If so, where should it bypass too?

Sean Hayes
Standards and Policy Team
Corporate Accessibility Group
Microsoft
Phone:
mob +44 7977 455002
office +44 117 9719730

From: Jared Smith
Date: Fri, Jun 08 2007 11:35 AM
Subject: Re: Bypassing content.

On 6/8/07, Sean Hayes < = EMAIL ADDRESS REMOVED = > wrote:
> I have a slight issue on the testability of : "A mechanism shall be available to bypass blocks of content that are repeated on multiple Web pages."
>
> What exactly constitutes a 'block of content'?

I interpret it to be any section of content that is the same across
two or more Web pages. WCAG 2.0 explanation at
http://www.w3.org/TR/UNDERSTANDING-WCAG20/Overview.html#navigation-mechanisms-skip
and techniques at
http://www.w3.org/WAI/WCAG20/quickref/20070517/Overview.php#navigation-mechanisms-skip

A 'mechanism' is intended to be headings or even proper structural
markup. This allows AT to navigate by headings, lists, paragraphs,
etc. This probably needs to be clarified in the wording. The last
thing we want is 'skip' links ("skip the logo", "skip this ad", "skip
the footer", etc.) scattered in multiple places throughout a page -
that's almost certain to make a page *less* accessible.

I'm not sure how best to reflect this in the wording. Also of note,
should "Web pages" be replaced with something less technology
specific?

Maybe something like, "Blocks of content that are repeated on multiple
pages or displays shall be marked up as distinct structural elements
or a link shall be provided to skip over them"?

Jared Smith

From: Hoffman, Allen
Date: Fri, Jun 08 2007 11:50 AM
Subject: Re: Bypassing content.

I'm trying to complete a large post to highlight several ways we can
complete the inclusion of non-HTML content requirements into our work
now.

jared wrote:
I'm not sure how best to reflect this in the wording. Also of note,
should "Web pages" be replaced with something less technology specific?

yes we should try less specific, like not use "mark up".

The drawback to removing any specific web-based requirements is that we
may loose the harmonization with W3C, and we may weaken them in that
context. The web provisions have been arguably the most successfully
implemented provisions of them all, so broadening them to make them
cover more variations may not be the most effective strategy.

I hope to post my set of three ways to get this accomplished by later
tomorrow as it is taking me quite some time to get done well enough to
share.



Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Jared
Smith
Sent: Friday, June 08, 2007 1:16 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

On 6/8/07, Sean Hayes < = EMAIL ADDRESS REMOVED = > wrote:
> I have a slight issue on the testability of : "A mechanism shall be
available to bypass blocks of content that are repeated on multiple Web
pages."
>
> What exactly constitutes a 'block of content'?

I interpret it to be any section of content that is the same across two
or more Web pages. WCAG 2.0 explanation at
http://www.w3.org/TR/UNDERSTANDING-WCAG20/Overview.html#navigation-mecha
nisms-skip
and techniques at
http://www.w3.org/WAI/WCAG20/quickref/20070517/Overview.php#navigation-m
echanisms-skip

A 'mechanism' is intended to be headings or even proper structural
markup. This allows AT to navigate by headings, lists, paragraphs, etc.
This probably needs to be clarified in the wording. The last thing we
want is 'skip' links ("skip the logo", "skip this ad", "skip the
footer", etc.) scattered in multiple places throughout a page - that's
almost certain to make a page *less* accessible.

I'm not sure how best to reflect this in the wording. Also of note,
should "Web pages" be replaced with something less technology specific?

Maybe something like, "Blocks of content that are repeated on multiple
pages or displays shall be marked up as distinct structural elements or
a link shall be provided to skip over them"?

Jared Smith

From: Sean Hayes
Date: Fri, Jun 08 2007 12:05 PM
Subject: Re: Bypassing content.

Yes, I understand what it's supposed to do, I question whether it actually does it.

As it stands, with a literal interpretation one would have to be able to skip every occurrence of the word "the", so clearly a word is not intended to be a 'block of content', is a <h2>Contents</h2> title a block? How big does a block have to get? Does it have to have a link in it?

As for your algorithm, I typically put the navigation menu at the end of the content, and then position it with CSS; obviating any need for a skip link. Would my content fail?

Rather than require going round the bits you don't want, perhaps a better approach would be to require discoverable entry points to the bits you do want.

Something along the lines of:

"The location of the start of each major block of a web page shall be programmatically determinable."

This still leaves the definition of 'major block' somewhat subjective, but I think it's a lot easier to state something in the positive than in the negative.

Sean Hayes
Standards and Policy Team
Corporate Accessibility Group
Microsoft
Phone:
mob +44 7977 455002
office +44 117 9719730

From: Hoffman, Allen
Date: Fri, Jun 08 2007 12:10 PM
Subject: Re: Bypassing content.

Hmmm, thinking cap on.

I am not convinced "logical tab order" is so difficult to convert to
specification/standard format, and am not convinced that this one isn't
either, yet.

Web pages that are coded to:
allow correct keyboard operation;
associate header/cell information;
expose control element information;
generally can meet this repetitive navigation requirement, but
could still have block of content to navigate through.

Screen readers now generally do this by either using algorithms to
evaluate pages for repetitive int3erface elements, or by allowing the
user to hit t to jump to next table, h for next header, or other letters
for other element types to jump through. To allow this functionality
what is important is to use consistent tagging of blocks so that, for
example, when you hit <next whatever> it doesn't actually skip extra
content, or skip in to the middle of the current block.

Imagine:

header: text.
header: text.
table: information.
header: text.
unordered list: information
header: text
table: links.

What command would one use to jump from "block" to block without a
specific consistent identifier being used for the screen reader to grab
on to and pick up the next part of the page to read?


I'm just writing this to give us all food for thought as we try and
finalize this requirement.
thanks.




Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Peter
Wallack
Sent: Friday, June 08, 2007 1:13 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

Allen --
I like your explanation of the intent of this standard far better than
the current wording. I'm not at all keen on the phrase 'repeated on
multiple web pages' as that seems more directed toward a situation where
the user is taking a linear path through a site, which is often not the
case. How about simply:

"A mechanism shall be available to bypass blocks of content"

leaving us only to haggle over the best wording for 'block of content',
which is a concept that I think we all intrinsically understand but we'd
like something a bit more scientific.

As for testability, this probably is no more testable than having a tab
sequence that is 'logical'. This just takes it to one level higher:
having a logical division of blocks of content that you can sequence
through.

Peter Wallack
Accessibility Program Director
Oracle Corporation



Hoffman, Allen wrote:
> This one is intended to allow people to basically use things like
> headers to jump from each block to the next quickly. I'm not
> convinced this language is exactly right, or better than the original
> CFR1194.22(o) either. it is testable, but certainly is not easy to
> test in an automated fashion.
>
> I have been trying to work out testing algorithm to count the number
> of links from start of page to each link on a page, take the average
> link-count, and then by re-analyzing the links in variations of
> sequence, determine if shorter paths indeed exist, and where to set
> upper and lower boundaries on the tests to get the answer, is there a
> way to shorten the link path. if anyone can put that into a provision

> we would have a testable check on lowering "link" or "tab" path. The
> reason "block of content" is good is that it doesn't really matter
> what stuff one has to skip over to get past with the voice, the
> important factors is to be able to skip it quickly when needed.
>
>
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sean
> Hayes
> Sent: Friday, June 08, 2007 11:49 AM
> To: TEITAC Web/Software Subcommittee
> Subject: [teitac-websoftware] Bypassing content.
>
> I have a slight issue on the testability of : "A mechanism shall be
> available to bypass blocks of content that are repeated on multiple
> Web pages."
>
> What exactly constitutes a 'block of content'?
>
> Also if the repeated content is the last thing on the page, is a
> bypass still necessary? If so, where should it bypass too?
>
> Sean Hayes
> Standards and Policy Team
> Corporate Accessibility Group
> Microsoft
> Phone:
> mob +44 7977 455002
> office +44 117 9719730
>

From: Hoffman, Allen
Date: Fri, Jun 08 2007 12:15 PM
Subject: Re: Bypassing content.

Good thinking Sean.



Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sean
Hayes
Sent: Friday, June 08, 2007 1:24 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

Yes, I understand what it's supposed to do, I question whether it
actually does it.

As it stands, with a literal interpretation one would have to be able to
skip every occurrence of the word "the", so clearly a word is not
intended to be a 'block of content', is a <h2>Contents</h2> title a
block? How big does a block have to get? Does it have to have a link in
it?

As for your algorithm, I typically put the navigation menu at the end of
the content, and then position it with CSS; obviating any need for a
skip link. Would my content fail?

Rather than require going round the bits you don't want, perhaps a
better approach would be to require discoverable entry points to the
bits you do want.

Something along the lines of:

"The location of the start of each major block of a web page shall be
programmatically determinable."

This still leaves the definition of 'major block' somewhat subjective,
but I think it's a lot easier to state something in the positive than in
the negative.

Sean Hayes
Standards and Policy Team
Corporate Accessibility Group
Microsoft
Phone:
mob +44 7977 455002
office +44 117 9719730

From: Walser, Kate
Date: Fri, Jun 08 2007 12:35 PM
Subject: Re: Bypassing content.

Sean, your suggestion to phrase things positively such as "The location
of the start of each major block of a web page shall be programmatically
determinable" sounds good and headed more towards a testable solution.

With respect to the "logical tab order" that's been suggested in a few
folks' e-mails - "logical" becomes tricky. "Logical" is clear to me as a
usability and accessibility-minded person, but to some well-intentioned
but new to accessibility person, it can be dangerous. Example: I saw an
application an agency had had developed for them in which the developers
set the recurring navigation menu on the page to be last in tab order
via tabindex. Unfortunately, this didn't mirror the visual layout of the
page or the order without CSS. The page had literally hundreds of links
on it as it was a more complex application. As a result, users with
screen readers heard the menu first, then the rest of the page contents,
and couldn't seem to get to that menu via tabbing as it was link #156 or
so down the page. Extreme example I know, but as we've ditched the
"readable without style sheet" language and in light of a judgment call
of "logical", becomes tricky. To these developers, the tab order they'd
set was completely logical and satisfied the standard about skipping
past repetitive links.

Cheers,
Kate

Kate Walser
Director, Usability Center of Excellence
SRA International, Inc.
4300 Fair Lakes Court
Fairfax, VA 22033
(703) 502-1170

From: Barrett, Don
Date: Fri, Jun 08 2007 12:45 PM
Subject: Re: Bypassing content.

I agree about the issue of the content being a footer. We might just
say something like:

"A mechanism shall be available to bypass blocks of content that are
repeated on multiple Web pages, except where such content constitutes
the bottom end of the web page."


Don Barrett
Section 508 Coordinator
U.S. Department of Education
(202)-205-8245
= EMAIL ADDRESS REMOVED =

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sean
Hayes
Sent: Friday, June 08, 2007 11:49 AM
To: TEITAC Web/Software Subcommittee
Subject: [teitac-websoftware] Bypassing content.

I have a slight issue on the testability of : "A mechanism shall be
available to bypass blocks of content that are repeated on multiple Web
pages."

What exactly constitutes a 'block of content'?

Also if the repeated content is the last thing on the page, is a bypass
still necessary? If so, where should it bypass too?

Sean Hayes
Standards and Policy Team
Corporate Accessibility Group
Microsoft
Phone:
mob +44 7977 455002
office +44 117 9719730

From: Barrett, Don
Date: Fri, Jun 08 2007 12:50 PM
Subject: Re: Bypassing content.

The original "skip repetitive navigation links" standard was a very
useful and easily implemented standard, as people could tell what was
meant immediately by reading the standard.
Just making major blocks of content programmatically determinable
doesn't necessarily mean they will be easily skippable, at least, as I
understand it.


Don Barrett
Section 508 Coordinator
U.S. Department of Education
(202)-205-8245
= EMAIL ADDRESS REMOVED =

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Walser,
Kate
Sent: Friday, June 08, 2007 2:30 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

Sean, your suggestion to phrase things positively such as "The location
of the start of each major block of a web page shall be programmatically
determinable" sounds good and headed more towards a testable solution.

With respect to the "logical tab order" that's been suggested in a few
folks' e-mails - "logical" becomes tricky. "Logical" is clear to me as a
usability and accessibility-minded person, but to some well-intentioned
but new to accessibility person, it can be dangerous. Example: I saw an
application an agency had had developed for them in which the developers
set the recurring navigation menu on the page to be last in tab order
via tabindex. Unfortunately, this didn't mirror the visual layout of the
page or the order without CSS. The page had literally hundreds of links
on it as it was a more complex application. As a result, users with
screen readers heard the menu first, then the rest of the page contents,
and couldn't seem to get to that menu via tabbing as it was link #156 or
so down the page. Extreme example I know, but as we've ditched the
"readable without style sheet" language and in light of a judgment call
of "logical", becomes tricky. To these developers, the tab order they'd
set was completely logical and satisfied the standard about skipping
past repetitive links.

Cheers,
Kate

Kate Walser
Director, Usability Center of Excellence SRA International, Inc.
4300 Fair Lakes Court
Fairfax, VA 22033
(703) 502-1170

From: David Poehlman
Date: Fri, Jun 08 2007 12:55 PM
Subject: Re: Bypassing content.

since there are other methods tand of course we have the others which can be
named as well. I hope this makes sense and I hope that the revised
standards reflect a more natural and positive progression. You will not
find "skip nav" in a daisy book.
There does need to be a way to move through a page with visibal navigations
aids through tab, but skipping repetative links problematic and instead, we
need targets for expressions rather than actions. For instance, one link
could be labeled navigation, another could be labeled main content, footer,
Programatically determinable would habe been as good then as it is now and
should have been employeed in the standards. han tabbing to navigate web
pages, this is not a major issue as such. Logical order is a well
established norm among developpers and it is my guess that in this case, it
was not actually employed. The danger was in the standard "skip past
repetative links" in the first place.
----- Original Message -----
From: "Walser, Kate" < = EMAIL ADDRESS REMOVED = >
To: "TEITAC Web/Software Subcommittee" < = EMAIL ADDRESS REMOVED = >
Sent: Friday, June 08, 2007 2:30 PM
Subject: Re: [teitac-websoftware] Bypassing content.


Sean, your suggestion to phrase things positively such as "The location
of the start of each major block of a web page shall be programmatically
determinable" sounds good and headed more towards a testable solution.

With respect to the "logical tab order" that's been suggested in a few
folks' e-mails - "logical" becomes tricky. "Logical" is clear to me as a
usability and accessibility-minded person, but to some well-intentioned
but new to accessibility person, it can be dangerous. Example: I saw an
application an agency had had developed for them in which the developers
set the recurring navigation menu on the page to be last in tab order
via tabindex. Unfortunately, this didn't mirror the visual layout of the
page or the order without CSS. The page had literally hundreds of links
on it as it was a more complex application. As a result, users with
screen readers heard the menu first, then the rest of the page contents,
and couldn't seem to get to that menu via tabbing as it was link #156 or
so down the page. Extreme example I know, but as we've ditched the
"readable without style sheet" language and in light of a judgment call
of "logical", becomes tricky. To these developers, the tab order they'd
set was completely logical and satisfied the standard about skipping
past repetitive links.

Cheers,
Kate

Kate Walser
Director, Usability Center of Excellence
SRA International, Inc.
4300 Fair Lakes Court
Fairfax, VA 22033
(703) 502-1170

From: Sailesh Panchang
Date: Fri, Jun 08 2007 1:00 PM
Subject: Re: Bypassing content.

