- From: Joe Clark <joeclark@joeclark.org>
- Date: Tue, 23 May 2006 17:14:19 +0000 (UTC)
- To: public-comments-wcag20@w3.org
<http://joeclark.org/access/webaccess/WCAG/response1_Techniques-WCAG2.html>
General
Instant-messaging applications are not Web content
An instant-messaging application uses two simple glyphs to indicate
a person's instant messaging status.
An instant-messaging application is not Web content. (Yes, I know,
it's possible to use the Web as a front end for instant messaging, as
with Google Chat and Meebo. Those are not applications.)
Text equivalents
alt=" " is officially permitted alongside the actually correct alt="".
Incorrect CSS or usage
The Techniques document gives incorrect CSS in places, or simply gives
unrealistic examples that don't match the practices of
standards-compliant developers.
1.
/* Rules for bidi */
HEBREW, HE-QUO {direction: rtl; unicode-bidi: embed}
ENGLISH {direction: ltr; unicode-bidi: embed}
/* Rules for presentation */
HEBREW, ENGLISH, PAR {display: block}
EMPH {font-weight: bold}
Each of those class names must begin with a dot (e.g., .HEBREW),
and all-lower-case is preferred, as XML documents have
[3]case-sensitive CSS parsing.
2.
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="<url>http://www.w3.org/1999/xhtml">http://www.w3.org/1999/xhtml</
title>">
<html>
<head>
<title>A study of population dynamics</title>
</head>
<body bgcolor="white">
<p> ... document body...</p>
</body>
</html>
body bgcolor="" is an incorrect usage; that's what we have CSS
for. (Same for body color="" later.) Note that the DOCTYPE URL is
mangled.
3.
Procedure
1. Examine the code of the Web unit.
2. Check to see if a foreground color is specified.
3. Check to see if a background color is specified.
Colours may have been specified in preceding selectors. The C in
"CSS" does stand for "cascading," after all. As written, this
technique authorizes pedants to write in and complain that not
every single element has foreground and background colours
explicitly stated. They don't have to be.
Trivial examples
One of the most common examples of using inconsistent labels for
components with the same function is to use a button that says
"search" in one page and to use a button that says "find" on
another page when they both serve the identical function.
And the words have identical meaning.
Failure Example 4:
An E-commerce application uses a printer icon that allows the user
to print receipts and invoices. In one part of the application, the
printer icon is labeled "Print receipt" and is used to print
receipts, while in another part it is labeled "Print invoice" and
is used to print invoices. The labeling is consistent ("Print x"),
but the labels are different to reflect the different functions of
the icons. Therefore, this example does not fail the success
criterion.
So why is it in there?
Testing in browsers
The blink value of the text-decoration property is not supported by
Internet Explorer.
For Windows or Mac?
It is supported in Netscape/Mozilla family browsers. Not tested in
others (e.g., Safari, Opera).
And why wasn't it tested in those "others"?
Blinking (but not flashing)
1.
CSS defines the blink value for the text-decoration property. When
used, it causes any text in elements with this property to blink at
a predetermined rate. This cannot be interrupted by the user, nor
can it be disabled as a user-agent preference.
User stylesheets would seem to be useless in this regard, as they
do not permit the rewriting of selectors. However, User Agent
Accessibility Guidelines [4]Checkpoint 3.3 requires the ability to
toggle blinking text. This appears to be a user-agent problem, in
whole or in part.
2.
On a page with moving or scrolling content,
1. Check that a mechanism is provided in the Web unit or user
agent to pause moving or scrolling content.
2. Use the pause mechanism to pause the moving or scrolling
content.
3. Check that the moving or scrolling has stopped and does not
restart by itself.
4. Check that a mechanism is provided in the Web unit or user
agent to restart the paused content.
5. Use the restart mechanism provided to restart the moving
content.
6. Check that the movement or scrolling has resumed from the
point where it was stopped.
But the Understanding document uses an example of a stock ticker:
A stock ticker has "pause" and "restart" buttons. Pausing the
ticker causes it to pause on the current stock. Restarting causes
the ticker to jump ahead to the current stock. Stocks that were
updated during the pause will not be displayed.
So which is it: Once you resume you have to start from where you
paused, or can you skip what you missed?
Multimedia
1.
Examples of text streams that are not captions include... subtitles
that do not include important sounds
No "subtitles" are "captions." By implication, this technique
permits subtitling into a foreign language as long as non-speech
information is included.
2.
Procedure
1. View the material with captioning turned on.
2. Check that all dialog is accompanied by a caption.
3. Check that all important sounds are captioned.
And how does a deaf person do this?
Keyboard access
Tests
Procedure
1. Using a keyboard, navigate through the content.
2. Check to see that the keyboard focus is not "trapped" and it is
possible to move keyboard focus out of the plug-in content without
closing the user agent or restarting the system.
What if this depends on the user agent, as it so often does? What if
nothing the author can do will ever permit the user to escape from the
trapping content? (This was a real example with accessible Flash.)
Popup windows
1.
Failure due to opening new windows when the user does not expect
them. New windows take the focus away from what the user is reading
or doing. This is fine when the user has interacted with a piece of
user interface and expects to get a new window, such as an options
dialogue. The failure comes when pop-ups appear unexpectedly.
Actually, WCAG 2 [5]bans all popup windows without explicit alert
beforehand (though at Level 3).
2.
A user clicks on a link, and a new window appears. The original
link has no associated text saying that it will open a new window.
[...] Check if elements that open new windows have associated text
saying that will happen. The text can be displayed in the link, or
available through a hidden association such as an HTML title
attribute.
target="_blank" is programmatically determinable and it is up to
the user agent to warn the user. JavaScript is another story and
should be addressed by the Techniques.
3.
A user clicks on the body of a page and a new window appears. No
indication that the area that was clicked has functionality is
present.
If that area were marked up as a or indeed as area, such warning
is programmatically determinable and it is up to the user agent to
warn the user.
Linguistics & typography
1.
The objective of this technique is to show how using a non-text
mark to convey information can make content difficult to
comprehend. A non-text mark may be non-text content such as an
image, or a font glyph which is not text nor an image of text.
Font glyphs are text. This whole section is nonsense.
2.
The objective of this technique is to describe how using blank
characters, such as space, tab, line break, or carriage return, to
format individual words visually can be a failure to present
meaningful sequences properly. Blank characters have no appearance
when rendered visually
Yes, they do. They're blank!
You are here: [6]joeclark.org -> [7]Captioning and media access ->
[8]Web accessibility -> [9]WCAG -> Response to `Techniques for WCAG
2.0'
Updated 2006.05.23
References
1. LYNXIMGMAP:http://joeclark.org/access/webaccess/WCAG/response1_Techniques-WCAG2.html#joeclark_angie_02IX_Map
2. http://www.w3.org/TR/WCAG20-TECHS/
3. http://www.w3.org/TR/CSS21/selector.html#q1
4. http://www.w3.org/TR/UAAG10-TECHS/guidelines.html#tech-on-off-blinking-text
5. http://joeclark.org/access/webaccess/WCAG/WX#%23context-changedef
6. http://joeclark.org/
7. http://joeclark.org/access/
8. http://joeclark.org/access/webaccess/
9. http://joeclark.org/access/webaccess/WCAG/
Received on Tuesday, 23 May 2006 17:14:47 UTC