This requirement is primarily designed to help users of screen reading
software.
User agents including assistive technologies provide ability to jump to
different elements (eg. lists, tables, frames,) on a page if structural
markup is used.
Structural markup for tables, headings lists, etc. is required by Para (g)
under 6.1 that reads "Information and relationships conveyed through
presentation shall be programmatically determinable, and notification of
changes to these is available to...". So using structural markup facilitates
AT to skip over blocks / skip to desired block.
Therefore I believe the skip block requirement being debated over
predominantly applies to navigational links. This was the intent of the 2001
S508 standards too.
Note that when groups of links are placed in a list or frame, etc. a
mechanism to skip over them is available to screen reader users- an explicit
skip nav link is not needed.

Conclusion: So I think the focus should just be "group" of navigational
links for the skipping over requirement. Consider requirement of the word
"repetitive" too.

Sailesh Panchang
Senior Accessibility Engineer
Deque Systems Inc. (www.deque.com)
11130 Sunrise Valley Drive, Suite #140,
Reston VA 20191
Phone: 703-225-0380 (ext 105)
E-mail: = EMAIL ADDRESS REMOVED =

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Hoffman,
Allen
Sent: Friday, June 08, 2007 1:28 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

Hmmm, thinking cap on.

I am not convinced "logical tab order" is so difficult to convert to
specification/standard format, and am not convinced that this one isn't
either, yet.

Web pages that are coded to:
allow correct keyboard operation;
associate header/cell information;
expose control element information;
generally can meet this repetitive navigation requirement, but
could still have block of content to navigate through.

Screen readers now generally do this by either using algorithms to
evaluate pages for repetitive int3erface elements, or by allowing the
user to hit t to jump to next table, h for next header, or other letters
for other element types to jump through. To allow this functionality
what is important is to use consistent tagging of blocks so that, for
example, when you hit <next whatever> it doesn't actually skip extra
content, or skip in to the middle of the current block.

Imagine:

header: text.
header: text.
table: information.
header: text.
unordered list: information
header: text
table: links.

What command would one use to jump from "block" to block without a
specific consistent identifier being used for the screen reader to grab
on to and pick up the next part of the page to read?


I'm just writing this to give us all food for thought as we try and
finalize this requirement.
thanks.




Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Peter
Wallack
Sent: Friday, June 08, 2007 1:13 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

Allen --
I like your explanation of the intent of this standard far better than
the current wording. I'm not at all keen on the phrase 'repeated on
multiple web pages' as that seems more directed toward a situation where
the user is taking a linear path through a site, which is often not the
case. How about simply:

"A mechanism shall be available to bypass blocks of content"

leaving us only to haggle over the best wording for 'block of content',
which is a concept that I think we all intrinsically understand but we'd
like something a bit more scientific.

As for testability, this probably is no more testable than having a tab
sequence that is 'logical'. This just takes it to one level higher:
having a logical division of blocks of content that you can sequence
through.

Peter Wallack
Accessibility Program Director
Oracle Corporation



Hoffman, Allen wrote:
> This one is intended to allow people to basically use things like
> headers to jump from each block to the next quickly. I'm not
> convinced this language is exactly right, or better than the original
> CFR1194.22(o) either. it is testable, but certainly is not easy to
> test in an automated fashion.
>
> I have been trying to work out testing algorithm to count the number
> of links from start of page to each link on a page, take the average
> link-count, and then by re-analyzing the links in variations of
> sequence, determine if shorter paths indeed exist, and where to set
> upper and lower boundaries on the tests to get the answer, is there a
> way to shorten the link path. if anyone can put that into a provision

> we would have a testable check on lowering "link" or "tab" path. The
> reason "block of content" is good is that it doesn't really matter
> what stuff one has to skip over to get past with the voice, the
> important factors is to be able to skip it quickly when needed.
>
>
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sean
> Hayes
> Sent: Friday, June 08, 2007 11:49 AM
> To: TEITAC Web/Software Subcommittee
> Subject: [teitac-websoftware] Bypassing content.
>
> I have a slight issue on the testability of : "A mechanism shall be
> available to bypass blocks of content that are repeated on multiple
> Web pages."
>
> What exactly constitutes a 'block of content'?
>
> Also if the repeated content is the last thing on the page, is a
> bypass still necessary? If so, where should it bypass too?
>
> Sean Hayes
> Standards and Policy Team
> Corporate Accessibility Group
> Microsoft
> Phone:
> mob +44 7977 455002
> office +44 117 9719730
>

From: Barrett, Don
Date: Fri, Jun 08 2007 1:10 PM
Subject: Re: Bypassing content.

The concept of being repetitive is critical; if a group of links or
seemingly erroneous content appears only once, the option to skip it
would negatively impact accessibility or usability. The repetitiveness
of a block of links or content means that it falls into the "seen once,
skip many" category, and that increases efficiency.


Don Barrett
Section 508 Coordinator
U.S. Department of Education
(202)-205-8245
= EMAIL ADDRESS REMOVED =

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sailesh
Panchang
Sent: Friday, June 08, 2007 3:02 PM
To: 'TEITAC Web/Software Subcommittee'
Subject: Re: [teitac-websoftware] Bypassing content.

This requirement is primarily designed to help users of screen reading
software.
User agents including assistive technologies provide ability to jump to
different elements (eg. lists, tables, frames,) on a page if structural
markup is used.
Structural markup for tables, headings lists, etc. is required by Para
(g) under 6.1 that reads "Information and relationships conveyed through
presentation shall be programmatically determinable, and notification of
changes to these is available to...". So using structural markup
facilitates AT to skip over blocks / skip to desired block.
Therefore I believe the skip block requirement being debated over
predominantly applies to navigational links. This was the intent of the
2001
S508 standards too.
Note that when groups of links are placed in a list or frame, etc. a
mechanism to skip over them is available to screen reader users- an
explicit skip nav link is not needed.

Conclusion: So I think the focus should just be "group" of navigational
links for the skipping over requirement. Consider requirement of the
word
"repetitive" too.

Sailesh Panchang
Senior Accessibility Engineer
Deque Systems Inc. (www.deque.com)
11130 Sunrise Valley Drive, Suite #140,
Reston VA 20191
Phone: 703-225-0380 (ext 105)
E-mail: = EMAIL ADDRESS REMOVED =

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
Hoffman, Allen
Sent: Friday, June 08, 2007 1:28 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

Hmmm, thinking cap on.

I am not convinced "logical tab order" is so difficult to convert to
specification/standard format, and am not convinced that this one isn't
either, yet.

Web pages that are coded to:
allow correct keyboard operation;
associate header/cell information;
expose control element information;
generally can meet this repetitive navigation requirement, but
could still have block of content to navigate through.

Screen readers now generally do this by either using algorithms to
evaluate pages for repetitive int3erface elements, or by allowing the
user to hit t to jump to next table, h for next header, or other letters
for other element types to jump through. To allow this functionality
what is important is to use consistent tagging of blocks so that, for
example, when you hit <next whatever> it doesn't actually skip extra
content, or skip in to the middle of the current block.

Imagine:

header: text.
header: text.
table: information.
header: text.
unordered list: information
header: text
table: links.

What command would one use to jump from "block" to block without a
specific consistent identifier being used for the screen reader to grab
on to and pick up the next part of the page to read?


I'm just writing this to give us all food for thought as we try and
finalize this requirement.
thanks.




Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Peter
Wallack
Sent: Friday, June 08, 2007 1:13 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

Allen --
I like your explanation of the intent of this standard far better than
the current wording. I'm not at all keen on the phrase 'repeated on
multiple web pages' as that seems more directed toward a situation where
the user is taking a linear path through a site, which is often not the
case. How about simply:

"A mechanism shall be available to bypass blocks of content"

leaving us only to haggle over the best wording for 'block of content',
which is a concept that I think we all intrinsically understand but we'd
like something a bit more scientific.

As for testability, this probably is no more testable than having a tab
sequence that is 'logical'. This just takes it to one level higher:
having a logical division of blocks of content that you can sequence
through.

Peter Wallack
Accessibility Program Director
Oracle Corporation



Hoffman, Allen wrote:
> This one is intended to allow people to basically use things like
> headers to jump from each block to the next quickly. I'm not
> convinced this language is exactly right, or better than the original
> CFR1194.22(o) either. it is testable, but certainly is not easy to
> test in an automated fashion.
>
> I have been trying to work out testing algorithm to count the number
> of links from start of page to each link on a page, take the average
> link-count, and then by re-analyzing the links in variations of
> sequence, determine if shorter paths indeed exist, and where to set
> upper and lower boundaries on the tests to get the answer, is there a
> way to shorten the link path. if anyone can put that into a provision

> we would have a testable check on lowering "link" or "tab" path. The
> reason "block of content" is good is that it doesn't really matter
> what stuff one has to skip over to get past with the voice, the
> important factors is to be able to skip it quickly when needed.
>
>
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sean
> Hayes
> Sent: Friday, June 08, 2007 11:49 AM
> To: TEITAC Web/Software Subcommittee
> Subject: [teitac-websoftware] Bypassing content.
>
> I have a slight issue on the testability of : "A mechanism shall be
> available to bypass blocks of content that are repeated on multiple
> Web pages."
>
> What exactly constitutes a 'block of content'?
>
> Also if the repeated content is the last thing on the page, is a
> bypass still necessary? If so, where should it bypass too?
>
> Sean Hayes
> Standards and Policy Team
> Corporate Accessibility Group
> Microsoft
> Phone:
> mob +44 7977 455002
> office +44 117 9719730
>

From: David Poehlman
Date: Fri, Jun 08 2007 1:20 PM
Subject: Re: Bypassing content.

I don't like bypass.

----- Original Message -----
From: "Peter Wallack" < = EMAIL ADDRESS REMOVED = >
To: "TEITAC Web/Software Subcommittee" < = EMAIL ADDRESS REMOVED = >
Sent: Friday, June 08, 2007 1:12 PM
Subject: Re: [teitac-websoftware] Bypassing content.


Allen --
I like your explanation of the intent of this standard far better than
the current wording. I'm not at all keen on the phrase 'repeated on
multiple web pages' as that seems more directed toward a situation where
the user is taking a linear path through a site, which is often not the
case. How about simply:

"A mechanism shall be available to bypass blocks of content"

leaving us only to haggle over the best wording for 'block of content',
which is a concept that I think we all intrinsically understand but we'd
like something a bit more scientific.

As for testability, this probably is no more testable than having a tab
sequence that is 'logical'. This just takes it to one level higher:
having a logical division of blocks of content that you can sequence
through.

Peter Wallack
Accessibility Program Director
Oracle Corporation



Hoffman, Allen wrote:
> This one is intended to allow people to basically use things like
> headers to jump from each block to the next quickly. I'm not convinced
> this language is exactly right, or better than the original
> CFR1194.22(o) either. it is testable, but certainly is not easy to test
> in an automated fashion.
>
> I have been trying to work out testing algorithm to count the number of
> links from start of page to each link on a page, take the average
> link-count, and then by re-analyzing the links in variations of
> sequence, determine if shorter paths indeed exist, and where to set
> upper and lower boundaries on the tests to get the answer, is there a
> way to shorten the link path. if anyone can put that into a provision
> we would have a testable check on lowering "link" or "tab" path. The
> reason "block of content" is good is that it doesn't really matter what
> stuff one has to skip over to get past with the voice, the important
> factors is to be able to skip it quickly when needed.
>
>
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sean
> Hayes
> Sent: Friday, June 08, 2007 11:49 AM
> To: TEITAC Web/Software Subcommittee
> Subject: [teitac-websoftware] Bypassing content.
>
> I have a slight issue on the testability of : "A mechanism shall be
> available to bypass blocks of content that are repeated on multiple Web
> pages."
>
> What exactly constitutes a 'block of content'?
>
> Also if the repeated content is the last thing on the page, is a bypass
> still necessary? If so, where should it bypass too?
>
> Sean Hayes
> Standards and Policy Team
> Corporate Accessibility Group
> Microsoft
> Phone:
> mob +44 7977 455002
> office +44 117 9719730
>

From: Peter Wallack
Date: Fri, Jun 08 2007 3:25 PM
Subject: Re: Bypassing content.

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
It seems to me we're all debating two different concepts, both of which
are perfectly legitimate:<br>
<ul>
<li>the ability to bypass repetitive content</li>
<li>the ability to jump to blocks of content</li>
</ul>
Solving the latter one would probably suffice to solve the former,
though not necessarily optimally. The current 508 only addresses the
first issue. So the big question is: <b>are we merely refining the
current standard, or proposing an additional concept?</b><br>
<pre class="moz-signature" cols="72">Peter Wallack
Accessibility Program Director
Oracle Corporation</pre>
<br>
<br>
David Poehlman wrote:
<blockquote cite="mid:002a01c7a9ff$f66d6eb0$0901a8c0@HANDS" type="cite">
<pre wrap="">I don't like bypass.

----- Original Message -----
From: "Peter Wallack" <a class="moz-txt-link-rfc2396E" href="mailto: = EMAIL ADDRESS REMOVED = ">&lt; = EMAIL ADDRESS REMOVED = &gt;</a>
To: "TEITAC Web/Software Subcommittee" <a class="moz-txt-link-rfc2396E" href="mailto: = EMAIL ADDRESS REMOVED = ">&lt; = EMAIL ADDRESS REMOVED = &gt;</a>
Sent: Friday, June 08, 2007 1:12 PM
Subject: Re: [teitac-websoftware] Bypassing content.


Allen --
I like your explanation of the intent of this standard far better than
the current wording. I'm not at all keen on the phrase 'repeated on
multiple web pages' as that seems more directed toward a situation where
the user is taking a linear path through a site, which is often not the
case. How about simply:

"A mechanism shall be available to bypass blocks of content"

leaving us only to haggle over the best wording for 'block of content',
which is a concept that I think we all intrinsically understand but we'd
like something a bit more scientific.

As for testability, this probably is no more testable than having a tab
sequence that is 'logical'. This just takes it to one level higher:
having a logical division of blocks of content that you can sequence
through.

Peter Wallack
Accessibility Program Director
Oracle Corporation



Hoffman, Allen wrote:
</pre>
<blockquote type="cite">
<pre wrap="">This one is intended to allow people to basically use things like
headers to jump from each block to the next quickly. I'm not convinced
this language is exactly right, or better than the original
CFR1194.22(o) either. it is testable, but certainly is not easy to test
in an automated fashion.

I have been trying to work out testing algorithm to count the number of
links from start of page to each link on a page, take the average
link-count, and then by re-analyzing the links in variations of
sequence, determine if shorter paths indeed exist, and where to set
upper and lower boundaries on the tests to get the answer, is there a
way to shorten the link path. if anyone can put that into a provision
we would have a testable check on lowering "link" or "tab" path. The
reason "block of content" is good is that it doesn't really matter what
stuff one has to skip over to get past with the voice, the important
factors is to be able to skip it quickly when needed.






Allen Hoffman -- <a class="moz-txt-link-abbreviated" href="mailto: = EMAIL ADDRESS REMOVED = "> = EMAIL ADDRESS REMOVED = </a>; v: 202-447-0303

-----Original Message-----
From: <a class="moz-txt-link-abbreviated" href="mailto: = EMAIL ADDRESS REMOVED = "> = EMAIL ADDRESS REMOVED = </a>
[<a class="moz-txt-link-freetext" href="mailto: = EMAIL ADDRESS REMOVED = ">mailto: = EMAIL ADDRESS REMOVED = </a>] On Behalf Of Sean
Hayes
Sent: Friday, June 08, 2007 11:49 AM
To: TEITAC Web/Software Subcommittee
Subject: [teitac-websoftware] Bypassing content.

I have a slight issue on the testability of : "A mechanism shall be
available to bypass blocks of content that are repeated on multiple Web
pages."

What exactly constitutes a 'block of content'?

Also if the repeated content is the last thing on the page, is a bypass
still necessary? If so, where should it bypass too?

Sean Hayes
Standards and Policy Team
Corporate Accessibility Group
Microsoft
Phone:
mob +44 7977 455002
office +44 117 9719730

From: Travis Roth
Date: Fri, Jun 08 2007 4:55 PM
Subject: Re: Bypassing content.

The skipping of groups of links, such as navigational links, is also
beneficial to other keyboard users who cannot use a mouse but also do not
use a screen reader.

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sailesh
Panchang
Sent: Friday, June 08, 2007 2:02 PM
To: 'TEITAC Web/Software Subcommittee'
Subject: Re: [teitac-websoftware] Bypassing content.


This requirement is primarily designed to help users of screen reading
software.
User agents including assistive technologies provide ability to jump to
different elements (eg. lists, tables, frames,) on a page if structural
markup is used. Structural markup for tables, headings lists, etc. is
required by Para (g) under 6.1 that reads "Information and relationships
conveyed through presentation shall be programmatically determinable, and
notification of changes to these is available to...". So using structural
markup facilitates AT to skip over blocks / skip to desired block.
Therefore I believe the skip block requirement being debated over
predominantly applies to navigational links. This was the intent of the 2001
S508 standards too.
Note that when groups of links are placed in a list or frame, etc. a
mechanism to skip over them is available to screen reader users- an explicit
skip nav link is not needed.

Conclusion: So I think the focus should just be "group" of navigational
links for the skipping over requirement. Consider requirement of the word
"repetitive" too.

Sailesh Panchang
Senior Accessibility Engineer
Deque Systems Inc. (www.deque.com)
11130 Sunrise Valley Drive, Suite #140,
Reston VA 20191
Phone: 703-225-0380 (ext 105)
E-mail: = EMAIL ADDRESS REMOVED =

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Hoffman,
Allen
Sent: Friday, June 08, 2007 1:28 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

Hmmm, thinking cap on.

I am not convinced "logical tab order" is so difficult to convert to
specification/standard format, and am not convinced that this one isn't
either, yet.

Web pages that are coded to:
allow correct keyboard operation;
associate header/cell information;
expose control element information;
generally can meet this repetitive navigation requirement, but could
still have block of content to navigate through.

Screen readers now generally do this by either using algorithms to evaluate
pages for repetitive int3erface elements, or by allowing the user to hit t
to jump to next table, h for next header, or other letters for other element
types to jump through. To allow this functionality what is important is to
use consistent tagging of blocks so that, for example, when you hit <next
whatever> it doesn't actually skip extra content, or skip in to the middle
of the current block.

Imagine:

header: text.
header: text.
table: information.
header: text.
unordered list: information
header: text
table: links.

What command would one use to jump from "block" to block without a specific
consistent identifier being used for the screen reader to grab on to and
pick up the next part of the page to read?


I'm just writing this to give us all food for thought as we try and finalize
this requirement. thanks.




Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Peter
Wallack
Sent: Friday, June 08, 2007 1:13 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

Allen --
I like your explanation of the intent of this standard far better than the
current wording. I'm not at all keen on the phrase 'repeated on multiple web
pages' as that seems more directed toward a situation where the user is
taking a linear path through a site, which is often not the case. How about
simply:

"A mechanism shall be available to bypass blocks of content"

leaving us only to haggle over the best wording for 'block of content',
which is a concept that I think we all intrinsically understand but we'd
like something a bit more scientific.

As for testability, this probably is no more testable than having a tab
sequence that is 'logical'. This just takes it to one level higher:
having a logical division of blocks of content that you can sequence
through.

Peter Wallack
Accessibility Program Director
Oracle Corporation



Hoffman, Allen wrote:
> This one is intended to allow people to basically use things like
> headers to jump from each block to the next quickly. I'm not
> convinced this language is exactly right, or better than the original
> CFR1194.22(o) either. it is testable, but certainly is not easy to
> test in an automated fashion.
>
> I have been trying to work out testing algorithm to count the number
> of links from start of page to each link on a page, take the average
> link-count, and then by re-analyzing the links in variations of
> sequence, determine if shorter paths indeed exist, and where to set
> upper and lower boundaries on the tests to get the answer, is there a
> way to shorten the link path. if anyone can put that into a provision

> we would have a testable check on lowering "link" or "tab" path. The
> reason "block of content" is good is that it doesn't really matter
> what stuff one has to skip over to get past with the voice, the
> important factors is to be able to skip it quickly when needed.
>
>
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sean
> Hayes
> Sent: Friday, June 08, 2007 11:49 AM
> To: TEITAC Web/Software Subcommittee
> Subject: [teitac-websoftware] Bypassing content.
>
> I have a slight issue on the testability of : "A mechanism shall be
> available to bypass blocks of content that are repeated on multiple
> Web pages."
>
> What exactly constitutes a 'block of content'?
>
> Also if the repeated content is the last thing on the page, is a
> bypass still necessary? If so, where should it bypass too?
>
> Sean Hayes
> Standards and Policy Team
> Corporate Accessibility Group
> Microsoft
> Phone:
> mob +44 7977 455002
> office +44 117 9719730
>

From: David Poehlman
Date: Sat, Jun 09 2007 7:00 AM
Subject: Re: Bypassing content.

My point here is it boils down to navigation so it would be most helpful for
the standard to suggest that navigation to the most important new content on
a page would be appropriate or even navigation to the start of new content
on the page but we do need to combine the two into one.

----- Original Message -----
From: "Peter Wallack" < = EMAIL ADDRESS REMOVED = >
To: "TEITAC Web/Software Subcommittee" < = EMAIL ADDRESS REMOVED = >
Sent: Friday, June 08, 2007 5:20 PM
Subject: Re: [teitac-websoftware] Bypassing content.


It seems to me we're all debating two different concepts, both of which are
perfectly legitimate:

a.. the ability to bypass repetitive content
b.. the ability to jump to blocks of content
Solving the latter one would probably suffice to solve the former, though
not necessarily optimally. The current 508 only addresses the first issue.
So the big question is: are we merely refining the current standard, or
proposing an additional concept?

Peter Wallack
Accessibility Program Director
Oracle Corporation

David Poehlman wrote:
I don't like bypass.

----- Original Message -----
From: "Peter Wallack" < = EMAIL ADDRESS REMOVED = >
To: "TEITAC Web/Software Subcommittee" < = EMAIL ADDRESS REMOVED = >
Sent: Friday, June 08, 2007 1:12 PM
Subject: Re: [teitac-websoftware] Bypassing content.


Allen --
I like your explanation of the intent of this standard far better than
the current wording. I'm not at all keen on the phrase 'repeated on
multiple web pages' as that seems more directed toward a situation where
the user is taking a linear path through a site, which is often not the
case. How about simply:

"A mechanism shall be available to bypass blocks of content"

leaving us only to haggle over the best wording for 'block of content',
which is a concept that I think we all intrinsically understand but we'd
like something a bit more scientific.

As for testability, this probably is no more testable than having a tab
sequence that is 'logical'. This just takes it to one level higher:
having a logical division of blocks of content that you can sequence
through.

Peter Wallack
Accessibility Program Director
Oracle Corporation



Hoffman, Allen wrote:
This one is intended to allow people to basically use things like
headers to jump from each block to the next quickly. I'm not convinced
this language is exactly right, or better than the original
CFR1194.22(o) either. it is testable, but certainly is not easy to test
in an automated fashion.

I have been trying to work out testing algorithm to count the number of
links from start of page to each link on a page, take the average
link-count, and then by re-analyzing the links in variations of
sequence, determine if shorter paths indeed exist, and where to set
upper and lower boundaries on the tests to get the answer, is there a
way to shorten the link path. if anyone can put that into a provision
we would have a testable check on lowering "link" or "tab" path. The
reason "block of content" is good is that it doesn't really matter what
stuff one has to skip over to get past with the voice, the important
factors is to be able to skip it quickly when needed.






Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sean
Hayes
Sent: Friday, June 08, 2007 11:49 AM
To: TEITAC Web/Software Subcommittee
Subject: [teitac-websoftware] Bypassing content.

I have a slight issue on the testability of : "A mechanism shall be
available to bypass blocks of content that are repeated on multiple Web
pages."

What exactly constitutes a 'block of content'?

Also if the repeated content is the last thing on the page, is a bypass
still necessary? If so, where should it bypass too?

Sean Hayes
Standards and Policy Team
Corporate Accessibility Group
Microsoft
Phone:
mob +44 7977 455002
office +44 117 9719730

From: Hoffman, Allen
Date: Mon, Jun 11 2007 7:10 AM
Subject: Re: Bypassing content.

I think we are very close to two distinct functionalities. They are
just very closely related.

it might be a good idea to consider as a recommendation adding the extra
functional requirement as a separate one after all. i can see merit in
that approach.



Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

From: Andrew Kirkpatrick
Date: Mon, Jun 11 2007 7:15 AM
Subject: Re: Bypassing content.

I'm willing to bet that most "repetitive content" is not completely
repetitive. We should be careful about how we define this - for
example, if the block of content is a navigation bar and when you click
on a link in that navigation bar you get a page that has almost the same
navigation bar, except that there is some small difference, such as a
different set of subnavigation links that are in view or something
similar. Is that still repetitive?

AWK

From: Hoffman, Allen
Date: Mon, Jun 11 2007 7:20 AM
Subject: Re: Bypassing content.

You are most definitely correct Andrew, its "nearly repetitive", or
"very similar" patterns of content.



Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

From: Hoffman, Allen
Date: Mon, Jun 11 2007 7:50 AM
Subject: Re: Bypassing content.

Something we have missed in our language is that the method must also be
visible. This is so often overlooked, or intentionally not wanted that
it MUST be said.




Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Travis
Roth
Sent: Friday, June 08, 2007 6:49 PM
To: = EMAIL ADDRESS REMOVED = ; 'TEITAC Web/Software Subcommittee'
Subject: Re: [teitac-websoftware] Bypassing content.

The skipping of groups of links, such as navigational links, is also
beneficial to other keyboard users who cannot use a mouse but also do
not use a screen reader.

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sailesh
Panchang
Sent: Friday, June 08, 2007 2:02 PM
To: 'TEITAC Web/Software Subcommittee'
Subject: Re: [teitac-websoftware] Bypassing content.


This requirement is primarily designed to help users of screen reading
software.
User agents including assistive technologies provide ability to jump to
different elements (eg. lists, tables, frames,) on a page if structural
markup is used. Structural markup for tables, headings lists, etc. is
required by Para (g) under 6.1 that reads "Information and relationships
conveyed through presentation shall be programmatically determinable,
and notification of changes to these is available to...". So using
structural markup facilitates AT to skip over blocks / skip to desired
block.
Therefore I believe the skip block requirement being debated over
predominantly applies to navigational links. This was the intent of the
2001
S508 standards too.
Note that when groups of links are placed in a list or frame, etc. a
mechanism to skip over them is available to screen reader users- an
explicit skip nav link is not needed.

Conclusion: So I think the focus should just be "group" of navigational
links for the skipping over requirement. Consider requirement of the
word
"repetitive" too.

Sailesh Panchang
Senior Accessibility Engineer
Deque Systems Inc. (www.deque.com)
11130 Sunrise Valley Drive, Suite #140,
Reston VA 20191
Phone: 703-225-0380 (ext 105)
E-mail: = EMAIL ADDRESS REMOVED =

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
Hoffman, Allen
Sent: Friday, June 08, 2007 1:28 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

Hmmm, thinking cap on.

I am not convinced "logical tab order" is so difficult to convert to
specification/standard format, and am not convinced that this one isn't
either, yet.

Web pages that are coded to:
allow correct keyboard operation;
associate header/cell information;
expose control element information;
generally can meet this repetitive navigation requirement, but
could still have block of content to navigate through.

Screen readers now generally do this by either using algorithms to
evaluate pages for repetitive int3erface elements, or by allowing the
user to hit t to jump to next table, h for next header, or other letters
for other element types to jump through. To allow this functionality
what is important is to use consistent tagging of blocks so that, for
example, when you hit <next
whatever> it doesn't actually skip extra content, or skip in to the
whatever> middle
of the current block.

Imagine:

header: text.
header: text.
table: information.
header: text.
unordered list: information
header: text
table: links.

What command would one use to jump from "block" to block without a
specific consistent identifier being used for the screen reader to grab
on to and pick up the next part of the page to read?


I'm just writing this to give us all food for thought as we try and
finalize this requirement. thanks.




Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Peter
Wallack
Sent: Friday, June 08, 2007 1:13 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

Allen --
I like your explanation of the intent of this standard far better than
the current wording. I'm not at all keen on the phrase 'repeated on
multiple web pages' as that seems more directed toward a situation where
the user is taking a linear path through a site, which is often not the
case. How about
simply:

"A mechanism shall be available to bypass blocks of content"

leaving us only to haggle over the best wording for 'block of content',
which is a concept that I think we all intrinsically understand but we'd
like something a bit more scientific.

As for testability, this probably is no more testable than having a tab
sequence that is 'logical'. This just takes it to one level higher:
having a logical division of blocks of content that you can sequence
through.

Peter Wallack
Accessibility Program Director
Oracle Corporation



Hoffman, Allen wrote:
> This one is intended to allow people to basically use things like
> headers to jump from each block to the next quickly. I'm not
> convinced this language is exactly right, or better than the original
> CFR1194.22(o) either. it is testable, but certainly is not easy to
> test in an automated fashion.
>
> I have been trying to work out testing algorithm to count the number
> of links from start of page to each link on a page, take the average
> link-count, and then by re-analyzing the links in variations of
> sequence, determine if shorter paths indeed exist, and where to set
> upper and lower boundaries on the tests to get the answer, is there a
> way to shorten the link path. if anyone can put that into a provision

> we would have a testable check on lowering "link" or "tab" path. The
> reason "block of content" is good is that it doesn't really matter
> what stuff one has to skip over to get past with the voice, the
> important factors is to be able to skip it quickly when needed.
>
>
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sean
> Hayes
> Sent: Friday, June 08, 2007 11:49 AM
> To: TEITAC Web/Software Subcommittee
> Subject: [teitac-websoftware] Bypassing content.
>
> I have a slight issue on the testability of : "A mechanism shall be
> available to bypass blocks of content that are repeated on multiple
> Web pages."
>
> What exactly constitutes a 'block of content'?
>
> Also if the repeated content is the last thing on the page, is a
> bypass still necessary? If so, where should it bypass too?
>
> Sean Hayes
> Standards and Policy Team
> Corporate Accessibility Group
> Microsoft
> Phone:
> mob +44 7977 455002
> office +44 117 9719730
>

From: Barrett, Don
Date: Mon, Jun 11 2007 7:55 AM
Subject: Re: Bypassing content.

Not, it is not.


Don Barrett
Section 508 Coordinator
U.S. Department of Education
(202)-205-8245
= EMAIL ADDRESS REMOVED =




_____

From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Andrew
Kirkpatrick
Sent: Monday, June 11, 2007 9:12 AM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.


I'm willing to bet that most "repetitive content" is not completely
repetitive. We should be careful about how we define this - for
example, if the block of content is a navigation bar and when you click
on a link in that navigation bar you get a page that has almost the same
navigation bar, except that there is some small difference, such as a
different set of subnavigation links that are in view or something
similar. Is that still repetitive?

AWK


_____

From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
Hoffman, Allen
Sent: Monday, June 11, 2007 9:06 AM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.


I think we are very close to two distinct functionalities. They
are just very closely related.

it might be a good idea to consider as a recommendation adding
the extra functional requirement as a separate one after all. i can see
merit in that approach.



Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303



_____

From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Peter
Wallack
Sent: Friday, June 08, 2007 5:21 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.


It seems to me we're all debating two different concepts, both
of which are perfectly legitimate:


* the ability to bypass repetitive content
* the ability to jump to blocks of content

Solving the latter one would probably suffice to solve the
former, though not necessarily optimally. The current 508 only addresses
the first issue. So the big question is: are we merely refining the
current standard, or proposing an additional concept?

Peter Wallack
Accessibility Program Director
Oracle Corporation


David Poehlman wrote:

I don't like bypass.

----- Original Message -----
From: "Peter Wallack" < = EMAIL ADDRESS REMOVED = >
<mailto: = EMAIL ADDRESS REMOVED = >
To: "TEITAC Web/Software Subcommittee"
< = EMAIL ADDRESS REMOVED = >
<mailto: = EMAIL ADDRESS REMOVED = >
Sent: Friday, June 08, 2007 1:12 PM
Subject: Re: [teitac-websoftware] Bypassing content.


Allen --
I like your explanation of the intent of this standard
far better than
the current wording. I'm not at all keen on the phrase
'repeated on
multiple web pages' as that seems more directed toward a
situation where
the user is taking a linear path through a site, which
is often not the
case. How about simply:

"A mechanism shall be available to bypass blocks of
content"

leaving us only to haggle over the best wording for
'block of content',
which is a concept that I think we all intrinsically
understand but we'd
like something a bit more scientific.

As for testability, this probably is no more testable
than having a tab
sequence that is 'logical'. This just takes it to one
level higher:
having a logical division of blocks of content that you
can sequence
through.

Peter Wallack
Accessibility Program Director
Oracle Corporation



Hoffman, Allen wrote:


This one is intended to allow people to
basically use things like
headers to jump from each block to the next
quickly. I'm not convinced
this language is exactly right, or better than
the original
CFR1194.22(o) either. it is testable, but
certainly is not easy to test
in an automated fashion.

I have been trying to work out testing algorithm
to count the number of
links from start of page to each link on a page,
take the average
link-count, and then by re-analyzing the links
in variations of
sequence, determine if shorter paths indeed
exist, and where to set
upper and lower boundaries on the tests to get
the answer, is there a
way to shorten the link path. if anyone can put
that into a provision
we would have a testable check on lowering
"link" or "tab" path. The
reason "block of content" is good is that it
doesn't really matter what
stuff one has to skip over to get past with the
voice, the important
factors is to be able to skip it quickly when
needed.






Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v:
202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =

[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sean
Hayes
Sent: Friday, June 08, 2007 11:49 AM
To: TEITAC Web/Software Subcommittee
Subject: [teitac-websoftware] Bypassing content.

I have a slight issue on the testability of : "A
mechanism shall be
available to bypass blocks of content that are
repeated on multiple Web
pages."

What exactly constitutes a 'block of content'?

Also if the repeated content is the last thing
on the page, is a bypass
still necessary? If so, where should it bypass
too?

Sean Hayes
Standards and Policy Team
Corporate Accessibility Group
Microsoft
Phone:
mob +44 7977 455002
office +44 117 9719730

From: Sean Hayes
Date: Mon, Jun 11 2007 8:00 AM
Subject: Re: Bypassing content.

So now we seem to be iterating towards "bits of content of indeterminate size which are almost but not exactly the same as content you may or may not have seen on other nearby web pages".

That doesn't sound very testable to me :)

I suggest either we focus on the specific functionality:
Web pages must provide a mechanism to bypass navigational content in situations where in a linear presentation or sequential navigation it would occur before the informational content.

Or we go to the more general one I suggested before.

Sean Hayes
Standards and Policy Team
Corporate Accessibility Group
Microsoft
Phone:
mob +44 7977 455002
office +44 117 9719730

From: Barrett, Don
Date: Mon, Jun 11 2007 8:35 AM
Subject: Re: Bypassing content.

Can we really require developers to use heading structures on all pages
just so we can satisfy the programmatically determinable aspect of the
site?


Don Barrett
Section 508 Coordinator
U.S. Department of Education
(202)-205-8245
= EMAIL ADDRESS REMOVED =

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Andrew
Kirkpatrick
Sent: Monday, June 11, 2007 10:29 AM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

Allen,
Some methods to meet this requirement might not be visible - for
example, providing a heading structure that allows users to navigate to
different areas of the content. I don't think that we should be think
just about "skip navigation" links.

AWK

> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> Hoffman, Allen
> Sent: Monday, June 11, 2007 9:33 AM
> To: = EMAIL ADDRESS REMOVED = ; TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Something we have missed in our language is that the method must also
> be visible. This is so often overlooked, or intentionally not wanted
> that it MUST be said.
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> Travis Roth
> Sent: Friday, June 08, 2007 6:49 PM
> To: = EMAIL ADDRESS REMOVED = ; 'TEITAC Web/Software Subcommittee'
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> The skipping of groups of links, such as navigational links, is also
> beneficial to other keyboard users who cannot use a mouse but also do
> not use a screen reader.
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> Sailesh Panchang
> Sent: Friday, June 08, 2007 2:02 PM
> To: 'TEITAC Web/Software Subcommittee'
> Subject: Re: [teitac-websoftware] Bypassing content.
>
>
> This requirement is primarily designed to help users of screen reading

> software.
> User agents including assistive technologies provide ability to jump
> to different elements (eg. lists, tables, frames,) on a page if
> structural markup is used. Structural markup for tables, headings
> lists, etc. is required by Para (g) under
> 6.1 that reads "Information and relationships conveyed through
> presentation shall be programmatically determinable, and notification
> of changes to these is available to...". So using structural markup
> facilitates AT to skip over blocks / skip to desired block.
> Therefore I believe the skip block requirement being debated over
> predominantly applies to navigational links. This was the intent of
> the
> 2001
> S508 standards too.
> Note that when groups of links are placed in a list or frame, etc. a
> mechanism to skip over them is available to screen reader users- an
> explicit skip nav link is not needed.
>
> Conclusion: So I think the focus should just be "group" of
> navigational links for the skipping over requirement.
> Consider requirement of the word "repetitive" too.
>
> Sailesh Panchang
> Senior Accessibility Engineer
> Deque Systems Inc. (www.deque.com)
> 11130 Sunrise Valley Drive, Suite #140, Reston VA 20191
> Phone: 703-225-0380 (ext 105)
> E-mail: = EMAIL ADDRESS REMOVED =
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> Hoffman, Allen
> Sent: Friday, June 08, 2007 1:28 PM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Hmmm, thinking cap on.
>
> I am not convinced "logical tab order" is so difficult to convert to
> specification/standard format, and am not convinced that this one
> isn't either, yet.
>
> Web pages that are coded to:
> allow correct keyboard operation;
> associate header/cell information;
> expose control element information;
> generally can meet this repetitive navigation requirement, but
could
> still have block of content to navigate through.
>
> Screen readers now generally do this by either using algorithms to
> evaluate pages for repetitive int3erface elements, or by allowing the
> user to hit t to jump to next table, h for next header, or other
> letters for other element types to jump through. To allow this
> functionality what is important is to use consistent tagging of blocks

> so that, for example, when you hit <next
> whatever> it doesn't actually skip extra content, or skip in to the
> whatever> middle
> of the current block.
>
> Imagine:
>
> header: text.
> header: text.
> table: information.
> header: text.
> unordered list: information
> header: text
> table: links.
>
> What command would one use to jump from "block" to block without a
> specific consistent identifier being used for the screen reader to
> grab on to and pick up the next part of the page to read?
>
>
> I'm just writing this to give us all food for thought as we
> try and finalize this requirement. thanks.
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Peter Wallack
> Sent: Friday, June 08, 2007 1:13 PM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Allen --
> I like your explanation of the intent of this standard far
> better than the current wording. I'm not at all keen on the
> phrase 'repeated on multiple web pages' as that seems more
> directed toward a situation where the user is taking a linear
> path through a site, which is often not the case. How about
> simply:
>
> "A mechanism shall be available to bypass blocks of content"
>
> leaving us only to haggle over the best wording for 'block of
> content', which is a concept that I think we all
> intrinsically understand but we'd like something a bit more
> scientific.
>
> As for testability, this probably is no more testable than
> having a tab sequence that is 'logical'. This just takes it
> to one level higher:
> having a logical division of blocks of content that you can
> sequence through.
>
> Peter Wallack
> Accessibility Program Director
> Oracle Corporation
>
>
>
> Hoffman, Allen wrote:
> > This one is intended to allow people to basically use things like
> > headers to jump from each block to the next quickly. I'm not
> > convinced this language is exactly right, or better than
> the original
> > CFR1194.22(o) either. it is testable, but certainly is not easy to
> > test in an automated fashion.
> >
> > I have been trying to work out testing algorithm to count
> the number
> > of links from start of page to each link on a page, take
> the average
> > link-count, and then by re-analyzing the links in variations of
> > sequence, determine if shorter paths indeed exist, and where to set
> > upper and lower boundaries on the tests to get the answer,
> is there a
> > way to shorten the link path. if anyone can put that into
> a provision
>
> > we would have a testable check on lowering "link" or "tab"
> path. The
> > reason "block of content" is good is that it doesn't really matter
> > what stuff one has to skip over to get past with the voice, the
> > important factors is to be able to skip it quickly when needed.
> >
> >
> >
> >
> >
> >
> > Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
> >
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On
> Behalf Of Sean
> > Hayes
> > Sent: Friday, June 08, 2007 11:49 AM
> > To: TEITAC Web/Software Subcommittee
> > Subject: [teitac-websoftware] Bypassing content.
> >
> > I have a slight issue on the testability of : "A mechanism shall be
> > available to bypass blocks of content that are repeated on multiple
> > Web pages."
> >
> > What exactly constitutes a 'block of content'?
> >
> > Also if the repeated content is the last thing on the page, is a
> > bypass still necessary? If so, where should it bypass too?
> >
> > Sean Hayes
> > Standards and Policy Team
> > Corporate Accessibility Group
> > Microsoft
> > Phone:
> > mob +44 7977 455002
> > office +44 117 9719730
> >

From: Andrew Kirkpatrick
Date: Mon, Jun 11 2007 8:40 AM
Subject: Re: Bypassing content.

No, but we can't make this technique not sufficient by requiring that
the solution needs to visible either?
AWK

> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Barrett, Don
> Sent: Monday, June 11, 2007 10:31 AM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Can we really require developers to use heading structures on
> all pages just so we can satisfy the programmatically
> determinable aspect of the site?
>
>
> Don Barrett
> Section 508 Coordinator
> U.S. Department of Education
> (202)-205-8245
> = EMAIL ADDRESS REMOVED =
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Andrew Kirkpatrick
> Sent: Monday, June 11, 2007 10:29 AM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Allen,
> Some methods to meet this requirement might not be visible -
> for example, providing a heading structure that allows users
> to navigate to different areas of the content. I don't think
> that we should be think just about "skip navigation" links.
>
> AWK
>
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> > Hoffman, Allen
> > Sent: Monday, June 11, 2007 9:33 AM
> > To: = EMAIL ADDRESS REMOVED = ; TEITAC Web/Software Subcommittee
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> > Something we have missed in our language is that the method
> must also
> > be visible. This is so often overlooked, or intentionally
> not wanted
> > that it MUST be said.
> >
> >
> >
> >
> > Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
> >
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> > Travis Roth
> > Sent: Friday, June 08, 2007 6:49 PM
> > To: = EMAIL ADDRESS REMOVED = ; 'TEITAC Web/Software Subcommittee'
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> > The skipping of groups of links, such as navigational
> links, is also
> > beneficial to other keyboard users who cannot use a mouse
> but also do
> > not use a screen reader.
> >
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> > Sailesh Panchang
> > Sent: Friday, June 08, 2007 2:02 PM
> > To: 'TEITAC Web/Software Subcommittee'
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> >
> > This requirement is primarily designed to help users of
> screen reading
>
> > software.
> > User agents including assistive technologies provide
> ability to jump
> > to different elements (eg. lists, tables, frames,) on a page if
> > structural markup is used. Structural markup for tables, headings
> > lists, etc. is required by Para (g) under
> > 6.1 that reads "Information and relationships conveyed through
> > presentation shall be programmatically determinable, and
> notification
> > of changes to these is available to...". So using structural markup
> > facilitates AT to skip over blocks / skip to desired block.
> > Therefore I believe the skip block requirement being debated over
> > predominantly applies to navigational links. This was the intent of
> > the
> > 2001
> > S508 standards too.
> > Note that when groups of links are placed in a list or
> frame, etc. a
> > mechanism to skip over them is available to screen reader users- an
> > explicit skip nav link is not needed.
> >
> > Conclusion: So I think the focus should just be "group" of
> > navigational links for the skipping over requirement.
> > Consider requirement of the word "repetitive" too.
> >
> > Sailesh Panchang
> > Senior Accessibility Engineer
> > Deque Systems Inc. (www.deque.com)
> > 11130 Sunrise Valley Drive, Suite #140, Reston VA 20191
> > Phone: 703-225-0380 (ext 105)
> > E-mail: = EMAIL ADDRESS REMOVED =
> >
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> > Hoffman, Allen
> > Sent: Friday, June 08, 2007 1:28 PM
> > To: TEITAC Web/Software Subcommittee
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> > Hmmm, thinking cap on.
> >
> > I am not convinced "logical tab order" is so difficult to
> convert to
> > specification/standard format, and am not convinced that this one
> > isn't either, yet.
> >
> > Web pages that are coded to:
> > allow correct keyboard operation;
> > associate header/cell information;
> > expose control element information;
> > generally can meet this repetitive navigation requirement, but
> could
> > still have block of content to navigate through.
> >
> > Screen readers now generally do this by either using algorithms to
> > evaluate pages for repetitive int3erface elements, or by
> allowing the
> > user to hit t to jump to next table, h for next header, or other
> > letters for other element types to jump through. To allow this
> > functionality what is important is to use consistent
> tagging of blocks
>
> > so that, for example, when you hit <next
> > whatever> it doesn't actually skip extra content, or skip in to the
> > whatever> middle
> > of the current block.
> >
> > Imagine:
> >
> > header: text.
> > header: text.
> > table: information.
> > header: text.
> > unordered list: information
> > header: text
> > table: links.
> >
> > What command would one use to jump from "block" to block without a
> > specific consistent identifier being used for the screen reader to
> > grab on to and pick up the next part of the page to read?
> >
> >
> > I'm just writing this to give us all food for thought as we try and
> > finalize this requirement. thanks.
> >
> >
> >
> >
> > Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
> >
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On
> Behalf Of Peter
> > Wallack
> > Sent: Friday, June 08, 2007 1:13 PM
> > To: TEITAC Web/Software Subcommittee
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> > Allen --
> > I like your explanation of the intent of this standard far
> better than
> > the current wording. I'm not at all keen on the phrase 'repeated on
> > multiple web pages' as that seems more directed toward a situation
> > where the user is taking a linear path through a site,
> which is often
> > not the case. How about
> > simply:
> >
> > "A mechanism shall be available to bypass blocks of content"
> >
> > leaving us only to haggle over the best wording for 'block of
> > content', which is a concept that I think we all intrinsically
> > understand but we'd like something a bit more scientific.
> >
> > As for testability, this probably is no more testable than having a
> > tab sequence that is 'logical'. This just takes it to one level
> > higher:
> > having a logical division of blocks of content that you can
> sequence
> > through.
> >
> > Peter Wallack
> > Accessibility Program Director
> > Oracle Corporation
> >
> >
> >
> > Hoffman, Allen wrote:
> > > This one is intended to allow people to basically use things like
> > > headers to jump from each block to the next quickly. I'm not
> > > convinced this language is exactly right, or better than
> > the original
> > > CFR1194.22(o) either. it is testable, but certainly is
> not easy to
> > > test in an automated fashion.
> > >
> > > I have been trying to work out testing algorithm to count
> > the number
> > > of links from start of page to each link on a page, take
> > the average
> > > link-count, and then by re-analyzing the links in variations of
> > > sequence, determine if shorter paths indeed exist, and
> where to set
> > > upper and lower boundaries on the tests to get the answer,
> > is there a
> > > way to shorten the link path. if anyone can put that into
> > a provision
> >
> > > we would have a testable check on lowering "link" or "tab"
> > path. The
> > > reason "block of content" is good is that it doesn't
> really matter
> > > what stuff one has to skip over to get past with the voice, the
> > > important factors is to be able to skip it quickly when needed.
> > >
> > >
> > >
> > >
> > >
> > >
> > > Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
> > >
> > > -----Original Message-----
> > > From: = EMAIL ADDRESS REMOVED =
> > > [mailto: = EMAIL ADDRESS REMOVED = ] On
> > Behalf Of Sean
> > > Hayes
> > > Sent: Friday, June 08, 2007 11:49 AM
> > > To: TEITAC Web/Software Subcommittee
> > > Subject: [teitac-websoftware] Bypassing content.
> > >
> > > I have a slight issue on the testability of : "A
> mechanism shall be
> > > available to bypass blocks of content that are repeated
> on multiple
> > > Web pages."
> > >
> > > What exactly constitutes a 'block of content'?
> > >
> > > Also if the repeated content is the last thing on the page, is a
> > > bypass still necessary? If so, where should it bypass too?
> > >
> > > Sean Hayes
> > > Standards and Policy Team
> > > Corporate Accessibility Group
> > > Microsoft
> > > Phone:
> > > mob +44 7977 455002
> > > office +44 117 9719730
> > >

From: Andrew Kirkpatrick
Date: Mon, Jun 11 2007 8:45 AM
Subject: Re: Bypassing content.

Allen,
Some methods to meet this requirement might not be visible - for
example, providing a heading structure that allows users to navigate to
different areas of the content. I don't think that we should be think
just about "skip navigation" links.

AWK

> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Hoffman, Allen
> Sent: Monday, June 11, 2007 9:33 AM
> To: = EMAIL ADDRESS REMOVED = ; TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Something we have missed in our language is that the method
> must also be visible. This is so often overlooked, or
> intentionally not wanted that it MUST be said.
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Travis Roth
> Sent: Friday, June 08, 2007 6:49 PM
> To: = EMAIL ADDRESS REMOVED = ; 'TEITAC Web/Software Subcommittee'
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> The skipping of groups of links, such as navigational links,
> is also beneficial to other keyboard users who cannot use a
> mouse but also do not use a screen reader.
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Sailesh Panchang
> Sent: Friday, June 08, 2007 2:02 PM
> To: 'TEITAC Web/Software Subcommittee'
> Subject: Re: [teitac-websoftware] Bypassing content.
>
>
> This requirement is primarily designed to help users of
> screen reading software.
> User agents including assistive technologies provide ability
> to jump to different elements (eg. lists, tables, frames,) on
> a page if structural markup is used. Structural markup for
> tables, headings lists, etc. is required by Para (g) under
> 6.1 that reads "Information and relationships conveyed
> through presentation shall be programmatically determinable,
> and notification of changes to these is available to...". So
> using structural markup facilitates AT to skip over blocks /
> skip to desired block.
> Therefore I believe the skip block requirement being debated
> over predominantly applies to navigational links. This was
> the intent of the
> 2001
> S508 standards too.
> Note that when groups of links are placed in a list or frame,
> etc. a mechanism to skip over them is available to screen
> reader users- an explicit skip nav link is not needed.
>
> Conclusion: So I think the focus should just be "group" of
> navigational links for the skipping over requirement.
> Consider requirement of the word "repetitive" too.
>
> Sailesh Panchang
> Senior Accessibility Engineer
> Deque Systems Inc. (www.deque.com)
> 11130 Sunrise Valley Drive, Suite #140,
> Reston VA 20191
> Phone: 703-225-0380 (ext 105)
> E-mail: = EMAIL ADDRESS REMOVED =
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Hoffman, Allen
> Sent: Friday, June 08, 2007 1:28 PM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Hmmm, thinking cap on.
>
> I am not convinced "logical tab order" is so difficult to
> convert to specification/standard format, and am not
> convinced that this one isn't either, yet.
>
> Web pages that are coded to:
> allow correct keyboard operation;
> associate header/cell information;
> expose control element information;
> generally can meet this repetitive navigation
> requirement, but could still have block of content to
> navigate through.
>
> Screen readers now generally do this by either using
> algorithms to evaluate pages for repetitive int3erface
> elements, or by allowing the user to hit t to jump to next
> table, h for next header, or other letters for other element
> types to jump through. To allow this functionality what is
> important is to use consistent tagging of blocks so that, for
> example, when you hit <next
> whatever> it doesn't actually skip extra content, or skip in to the
> whatever> middle
> of the current block.
>
> Imagine:
>
> header: text.
> header: text.
> table: information.
> header: text.
> unordered list: information
> header: text
> table: links.
>
> What command would one use to jump from "block" to block
> without a specific consistent identifier being used for the
> screen reader to grab on to and pick up the next part of the
> page to read?
>
>
> I'm just writing this to give us all food for thought as we
> try and finalize this requirement. thanks.
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Peter Wallack
> Sent: Friday, June 08, 2007 1:13 PM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Allen --
> I like your explanation of the intent of this standard far
> better than the current wording. I'm not at all keen on the
> phrase 'repeated on multiple web pages' as that seems more
> directed toward a situation where the user is taking a linear
> path through a site, which is often not the case. How about
> simply:
>
> "A mechanism shall be available to bypass blocks of content"
>
> leaving us only to haggle over the best wording for 'block of
> content', which is a concept that I think we all
> intrinsically understand but we'd like something a bit more
> scientific.
>
> As for testability, this probably is no more testable than
> having a tab sequence that is 'logical'. This just takes it
> to one level higher:
> having a logical division of blocks of content that you can
> sequence through.
>
> Peter Wallack
> Accessibility Program Director
> Oracle Corporation
>
>
>
> Hoffman, Allen wrote:
> > This one is intended to allow people to basically use things like
> > headers to jump from each block to the next quickly. I'm not
> > convinced this language is exactly right, or better than
> the original
> > CFR1194.22(o) either. it is testable, but certainly is not easy to
> > test in an automated fashion.
> >
> > I have been trying to work out testing algorithm to count
> the number
> > of links from start of page to each link on a page, take
> the average
> > link-count, and then by re-analyzing the links in variations of
> > sequence, determine if shorter paths indeed exist, and where to set
> > upper and lower boundaries on the tests to get the answer,
> is there a
> > way to shorten the link path. if anyone can put that into
> a provision
>
> > we would have a testable check on lowering "link" or "tab"
> path. The
> > reason "block of content" is good is that it doesn't really matter
> > what stuff one has to skip over to get past with the voice, the
> > important factors is to be able to skip it quickly when needed.
> >
> >
> >
> >
> >
> >
> > Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
> >
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On
> Behalf Of Sean
> > Hayes
> > Sent: Friday, June 08, 2007 11:49 AM
> > To: TEITAC Web/Software Subcommittee
> > Subject: [teitac-websoftware] Bypassing content.
> >
> > I have a slight issue on the testability of : "A mechanism shall be
> > available to bypass blocks of content that are repeated on multiple
> > Web pages."
> >
> > What exactly constitutes a 'block of content'?
> >
> > Also if the repeated content is the last thing on the page, is a
> > bypass still necessary? If so, where should it bypass too?
> >
> > Sean Hayes
> > Standards and Policy Team
> > Corporate Accessibility Group
> > Microsoft
> > Phone:
> > mob +44 7977 455002
> > office +44 117 9719730
> >

From: Hoffman, Allen
Date: Mon, Jun 11 2007 8:50 AM
Subject: Re: Bypassing content.

I'm not following this thread well.

Andrew:
Are you saying method to skip repetitive (or blocks) does not need to be
visible? it won't meet functional performance criteria if it isn't made
visible.

Don:
> Can we really require developers to use heading structures on all
> pages just so we can satisfy the programmatically determinable aspect
> of the site?
>
Don, the idea for programmatically determinable is there to allow for
AT to re-interpret/present the information in a more navigable fashion
in various modes. I don't know about requiring a specific methodology
as in use <h#> tagging, but I tend to agree with the concept that this
needs to be provided. This could be meta data, or other solutions.



Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Andrew
Kirkpatrick
Sent: Monday, June 11, 2007 10:33 AM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

No, but we can't make this technique not sufficient by requiring that
the solution needs to visible either?
AWK

> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> Barrett, Don
> Sent: Monday, June 11, 2007 10:31 AM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Can we really require developers to use heading structures on all
> pages just so we can satisfy the programmatically determinable aspect
> of the site?
>
>
> Don Barrett
> Section 508 Coordinator
> U.S. Department of Education
> (202)-205-8245
> = EMAIL ADDRESS REMOVED =
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> Andrew Kirkpatrick
> Sent: Monday, June 11, 2007 10:29 AM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Allen,
> Some methods to meet this requirement might not be visible - for
> example, providing a heading structure that allows users to navigate
> to different areas of the content. I don't think that we should be
> think just about "skip navigation" links.
>
> AWK
>
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> > Hoffman, Allen
> > Sent: Monday, June 11, 2007 9:33 AM
> > To: = EMAIL ADDRESS REMOVED = ; TEITAC Web/Software Subcommittee
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> > Something we have missed in our language is that the method
> must also
> > be visible. This is so often overlooked, or intentionally
> not wanted
> > that it MUST be said.
> >
> >
> >
> >
> > Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
> >
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> > Travis Roth
> > Sent: Friday, June 08, 2007 6:49 PM
> > To: = EMAIL ADDRESS REMOVED = ; 'TEITAC Web/Software Subcommittee'
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> > The skipping of groups of links, such as navigational
> links, is also
> > beneficial to other keyboard users who cannot use a mouse
> but also do
> > not use a screen reader.
> >
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> > Sailesh Panchang
> > Sent: Friday, June 08, 2007 2:02 PM
> > To: 'TEITAC Web/Software Subcommittee'
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> >
> > This requirement is primarily designed to help users of
> screen reading
>
> > software.
> > User agents including assistive technologies provide
> ability to jump
> > to different elements (eg. lists, tables, frames,) on a page if
> > structural markup is used. Structural markup for tables, headings
> > lists, etc. is required by Para (g) under
> > 6.1 that reads "Information and relationships conveyed through
> > presentation shall be programmatically determinable, and
> notification
> > of changes to these is available to...". So using structural markup
> > facilitates AT to skip over blocks / skip to desired block.
> > Therefore I believe the skip block requirement being debated over
> > predominantly applies to navigational links. This was the intent of
> > the
> > 2001
> > S508 standards too.
> > Note that when groups of links are placed in a list or
> frame, etc. a
> > mechanism to skip over them is available to screen reader users- an
> > explicit skip nav link is not needed.
> >
> > Conclusion: So I think the focus should just be "group" of
> > navigational links for the skipping over requirement.
> > Consider requirement of the word "repetitive" too.
> >
> > Sailesh Panchang
> > Senior Accessibility Engineer
> > Deque Systems Inc. (www.deque.com)
> > 11130 Sunrise Valley Drive, Suite #140, Reston VA 20191
> > Phone: 703-225-0380 (ext 105)
> > E-mail: = EMAIL ADDRESS REMOVED =
> >
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> > Hoffman, Allen
> > Sent: Friday, June 08, 2007 1:28 PM
> > To: TEITAC Web/Software Subcommittee
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> > Hmmm, thinking cap on.
> >
> > I am not convinced "logical tab order" is so difficult to
> convert to
> > specification/standard format, and am not convinced that this one
> > isn't either, yet.
> >
> > Web pages that are coded to:
> > allow correct keyboard operation;
> > associate header/cell information;
> > expose control element information;
> > generally can meet this repetitive navigation requirement, but
> could
> > still have block of content to navigate through.
> >
> > Screen readers now generally do this by either using algorithms to
> > evaluate pages for repetitive int3erface elements, or by
> allowing the
> > user to hit t to jump to next table, h for next header, or other
> > letters for other element types to jump through. To allow this
> > functionality what is important is to use consistent
> tagging of blocks
>
> > so that, for example, when you hit <next
> > whatever> it doesn't actually skip extra content, or skip in to the
> > whatever> middle
> > of the current block.
> >
> > Imagine:
> >
> > header: text.
> > header: text.
> > table: information.
> > header: text.
> > unordered list: information
> > header: text
> > table: links.
> >
> > What command would one use to jump from "block" to block without a
> > specific consistent identifier being used for the screen reader to
> > grab on to and pick up the next part of the page to read?
> >
> >
> > I'm just writing this to give us all food for thought as we try and
> > finalize this requirement. thanks.
> >
> >
> >
> >
> > Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
> >
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On
> Behalf Of Peter
> > Wallack
> > Sent: Friday, June 08, 2007 1:13 PM
> > To: TEITAC Web/Software Subcommittee
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> > Allen --
> > I like your explanation of the intent of this standard far
> better than
> > the current wording. I'm not at all keen on the phrase 'repeated on
> > multiple web pages' as that seems more directed toward a situation
> > where the user is taking a linear path through a site,
> which is often
> > not the case. How about
> > simply:
> >
> > "A mechanism shall be available to bypass blocks of content"
> >
> > leaving us only to haggle over the best wording for 'block of
> > content', which is a concept that I think we all intrinsically
> > understand but we'd like something a bit more scientific.
> >
> > As for testability, this probably is no more testable than having a
> > tab sequence that is 'logical'. This just takes it to one level
> > higher:
> > having a logical division of blocks of content that you can
> sequence
> > through.
> >
> > Peter Wallack
> > Accessibility Program Director
> > Oracle Corporation
> >
> >
> >
> > Hoffman, Allen wrote:
> > > This one is intended to allow people to basically use things like
> > > headers to jump from each block to the next quickly. I'm not
> > > convinced this language is exactly right, or better than
> > the original
> > > CFR1194.22(o) either. it is testable, but certainly is
> not easy to
> > > test in an automated fashion.
> > >
> > > I have been trying to work out testing algorithm to count
> > the number
> > > of links from start of page to each link on a page, take
> > the average
> > > link-count, and then by re-analyzing the links in variations of
> > > sequence, determine if shorter paths indeed exist, and
> where to set
> > > upper and lower boundaries on the tests to get the answer,
> > is there a
> > > way to shorten the link path. if anyone can put that into
> > a provision
> >
> > > we would have a testable check on lowering "link" or "tab"
> > path. The
> > > reason "block of content" is good is that it doesn't
> really matter
> > > what stuff one has to skip over to get past with the voice, the
> > > important factors is to be able to skip it quickly when needed.
> > >
> > >
> > >
> > >
> > >
> > >
> > > Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
> > >
> > > -----Original Message-----
> > > From: = EMAIL ADDRESS REMOVED =
> > > [mailto: = EMAIL ADDRESS REMOVED = ] On
> > Behalf Of Sean
> > > Hayes
> > > Sent: Friday, June 08, 2007 11:49 AM
> > > To: TEITAC Web/Software Subcommittee
> > > Subject: [teitac-websoftware] Bypassing content.
> > >
> > > I have a slight issue on the testability of : "A
> mechanism shall be
> > > available to bypass blocks of content that are repeated
> on multiple
> > > Web pages."
> > >
> > > What exactly constitutes a 'block of content'?
> > >
> > > Also if the repeated content is the last thing on the page, is a
> > > bypass still necessary? If so, where should it bypass too?
> > >
> > > Sean Hayes
> > > Standards and Policy Team
> > > Corporate Accessibility Group
> > > Microsoft
> > > Phone:
> > > mob +44 7977 455002
> > > office +44 117 9719730
> > >

From: David Poehlman
Date: Mon, Jun 11 2007 9:25 AM
Subject: Re: Bypassing content.

I think the method should be visible and service navigation rather than a
special case of navigation.

----- Original Message -----
From: "Andrew Kirkpatrick" < = EMAIL ADDRESS REMOVED = >
To: "TEITAC Web/Software Subcommittee" < = EMAIL ADDRESS REMOVED = >
Sent: Monday, June 11, 2007 10:29 AM
Subject: Re: [teitac-websoftware] Bypassing content.


Allen,
Some methods to meet this requirement might not be visible - for
example, providing a heading structure that allows users to navigate to
different areas of the content. I don't think that we should be think
just about "skip navigation" links.

AWK

> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Hoffman, Allen
> Sent: Monday, June 11, 2007 9:33 AM
> To: = EMAIL ADDRESS REMOVED = ; TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Something we have missed in our language is that the method
> must also be visible. This is so often overlooked, or
> intentionally not wanted that it MUST be said.
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Travis Roth
> Sent: Friday, June 08, 2007 6:49 PM
> To: = EMAIL ADDRESS REMOVED = ; 'TEITAC Web/Software Subcommittee'
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> The skipping of groups of links, such as navigational links,
> is also beneficial to other keyboard users who cannot use a
> mouse but also do not use a screen reader.
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Sailesh Panchang
> Sent: Friday, June 08, 2007 2:02 PM
> To: 'TEITAC Web/Software Subcommittee'
> Subject: Re: [teitac-websoftware] Bypassing content.
>
>
> This requirement is primarily designed to help users of
> screen reading software.
> User agents including assistive technologies provide ability
> to jump to different elements (eg. lists, tables, frames,) on
> a page if structural markup is used. Structural markup for
> tables, headings lists, etc. is required by Para (g) under
> 6.1 that reads "Information and relationships conveyed
> through presentation shall be programmatically determinable,
> and notification of changes to these is available to...". So
> using structural markup facilitates AT to skip over blocks /
> skip to desired block.
> Therefore I believe the skip block requirement being debated
> over predominantly applies to navigational links. This was
> the intent of the
> 2001
> S508 standards too.
> Note that when groups of links are placed in a list or frame,
> etc. a mechanism to skip over them is available to screen
> reader users- an explicit skip nav link is not needed.
>
> Conclusion: So I think the focus should just be "group" of
> navigational links for the skipping over requirement.
> Consider requirement of the word "repetitive" too.
>
> Sailesh Panchang
> Senior Accessibility Engineer
> Deque Systems Inc. (www.deque.com)
> 11130 Sunrise Valley Drive, Suite #140,
> Reston VA 20191
> Phone: 703-225-0380 (ext 105)
> E-mail: = EMAIL ADDRESS REMOVED =
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Hoffman, Allen
> Sent: Friday, June 08, 2007 1:28 PM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Hmmm, thinking cap on.
>
> I am not convinced "logical tab order" is so difficult to
> convert to specification/standard format, and am not
> convinced that this one isn't either, yet.
>
> Web pages that are coded to:
> allow correct keyboard operation;
> associate header/cell information;
> expose control element information;
> generally can meet this repetitive navigation
> requirement, but could still have block of content to
> navigate through.
>
> Screen readers now generally do this by either using
> algorithms to evaluate pages for repetitive int3erface
> elements, or by allowing the user to hit t to jump to next
> table, h for next header, or other letters for other element
> types to jump through. To allow this functionality what is
> important is to use consistent tagging of blocks so that, for
> example, when you hit <next
> whatever> it doesn't actually skip extra content, or skip in to the
> whatever> middle
> of the current block.
>
> Imagine:
>
> header: text.
> header: text.
> table: information.
> header: text.
> unordered list: information
> header: text
> table: links.
>
> What command would one use to jump from "block" to block
> without a specific consistent identifier being used for the
> screen reader to grab on to and pick up the next part of the
> page to read?
>
>
> I'm just writing this to give us all food for thought as we
> try and finalize this requirement. thanks.
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Peter Wallack
> Sent: Friday, June 08, 2007 1:13 PM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Allen --
> I like your explanation of the intent of this standard far
> better than the current wording. I'm not at all keen on the
> phrase 'repeated on multiple web pages' as that seems more
> directed toward a situation where the user is taking a linear
> path through a site, which is often not the case. How about
> simply:
>
> "A mechanism shall be available to bypass blocks of content"
>
> leaving us only to haggle over the best wording for 'block of
> content', which is a concept that I think we all
> intrinsically understand but we'd like something a bit more
> scientific.
>
> As for testability, this probably is no more testable than
> having a tab sequence that is 'logical'. This just takes it
> to one level higher:
> having a logical division of blocks of content that you can
> sequence through.
>
> Peter Wallack
> Accessibility Program Director
> Oracle Corporation
>
>
>
> Hoffman, Allen wrote:
> > This one is intended to allow people to basically use things like
> > headers to jump from each block to the next quickly. I'm not
> > convinced this language is exactly right, or better than
> the original
> > CFR1194.22(o) either. it is testable, but certainly is not easy to
> > test in an automated fashion.
> >
> > I have been trying to work out testing algorithm to count
> the number
> > of links from start of page to each link on a page, take
> the average
> > link-count, and then by re-analyzing the links in variations of
> > sequence, determine if shorter paths indeed exist, and where to set
> > upper and lower boundaries on the tests to get the answer,
> is there a
> > way to shorten the link path. if anyone can put that into
> a provision
>
> > we would have a testable check on lowering "link" or "tab"
> path. The
> > reason "block of content" is good is that it doesn't really matter
> > what stuff one has to skip over to get past with the voice, the
> > important factors is to be able to skip it quickly when needed.
> >
> >
> >
> >
> >
> >
> > Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
> >
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On
> Behalf Of Sean
> > Hayes
> > Sent: Friday, June 08, 2007 11:49 AM
> > To: TEITAC Web/Software Subcommittee
> > Subject: [teitac-websoftware] Bypassing content.
> >
> > I have a slight issue on the testability of : "A mechanism shall be
> > available to bypass blocks of content that are repeated on multiple
> > Web pages."
> >
> > What exactly constitutes a 'block of content'?
> >
> > Also if the repeated content is the last thing on the page, is a
> > bypass still necessary? If so, where should it bypass too?
> >
> > Sean Hayes
> > Standards and Policy Team
> > Corporate Accessibility Group
> > Microsoft
> > Phone:
> > mob +44 7977 455002
> > office +44 117 9719730
> >

From: David Poehlman
Date: Mon, Jun 11 2007 9:30 AM
Subject: Re: Bypassing content.

Well, we required a skip nav link. I would suggest we not require mark up
but that we require the functionality and supply techniques if warranted.

----- Original Message -----
From: "Barrett, Don" < = EMAIL ADDRESS REMOVED = >
To: "TEITAC Web/Software Subcommittee" < = EMAIL ADDRESS REMOVED = >
Sent: Monday, June 11, 2007 10:31 AM
Subject: Re: [teitac-websoftware] Bypassing content.


Can we really require developers to use heading structures on all pages
just so we can satisfy the programmatically determinable aspect of the
site?


Don Barrett
Section 508 Coordinator
U.S. Department of Education
(202)-205-8245
= EMAIL ADDRESS REMOVED =

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Andrew
Kirkpatrick
Sent: Monday, June 11, 2007 10:29 AM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

Allen,
Some methods to meet this requirement might not be visible - for
example, providing a heading structure that allows users to navigate to
different areas of the content. I don't think that we should be think
just about "skip navigation" links.

AWK

> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> Hoffman, Allen
> Sent: Monday, June 11, 2007 9:33 AM
> To: = EMAIL ADDRESS REMOVED = ; TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Something we have missed in our language is that the method must also
> be visible. This is so often overlooked, or intentionally not wanted
> that it MUST be said.
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> Travis Roth
> Sent: Friday, June 08, 2007 6:49 PM
> To: = EMAIL ADDRESS REMOVED = ; 'TEITAC Web/Software Subcommittee'
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> The skipping of groups of links, such as navigational links, is also
> beneficial to other keyboard users who cannot use a mouse but also do
> not use a screen reader.
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> Sailesh Panchang
> Sent: Friday, June 08, 2007 2:02 PM
> To: 'TEITAC Web/Software Subcommittee'
> Subject: Re: [teitac-websoftware] Bypassing content.
>
>
> This requirement is primarily designed to help users of screen reading

> software.
> User agents including assistive technologies provide ability to jump
> to different elements (eg. lists, tables, frames,) on a page if
> structural markup is used. Structural markup for tables, headings
> lists, etc. is required by Para (g) under
> 6.1 that reads "Information and relationships conveyed through
> presentation shall be programmatically determinable, and notification
> of changes to these is available to...". So using structural markup
> facilitates AT to skip over blocks / skip to desired block.
> Therefore I believe the skip block requirement being debated over
> predominantly applies to navigational links. This was the intent of
> the
> 2001
> S508 standards too.
> Note that when groups of links are placed in a list or frame, etc. a
> mechanism to skip over them is available to screen reader users- an
> explicit skip nav link is not needed.
>
> Conclusion: So I think the focus should just be "group" of
> navigational links for the skipping over requirement.
> Consider requirement of the word "repetitive" too.
>
> Sailesh Panchang
> Senior Accessibility Engineer
> Deque Systems Inc. (www.deque.com)
> 11130 Sunrise Valley Drive, Suite #140, Reston VA 20191
> Phone: 703-225-0380 (ext 105)
> E-mail: = EMAIL ADDRESS REMOVED =
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> Hoffman, Allen
> Sent: Friday, June 08, 2007 1:28 PM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Hmmm, thinking cap on.
>
> I am not convinced "logical tab order" is so difficult to convert to
> specification/standard format, and am not convinced that this one
> isn't either, yet.
>
> Web pages that are coded to:
> allow correct keyboard operation;
> associate header/cell information;
> expose control element information;
> generally can meet this repetitive navigation requirement, but
could
> still have block of content to navigate through.
>
> Screen readers now generally do this by either using algorithms to
> evaluate pages for repetitive int3erface elements, or by allowing the
> user to hit t to jump to next table, h for next header, or other
> letters for other element types to jump through. To allow this
> functionality what is important is to use consistent tagging of blocks

> so that, for example, when you hit <next
> whatever> it doesn't actually skip extra content, or skip in to the
> whatever> middle
> of the current block.
>
> Imagine:
>
> header: text.
> header: text.
> table: information.
> header: text.
> unordered list: information
> header: text
> table: links.
>
> What command would one use to jump from "block" to block without a
> specific consistent identifier being used for the screen reader to
> grab on to and pick up the next part of the page to read?
>
>
> I'm just writing this to give us all food for thought as we
> try and finalize this requirement. thanks.
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Peter Wallack
> Sent: Friday, June 08, 2007 1:13 PM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Allen --
> I like your explanation of the intent of this standard far
> better than the current wording. I'm not at all keen on the
> phrase 'repeated on multiple web pages' as that seems more
> directed toward a situation where the user is taking a linear
> path through a site, which is often not the case. How about
> simply:
>
> "A mechanism shall be available to bypass blocks of content"
>
> leaving us only to haggle over the best wording for 'block of
> content', which is a concept that I think we all
> intrinsically understand but we'd like something a bit more
> scientific.
>
> As for testability, this probably is no more testable than
> having a tab sequence that is 'logical'. This just takes it
> to one level higher:
> having a logical division of blocks of content that you can
> sequence through.
>
> Peter Wallack
> Accessibility Program Director
> Oracle Corporation
>
>
>
> Hoffman, Allen wrote:
> > This one is intended to allow people to basically use things like
> > headers to jump from each block to the next quickly. I'm not
> > convinced this language is exactly right, or better than
> the original
> > CFR1194.22(o) either. it is testable, but certainly is not easy to
> > test in an automated fashion.
> >
> > I have been trying to work out testing algorithm to count
> the number
> > of links from start of page to each link on a page, take
> the average
> > link-count, and then by re-analyzing the links in variations of
> > sequence, determine if shorter paths indeed exist, and where to set
> > upper and lower boundaries on the tests to get the answer,
> is there a
> > way to shorten the link path. if anyone can put that into
> a provision
>
> > we would have a testable check on lowering "link" or "tab"
> path. The
> > reason "block of content" is good is that it doesn't really matter
> > what stuff one has to skip over to get past with the voice, the
> > important factors is to be able to skip it quickly when needed.
> >
> >
> >
> >
> >
> >
> > Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
> >
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On
> Behalf Of Sean
> > Hayes
> > Sent: Friday, June 08, 2007 11:49 AM
> > To: TEITAC Web/Software Subcommittee
> > Subject: [teitac-websoftware] Bypassing content.
> >
> > I have a slight issue on the testability of : "A mechanism shall be
> > available to bypass blocks of content that are repeated on multiple
> > Web pages."
> >
> > What exactly constitutes a 'block of content'?
> >
> > Also if the repeated content is the last thing on the page, is a
> > bypass still necessary? If so, where should it bypass too?
> >
> > Sean Hayes
> > Standards and Policy Team
> > Corporate Accessibility Group
> > Microsoft
> > Phone:
> > mob +44 7977 455002
> > office +44 117 9719730
> >

From: David Poehlman
Date: Mon, Jun 11 2007 9:55 AM
Subject: Re: Bypassing content.

and is it still repetative if a different set of graphics are used?

----- Original Message -----
From: "Andrew Kirkpatrick" < = EMAIL ADDRESS REMOVED = >
To: "TEITAC Web/Software Subcommittee" < = EMAIL ADDRESS REMOVED = >
Sent: Monday, June 11, 2007 9:12 AM
Subject: Re: [teitac-websoftware] Bypassing content.


I'm willing to bet that most "repetitive content" is not completely
repetitive. We should be careful about how we define this - for
example, if the block of content is a navigation bar and when you click
on a link in that navigation bar you get a page that has almost the same
navigation bar, except that there is some small difference, such as a
different set of subnavigation links that are in view or something
similar. Is that still repetitive?

AWK

From: Andrew Kirkpatrick
Date: Mon, Jun 11 2007 10:00 AM
Subject: Re: Bypassing content.

This is what I'm asking - if we say "repetitive", it implies that the
content is the same. If the content isn't the same, even in some
trivial way, some people will say that it isn't repetitive.

AWK

> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of David Poehlman
> Sent: Monday, June 11, 2007 11:54 AM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> and is it still repetative if a different set of graphics are used?
>
> ----- Original Message -----
> From: "Andrew Kirkpatrick" < = EMAIL ADDRESS REMOVED = >
> To: "TEITAC Web/Software Subcommittee"
> < = EMAIL ADDRESS REMOVED = >
> Sent: Monday, June 11, 2007 9:12 AM
> Subject: Re: [teitac-websoftware] Bypassing content.
>
>
> I'm willing to bet that most "repetitive content" is not completely
> repetitive. We should be careful about how we define this - for
> example, if the block of content is a navigation bar and when
> you click
> on a link in that navigation bar you get a page that has
> almost the same
> navigation bar, except that there is some small difference, such as a
> different set of subnavigation links that are in view or something
> similar. Is that still repetitive?
>
> AWK
>
>
>

From: Andrew Kirkpatrick
Date: Mon, Jun 11 2007 10:05 AM
Subject: Re: Bypassing content.

> Andrew:
> Are you saying method to skip repetitive (or blocks) does not
> need to be visible? it won't meet functional performance
> criteria if it isn't made visible.

An invisible skip link wouldn't satisfy the Functional perfomance
criteria, but neither would any invisible UI control. However,
functionality that allows users to move to different locations using the
semantic structure in a document or application is neither visible or
invisible, so I don't think that the requirement this this be visible is
necessary.
AWK

From: Brad Hodges
Date: Mon, Jun 11 2007 10:30 AM
Subject: Re: Bypassing content.

Greetings:

While invisible strategies may be of use in some instances, strategies
which use the native structure to facilitate nonvisual navigation in the
same way these same elements support visual navigation are clearly
preferable.

At the same time it should be noted that skip nav links often do not
work properly.

Brad


-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of David
Poehlman
Sent: Monday, June 11, 2007 11:23 AM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

I think the method should be visible and service navigation rather than
a special case of navigation.

----- Original Message -----
From: "Andrew Kirkpatrick" < = EMAIL ADDRESS REMOVED = >
To: "TEITAC Web/Software Subcommittee"
< = EMAIL ADDRESS REMOVED = >
Sent: Monday, June 11, 2007 10:29 AM
Subject: Re: [teitac-websoftware] Bypassing content.


Allen,
Some methods to meet this requirement might not be visible - for
example, providing a heading structure that allows users to navigate to
different areas of the content. I don't think that we should be think
just about "skip navigation" links.

AWK

> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Hoffman, Allen
> Sent: Monday, June 11, 2007 9:33 AM
> To: = EMAIL ADDRESS REMOVED = ; TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Something we have missed in our language is that the method
> must also be visible. This is so often overlooked, or
> intentionally not wanted that it MUST be said.
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Travis Roth
> Sent: Friday, June 08, 2007 6:49 PM
> To: = EMAIL ADDRESS REMOVED = ; 'TEITAC Web/Software Subcommittee'
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> The skipping of groups of links, such as navigational links,
> is also beneficial to other keyboard users who cannot use a
> mouse but also do not use a screen reader.
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Sailesh Panchang
> Sent: Friday, June 08, 2007 2:02 PM
> To: 'TEITAC Web/Software Subcommittee'
> Subject: Re: [teitac-websoftware] Bypassing content.
>
>
> This requirement is primarily designed to help users of
> screen reading software.
> User agents including assistive technologies provide ability
> to jump to different elements (eg. lists, tables, frames,) on
> a page if structural markup is used. Structural markup for
> tables, headings lists, etc. is required by Para (g) under
> 6.1 that reads "Information and relationships conveyed
> through presentation shall be programmatically determinable,
> and notification of changes to these is available to...". So
> using structural markup facilitates AT to skip over blocks /
> skip to desired block.
> Therefore I believe the skip block requirement being debated
> over predominantly applies to navigational links. This was
> the intent of the
> 2001
> S508 standards too.
> Note that when groups of links are placed in a list or frame,
> etc. a mechanism to skip over them is available to screen
> reader users- an explicit skip nav link is not needed.
>
> Conclusion: So I think the focus should just be "group" of
> navigational links for the skipping over requirement.
> Consider requirement of the word "repetitive" too.
>
> Sailesh Panchang
> Senior Accessibility Engineer
> Deque Systems Inc. (www.deque.com)
> 11130 Sunrise Valley Drive, Suite #140,
> Reston VA 20191
> Phone: 703-225-0380 (ext 105)
> E-mail: = EMAIL ADDRESS REMOVED =
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Hoffman, Allen
> Sent: Friday, June 08, 2007 1:28 PM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Hmmm, thinking cap on.
>
> I am not convinced "logical tab order" is so difficult to
> convert to specification/standard format, and am not
> convinced that this one isn't either, yet.
>
> Web pages that are coded to:
> allow correct keyboard operation;
> associate header/cell information;
> expose control element information;
> generally can meet this repetitive navigation
> requirement, but could still have block of content to
> navigate through.
>
> Screen readers now generally do this by either using
> algorithms to evaluate pages for repetitive int3erface
> elements, or by allowing the user to hit t to jump to next
> table, h for next header, or other letters for other element
> types to jump through. To allow this functionality what is
> important is to use consistent tagging of blocks so that, for
> example, when you hit <next
> whatever> it doesn't actually skip extra content, or skip in to the
> whatever> middle
> of the current block.
>
> Imagine:
>
> header: text.
> header: text.
> table: information.
> header: text.
> unordered list: information
> header: text
> table: links.
>
> What command would one use to jump from "block" to block
> without a specific consistent identifier being used for the
> screen reader to grab on to and pick up the next part of the
> page to read?
>
>
> I'm just writing this to give us all food for thought as we
> try and finalize this requirement. thanks.
>
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Peter Wallack
> Sent: Friday, June 08, 2007 1:13 PM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Allen --
> I like your explanation of the intent of this standard far
> better than the current wording. I'm not at all keen on the
> phrase 'repeated on multiple web pages' as that seems more
> directed toward a situation where the user is taking a linear
> path through a site, which is often not the case. How about
> simply:
>
> "A mechanism shall be available to bypass blocks of content"
>
> leaving us only to haggle over the best wording for 'block of
> content', which is a concept that I think we all
> intrinsically understand but we'd like something a bit more
> scientific.
>
> As for testability, this probably is no more testable than
> having a tab sequence that is 'logical'. This just takes it
> to one level higher:
> having a logical division of blocks of content that you can
> sequence through.
>
> Peter Wallack
> Accessibility Program Director
> Oracle Corporation
>
>
>
> Hoffman, Allen wrote:
> > This one is intended to allow people to basically use things like
> > headers to jump from each block to the next quickly. I'm not
> > convinced this language is exactly right, or better than
> the original
> > CFR1194.22(o) either. it is testable, but certainly is not easy to
> > test in an automated fashion.
> >
> > I have been trying to work out testing algorithm to count
> the number
> > of links from start of page to each link on a page, take
> the average
> > link-count, and then by re-analyzing the links in variations of
> > sequence, determine if shorter paths indeed exist, and where to set
> > upper and lower boundaries on the tests to get the answer,
> is there a
> > way to shorten the link path. if anyone can put that into
> a provision
>
> > we would have a testable check on lowering "link" or "tab"
> path. The
> > reason "block of content" is good is that it doesn't really matter
> > what stuff one has to skip over to get past with the voice, the
> > important factors is to be able to skip it quickly when needed.
> >
> >
> >
> >
> >
> >
> > Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
> >
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On
> Behalf Of Sean
> > Hayes
> > Sent: Friday, June 08, 2007 11:49 AM
> > To: TEITAC Web/Software Subcommittee
> > Subject: [teitac-websoftware] Bypassing content.
> >
> > I have a slight issue on the testability of : "A mechanism shall be
> > available to bypass blocks of content that are repeated on multiple
> > Web pages."
> >
> > What exactly constitutes a 'block of content'?
> >
> > Also if the repeated content is the last thing on the page, is a
> > bypass still necessary? If so, where should it bypass too?
> >
> > Sean Hayes
> > Standards and Policy Team
> > Corporate Accessibility Group
> > Microsoft
> > Phone:
> > mob +44 7977 455002
> > office +44 117 9719730
> >

From: Sailesh Panchang
Date: Mon, Jun 11 2007 10:35 AM
Subject: Re: Bypassing content.

It is because of these complexities in determining repetitiveness, I
suggested we need to consider the use of the word "repetitive".
When links (- repetitive or not) are not within a list or grouped under a
header label etc. it Is helpful if there is a skip nav link. If markup
surrounding the links allow user agents including AT to provide the ability
to skip, then a skip nav link may not be necessary.
A group may be defined as a set of at least 4 links.

Sailesh Panchang
Senior Accessibility Engineer
Deque Systems Inc. (www.deque.com)
11130 Sunrise Valley Drive, Suite #140,
Reston VA 20191
Phone: 703-225-0380 (ext 105)
E-mail: = EMAIL ADDRESS REMOVED =
-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Andrew
Kirkpatrick
Sent: Monday, June 11, 2007 11:56 AM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

This is what I'm asking - if we say "repetitive", it implies that the
content is the same. If the content isn't the same, even in some
trivial way, some people will say that it isn't repetitive.

AWK

> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of David Poehlman
> Sent: Monday, June 11, 2007 11:54 AM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> and is it still repetative if a different set of graphics are used?
>
> ----- Original Message -----
> From: "Andrew Kirkpatrick" < = EMAIL ADDRESS REMOVED = >
> To: "TEITAC Web/Software Subcommittee"
> < = EMAIL ADDRESS REMOVED = >
> Sent: Monday, June 11, 2007 9:12 AM
> Subject: Re: [teitac-websoftware] Bypassing content.
>
>
> I'm willing to bet that most "repetitive content" is not completely
> repetitive. We should be careful about how we define this - for
> example, if the block of content is a navigation bar and when
> you click
> on a link in that navigation bar you get a page that has
> almost the same
> navigation bar, except that there is some small difference, such as a
> different set of subnavigation links that are in view or something
> similar. Is that still repetitive?
>
> AWK
>
>
>

From: Hoffman, Allen
Date: Mon, Jun 11 2007 11:15 AM
Subject: Re: Bypassing content.

OK I'm following the logic.
This does highlight that these skip-nav and consistently using structure
to navigate may be indeed separable concepts.



Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Andrew
Kirkpatrick
Sent: Monday, June 11, 2007 11:50 AM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

> Andrew:
> Are you saying method to skip repetitive (or blocks) does not need to
> be visible? it won't meet functional performance criteria if it isn't

> made visible.

An invisible skip link wouldn't satisfy the Functional perfomance
criteria, but neither would any invisible UI control. However,
functionality that allows users to move to different locations using the
semantic structure in a document or application is neither visible or
invisible, so I don't think that the requirement this this be visible is
necessary.
AWK

From: Peter Wallack
Date: Mon, Jun 11 2007 11:35 AM
Subject: Re: Bypassing content.

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
But what structure are we going to recommend? If I markup a page with
&lt;H#&gt; tags, then screen readers do a beautiful job of letting me
jump to 'blocks of content' (i.e. press 'h' in JAWS to get a list of
all the headers in the page). But what about the keyboard-only user who
isn't running AT? Are we really going to require that every web page
code something in addition to the &lt;H#&gt; tags simply because the
(mainstream) browsers themselves don't provide the same navigation
mechanisms as the screen readers? And how would we make these visible -
yet another navigation bar on the page? (that itself would need to be
skippable).<br>
<br>
For fun, I just tried using section508.gov with the keyboard only.
Really awful. (I really like the popups that say you are leaving the
site, like on 'FAQs', even though all you did was navigate onto the
link.) It might help this discussion, particularly in defining 'block
of content' if someone could:<br>
<ul>
<li>tell us a site that they think has done it well</li>
<li>pick a site and describe how they think it should work</li>
</ul>
<pre class="moz-signature" cols="72">Peter Wallack
Accessibility Program Director
Oracle Corporation</pre>
<br>
<br>
Hoffman, Allen wrote:
<blockquote
cite="mid: = EMAIL ADDRESS REMOVED = "
type="cite">
<pre wrap="">OK I'm following the logic.
This does highlight that these skip-nav and consistently using structure
to navigate may be indeed separable concepts.



Allen Hoffman -- <a class="moz-txt-link-abbreviated" href="mailto: = EMAIL ADDRESS REMOVED = "> = EMAIL ADDRESS REMOVED = </a>; v: 202-447-0303

-----Original Message-----
From: <a class="moz-txt-link-abbreviated" href="mailto: = EMAIL ADDRESS REMOVED = "> = EMAIL ADDRESS REMOVED = </a>
[<a class="moz-txt-link-freetext" href="mailto: = EMAIL ADDRESS REMOVED = ">mailto: = EMAIL ADDRESS REMOVED = </a>] On Behalf Of Andrew
Kirkpatrick
Sent: Monday, June 11, 2007 11:50 AM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

</pre>
<blockquote type="cite">
<pre wrap="">Andrew:
Are you saying method to skip repetitive (or blocks) does not need to
be visible? it won't meet functional performance criteria if it isn't
</pre>
</blockquote>
<pre wrap=""><!---->
</pre>
<blockquote type="cite">
<pre wrap="">made visible.
</pre>
</blockquote>
<pre wrap=""><!---->
An invisible skip link wouldn't satisfy the Functional perfomance
criteria, but neither would any invisible UI control. However,
functionality that allows users to move to different locations using the
semantic structure in a document or application is neither visible or
invisible, so I don't think that the requirement this this be visible is
necessary.
AWK

From: David Poehlman
Date: Mon, Jun 11 2007 11:40 AM
Subject: Re: Bypassing content.

right, the user agent guidelines require browsers to provide a nav mechanism
via the keyboard.

----- Original Message -----
From: "Peter Wallack" < = EMAIL ADDRESS REMOVED = >
To: "TEITAC Web/Software Subcommittee" < = EMAIL ADDRESS REMOVED = >
Sent: Monday, June 11, 2007 1:28 PM
Subject: Re: [teitac-websoftware] Bypassing content.


But what structure are we going to recommend? If I markup a page with <H#>
tags, then screen readers do a beautiful job of letting me jump to 'blocks
of content' (i.e. press 'h' in JAWS to get a list of all the headers in the
page). But what about the keyboard-only user who isn't running AT? Are we
really going to require that every web page code something in addition to
the <H#> tags simply because the (mainstream) browsers themselves don't
provide the same navigation mechanisms as the screen readers? And how would
we make these visible - yet another navigation bar on the page? (that itself
would need to be skippable).

For fun, I just tried using section508.gov with the keyboard only. Really
awful. (I really like the popups that say you are leaving the site, like on
'FAQs', even though all you did was navigate onto the link.) It might help
this discussion, particularly in defining 'block of content' if someone
could:

a.. tell us a site that they think has done it well
b.. pick a site and describe how they think it should work
Peter Wallack
Accessibility Program Director
Oracle Corporation

Hoffman, Allen wrote:
OK I'm following the logic.
This does highlight that these skip-nav and consistently using structure
to navigate may be indeed separable concepts.



Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Andrew
Kirkpatrick
Sent: Monday, June 11, 2007 11:50 AM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

Andrew:
Are you saying method to skip repetitive (or blocks) does not need to
be visible? it won't meet functional performance criteria if it isn't

made visible.

An invisible skip link wouldn't satisfy the Functional perfomance
criteria, but neither would any invisible UI control. However,
functionality that allows users to move to different locations using the
semantic structure in a document or application is neither visible or
invisible, so I don't think that the requirement this this be visible is
necessary.
AWK

From: David Poehlman
Date: Mon, Jun 11 2007 11:45 AM
Subject: Re: Bypassing content.

this is why I am in favor of justThere are lots of ways to separate content
and to provide for movement/navigation to different parts of a document.
requiring structural mark up and having the user agents help out here.
----- Original Message -----
From: "Andrew Kirkpatrick" < = EMAIL ADDRESS REMOVED = >
To: "TEITAC Web/Software Subcommittee" < = EMAIL ADDRESS REMOVED = >
Sent: Monday, June 11, 2007 11:56 AM
Subject: Re: [teitac-websoftware] Bypassing content.


This is what I'm asking - if we say "repetitive", it implies that the
content is the same. If the content isn't the same, even in some
trivial way, some people will say that it isn't repetitive.

AWK

> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of David Poehlman
> Sent: Monday, June 11, 2007 11:54 AM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> and is it still repetative if a different set of graphics are used?
>
> ----- Original Message -----
> From: "Andrew Kirkpatrick" < = EMAIL ADDRESS REMOVED = >
> To: "TEITAC Web/Software Subcommittee"
> < = EMAIL ADDRESS REMOVED = >
> Sent: Monday, June 11, 2007 9:12 AM
> Subject: Re: [teitac-websoftware] Bypassing content.
>
>
> I'm willing to bet that most "repetitive content" is not completely
> repetitive. We should be careful about how we define this - for
> example, if the block of content is a navigation bar and when
> you click
> on a link in that navigation bar you get a page that has
> almost the same
> navigation bar, except that there is some small difference, such as a
> different set of subnavigation links that are in view or something
> similar. Is that still repetitive?
>
> AWK
>
>
>

From: Langum, Michael J
Date: Tue, Jun 12 2007 6:25 AM
Subject: Re: Bypassing content.

AWK wrote:
>"An invisible skip link wouldn't satisfy the Functional perfomance
criteria,. . ."

Have I missed something?

I thought the "function" of the "skip link" was to allow users of AT to
skip past banners, and standard navigation elements so that they could
quickly and easily get to the main content of a page. Since sighted
users can achieve the same "function" by simply examining the page and
choosing to begin reading at whatever point they wish (and hence don't
rely on the "skip link"), I don't see how an "invisible skip link
wouldn't satisfy the Functional performance criteria."

-- Mike Langum
Asst. Webmaster, WWW.OPM.GOV
U.S. Office of Personnel Management

From: David Poehlman
Date: Tue, Jun 12 2007 6:30 AM
Subject: Re: Bypassing content.

individuals with low vision who use the keyboard will need it. there are
also other circumstances where it would be handy.

----- Original Message -----
From: "Langum, Michael J" < = EMAIL ADDRESS REMOVED = >
To: < = EMAIL ADDRESS REMOVED = >
Sent: Tuesday, June 12, 2007 8:21 AM
Subject: Re: [teitac-websoftware] Bypassing content.



AWK wrote:
>"An invisible skip link wouldn't satisfy the Functional perfomance
criteria,. . ."

Have I missed something?

I thought the "function" of the "skip link" was to allow users of AT to
skip past banners, and standard navigation elements so that they could
quickly and easily get to the main content of a page. Since sighted
users can achieve the same "function" by simply examining the page and
choosing to begin reading at whatever point they wish (and hence don't
rely on the "skip link"), I don't see how an "invisible skip link
wouldn't satisfy the Functional performance criteria."

-- Mike Langum
Asst. Webmaster, WWW.OPM.GOV
U.S. Office of Personnel Management

From: Hoffman, Allen
Date: Tue, Jun 12 2007 6:45 AM
Subject: Re: Bypassing content.

People with low vision are restricted to small portions of the screen,
so focus tracking is critical--visually scanning the page via a tiny
window, imagine a word at a time, slows one down enormously, while
having a link to skip much of that will greatly increase someone in that
position's ability to use the information.

I think this was originally designed for screen reader users, but once
it is examined carefully, functional performance criteria would apply,
and it isn't just for totally blind folks really.




Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Langum,
Michael J
Sent: Tuesday, June 12, 2007 8:22 AM
To: = EMAIL ADDRESS REMOVED =
Subject: Re: [teitac-websoftware] Bypassing content.


AWK wrote:
>"An invisible skip link wouldn't satisfy the Functional perfomance
criteria,. . ."

Have I missed something?

I thought the "function" of the "skip link" was to allow users of AT to
skip past banners, and standard navigation elements so that they could
quickly and easily get to the main content of a page. Since sighted
users can achieve the same "function" by simply examining the page and
choosing to begin reading at whatever point they wish (and hence don't
rely on the "skip link"), I don't see how an "invisible skip link
wouldn't satisfy the Functional performance criteria."

-- Mike Langum
Asst. Webmaster, WWW.OPM.GOV
U.S. Office of Personnel Management

From: Katie Haritos-Shea
Date: Tue, Jun 12 2007 6:55 AM
Subject: Re: Bypassing content.

Mike,

The visible skip nav is also useful for persons with limited mobility, and the AT that they use. In that case the skip nav link must be visible.

Katie

-----Original Message-----
>From: "Langum, Michael J" < = EMAIL ADDRESS REMOVED = >
>Sent: Jun 12, 2007 8:21 AM
>To: = EMAIL ADDRESS REMOVED =
>Subject: Re: [teitac-websoftware] Bypassing content.
>
>
>AWK wrote:
>>"An invisible skip link wouldn't satisfy the Functional perfomance
>criteria,. . ."
>
>Have I missed something?
>
>I thought the "function" of the "skip link" was to allow users of AT to
>skip past banners, and standard navigation elements so that they could
>quickly and easily get to the main content of a page. Since sighted
>users can achieve the same "function" by simply examining the page and
>choosing to begin reading at whatever point they wish (and hence don't
>rely on the "skip link"), I don't see how an "invisible skip link
>wouldn't satisfy the Functional performance criteria."
>
>-- Mike Langum
> Asst. Webmaster, WWW.OPM.GOV
> U.S. Office of Personnel Management
>
>
>

From: Peter Wallack
Date: Tue, Jun 12 2007 9:55 AM
Subject: Re: Bypassing content.

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
I'd argue that it's useful far beyond that. Any keyboard-only user,
such as a 'power user', would appreciate the feature. I believe it
needs to be visible until User Agents provide a standard mechanism to
invoke it.<br>
<pre class="moz-signature" cols="72">Peter Wallack
Accessibility Program Director
Oracle Corporation</pre>
<br>
<br>
Katie Haritos-Shea wrote:
<blockquote
cite="mid: = EMAIL ADDRESS REMOVED = "
type="cite">
<pre wrap="">Mike,

The visible skip nav is also useful for persons with limited mobility, and the AT that they use. In that case the skip nav link must be visible.

Katie

-----Original Message-----
</pre>
<blockquote type="cite">
<pre wrap="">From: "Langum, Michael J" <a class="moz-txt-link-rfc2396E" href="mailto: = EMAIL ADDRESS REMOVED = ">&lt; = EMAIL ADDRESS REMOVED = &gt;</a>
Sent: Jun 12, 2007 8:21 AM
To: <a class="moz-txt-link-abbreviated" href="mailto: = EMAIL ADDRESS REMOVED = "> = EMAIL ADDRESS REMOVED = </a>
Subject: Re: [teitac-websoftware] Bypassing content.


AWK wrote:
</pre>
<blockquote type="cite">
<pre wrap="">"An invisible skip link wouldn't satisfy the Functional perfomance
</pre>
</blockquote>
<pre wrap="">criteria,. . ."

Have I missed something?

I thought the "function" of the "skip link" was to allow users of AT to
skip past banners, and standard navigation elements so that they could
quickly and easily get to the main content of a page. Since sighted
users can achieve the same "function" by simply examining the page and
choosing to begin reading at whatever point they wish (and hence don't
rely on the "skip link"), I don't see how an "invisible skip link
wouldn't satisfy the Functional performance criteria."

-- Mike Langum
Asst. Webmaster, <a class="moz-txt-link-abbreviated" href="http://WWW.OPM.GOV">WWW.OPM.GOV</a>
U.S. Office of Personnel Management

From: Andrew Kirkpatrick
Date: Tue, Jun 12 2007 2:15 PM
Subject: Re: Bypassing content.

The skip nav link offers very poor control - you get one option and that
is to move focus to where the link is and nowhere else.
AWK

> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Hoffman, Allen
> Sent: Tuesday, June 12, 2007 4:08 PM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> I concur with Andrew, but.
>
> The skip-nav requirement is intended to allow the end-user
> more control of directing the focus. if an application
> allows focus to be moved without inefficient keyboard or
> other driving, then it meets the standard already.
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf
> Of Andrew Kirkpatrick
> Sent: Tuesday, June 12, 2007 3:48 PM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Exactly. The concept of the skip link is a possible success
> technique, but we should not skew the requirement to
> eliminate other success techniques that people are able to
> use presently.
> AWK
>
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On
> Behalf Of David
>
> > Poehlman
> > Sent: Tuesday, June 12, 2007 3:45 PM
> > To: TEITAC Web/Software Subcommittee; Katie Haritos-Shea
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> > I just know there is already methodology for dealing with
> this without
>
> > using a special link though.
> >
> > ----- Original Message -----
> > From: "Peter Wallack" < = EMAIL ADDRESS REMOVED = >
> > To: "Katie Haritos-Shea" < = EMAIL ADDRESS REMOVED = >; "TEITAC
> Web/Software
>
> > Subcommittee" < = EMAIL ADDRESS REMOVED = >
> > Sent: Tuesday, June 12, 2007 11:52 AM
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> >
> > I'd argue that it's useful far beyond that. Any keyboard-only user,
> > such as a 'power user', would appreciate the feature. I believe it
> > needs to be visible until User Agents provide a standard
> mechanism to
> > invoke it.
> >
> > Peter Wallack
> > Accessibility Program Director
> > Oracle Corporation
> >
> > Katie Haritos-Shea wrote:
> > Mike,
> >
> > The visible skip nav is also useful for persons with
> limited mobility,
>
> > and the AT that they use. In that case the skip nav link must be
> > visible.
> >
> > Katie
> >
> > -----Original Message-----
> > From: "Langum, Michael J" < = EMAIL ADDRESS REMOVED = >
> > Sent: Jun 12, 2007 8:21 AM
> > To: = EMAIL ADDRESS REMOVED =
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> >
> > AWK wrote:
> > "An invisible skip link wouldn't satisfy the Functional
> perfomance
> > criteria,. . ."
> >
> > Have I missed something?
> >
> > I thought the "function" of the "skip link" was to allow
> users of AT
> > to skip past banners, and standard navigation elements so that they
> > could quickly and easily get to the main content of a page. Since
> > sighted users can achieve the same "function" by simply
> examining the
> > page and choosing to begin reading at whatever point they wish (and
> > hence don't rely on the "skip link"), I don't see how an "invisible
> > skip link wouldn't satisfy the Functional performance criteria."
> >
> > -- Mike Langum
> > Asst. Webmaster, WWW.OPM.GOV
> > U.S. Office of Personnel Management
> >
> >
> >

From: Barrett, Don
Date: Tue, Jun 12 2007 2:30 PM
Subject: Re: Bypassing content.

Hmmmm, for internal apps, we'd probably accept lists as we do headings.
We initially tried to only accept browser rather than screen reader
methods so in that sense, headings and lists wouldn't work, but since
major screen readers all support lists and headings, I imagine we would
be a little more forgiving. How's that for non-committal?


Don Barrett
Section 508 Coordinator
U.S. Department of Education
(202)-205-8245
= EMAIL ADDRESS REMOVED =

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Sailesh
Panchang
Sent: Tuesday, June 12, 2007 4:29 PM
To: 'TEITAC Web/Software Subcommittee'
Subject: Re: [teitac-websoftware] Bypassing content.

Don,
>We have accepted frames, headings, and skip nav links as methods of
>meeting the standard. All the standard has required is a method to
>skip repetitive nav links; it's been very generic and very helpful.
Do lists not make the cut in your opinion?

Sailesh Panchang
Senior Accessibility Engineer
Deque Systems Inc. (www.deque.com)
11130 Sunrise Valley Drive, Suite #140,
Reston VA 20191
Phone: 703-225-0380 (ext 105)
E-mail: = EMAIL ADDRESS REMOVED =

From: Sailesh Panchang
Date: Tue, Jun 12 2007 2:35 PM
Subject: Re: Bypassing content.

>The skip nav link offers very poor control - you get one option and that
>is to move focus to where the link is and nowhere else.
Well that is what all links do- move focus to a particular target. The skip
nav link moves focus to after end of group of links or repetitive block of
content if one prefers that. It never purported to do anything else and is
an effective vehicle for that. It is but one option when other structural
markup is absent.

Sailesh Panchang
Senior Accessibility Engineer
Deque Systems Inc. (www.deque.com)
11130 Sunrise Valley Drive, Suite #140,
Reston VA 20191
Phone: 703-225-0380 (ext 105)
E-mail: = EMAIL ADDRESS REMOVED =

From: Hoffman, Allen
Date: Tue, Jun 12 2007 2:40 PM
Subject: Re: Bypassing content.

The problem with skip-nav as is is that it is difficult to test in an
automated fashion as unless a consistent text is used, automated tools
have a hard time evaluating the tab order and functionality of the page
to shorten it effectively.

So:

how about:

Web pages shall provide a visible method that is programmatically
determinable to skip repetitive navigation and blocks of content.

Sufficient techniques may include:

use of anchor tags in HTML.
example should be inserted.
Use of scripting to make links visible when tabbed to.
example...
Use of tables of contents in non-HTML formats.
example...
use of scripting to appropriate direct focus dependent upon inputs.
example... yeah right.
use of consistent content block mark up such has headers.
example...







Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
Barrett, Don
Sent: Tuesday, June 12, 2007 4:17 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

We have accepted frames, headings, and skip nav links as methods of
meeting the standard. All the standard has required is a method to skip
repetitive nav links; it's been very generic and very helpful.


Don Barrett
Section 508 Coordinator
U.S. Department of Education
(202)-205-8245
= EMAIL ADDRESS REMOVED =

-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Andrew
Kirkpatrick
Sent: Tuesday, June 12, 2007 4:12 PM
To: TEITAC Web/Software Subcommittee
Subject: Re: [teitac-websoftware] Bypassing content.

The skip nav link offers very poor control - you get one option and that
is to move focus to where the link is and nowhere else.
AWK

> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> Hoffman, Allen
> Sent: Tuesday, June 12, 2007 4:08 PM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> I concur with Andrew, but.
>
> The skip-nav requirement is intended to allow the end-user more
> control of directing the focus. if an application allows focus to be
> moved without inefficient keyboard or other driving, then it meets the

> standard already.
>
>
>
> Allen Hoffman -- = EMAIL ADDRESS REMOVED = ; v: 202-447-0303
>
> -----Original Message-----
> From: = EMAIL ADDRESS REMOVED =
> [mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of
> Andrew Kirkpatrick
> Sent: Tuesday, June 12, 2007 3:48 PM
> To: TEITAC Web/Software Subcommittee
> Subject: Re: [teitac-websoftware] Bypassing content.
>
> Exactly. The concept of the skip link is a possible success
> technique, but we should not skew the requirement to eliminate other
> success techniques that people are able to use presently.
> AWK
>
> > -----Original Message-----
> > From: = EMAIL ADDRESS REMOVED =
> > [mailto: = EMAIL ADDRESS REMOVED = ] On
> Behalf Of David
>
> > Poehlman
> > Sent: Tuesday, June 12, 2007 3:45 PM
> > To: TEITAC Web/Software Subcommittee; Katie Haritos-Shea
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> > I just know there is already methodology for dealing with
> this without
>
> > using a special link though.
> >
> > ----- Original Message -----
> > From: "Peter Wallack" < = EMAIL ADDRESS REMOVED = >
> > To: "Katie Haritos-Shea" < = EMAIL ADDRESS REMOVED = >; "TEITAC
> Web/Software
>
> > Subcommittee" < = EMAIL ADDRESS REMOVED = >
> > Sent: Tuesday, June 12, 2007 11:52 AM
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> >
> > I'd argue that it's useful far beyond that. Any keyboard-only user,
> > such as a 'power user', would appreciate the feature. I believe it
> > needs to be visible until User Agents provide a standard
> mechanism to
> > invoke it.
> >
> > Peter Wallack
> > Accessibility Program Director
> > Oracle Corporation
> >
> > Katie Haritos-Shea wrote:
> > Mike,
> >
> > The visible skip nav is also useful for persons with
> limited mobility,
>
> > and the AT that they use. In that case the skip nav link must be
> > visible.
> >
> > Katie
> >
> > -----Original Message-----
> > From: "Langum, Michael J" < = EMAIL ADDRESS REMOVED = >
> > Sent: Jun 12, 2007 8:21 AM
> > To: = EMAIL ADDRESS REMOVED =
> > Subject: Re: [teitac-websoftware] Bypassing content.
> >
> >
> > AWK wrote:
> > "An invisible skip link wouldn't satisfy the Functional
> perfomance
> > criteria,. . ."
> >
> > Have I missed something?
> >
> > I thought the "function" of the "skip link" was to allow
> users of AT
> > to skip past banners, and standard navigation elements so that they
> > could quickly and easily get to the main content of a page. Since
> > sighted users can achieve the same "function" by simply
> examining the
> > page and choosing to begin reading at whatever point they wish (and
> > hence don't rely on the "skip link"), I don't see how an "invisible
> > skip link wouldn't satisfy the Functional performance criteria."
> >
> > -- Mike Langum
> > Asst. Webmaster, WWW.OPM.GOV
> > U.S. Office of Personnel Management
> >
> >
> >

From: Sailesh Panchang
Date: Tue, Jun 12 2007 2:45 PM
Subject: Re: Bypassing content.

Don,
>We have accepted frames, headings, and skip nav links as methods of
>meeting the standard. All the standard has required is a method to skip
>repetitive nav links; it's been very generic and very helpful.
Do lists not make the cut in your opinion?

Sailesh Panchang
Senior Accessibility Engineer
Deque Systems Inc. (www.deque.com)
11130 Sunrise Valley Drive, Suite #140,
Reston VA 20191
Phone: 703-225-0380 (ext 105)
E-mail: = EMAIL ADDRESS REMOVED =