E-mail List Archives
Thread: What assistive technology does (and doesn't do)
Number of posts in this thread: 11 (In chronological order)
From: Rachel Tanenhaus
Date: Tue, Dec 16 2003 1:01PM
Subject: What assistive technology does (and doesn't do)
No previous message | Next message →
Hello!
While giving a web accessibility workshop last week, I was approached by
a few web designers who had what I thought was an excellent question. I
had been explaining that there was no need to reference a link in the
alt text on an image map (in other words, one doesn't have to include
"link to <foo>" in the alt text), because user agents, including various
forms of assistive technology, were capable of identifying links.
"So all assistive technology [used for browsing the web] can identify
when something is a link?" they asked.
I answered in the affirmative - at least I *think* it all can - and they
wanted to know if there existed some list of functions that all AT could
perform, so that they could design their web pages accordingly. (frex:
they can all identify when something's a link, so you don't have to) Is
there a list of standards to which screen-readers, refreshable Braille
displays, etc. are designed? I don't know of one, but then again,
there's lots of stuff I don't know. :) And if no such list exists,
then how do we know that "all AT" performs a given function, and at what
point do we say that enough AT does something that we can design web
pages under the assumption that most AT can do that thing?
Thanks muchly,
Rachel
Rachel H. Tanenhaus, MPH
Information Specialist
New England ADA & Accessible IT Center
374 Congress Street, Suite 301
Boston, MA 02210
Phone: (617) 695-0085 (v/tty)
or (800) 949-4232 (v/tty) (in New England)
Fax: (617) 482-8099
E-mail: = EMAIL ADDRESS REMOVED =
URL: www.NewEnglandADA.org
----
To subscribe, unsubscribe, suspend, or view list archives,
visit http://www.webaim.org/discussion/
From: julian.rickards
Date: Tue, Dec 16 2003 3:39PM
Subject: RE: What assistive technology does (and doesn't do)
← Previous message | Next message →
Hi Rachel:
There are two parts to your question.
(1) Links in image maps. There seems to be a bit of a conflict in
information between WCAG check points 1.1 and 1.5. Check point 1.1 (Priority
1) states that images and regions in image maps are to be given alt text
which seems to suggest that hotspot regions are "visible" to ATs. However,
1.5 (Priority 3) states that equivalent text links should be provided for
all hotspot regions in image maps which seems to suggest that at least some
ATs may not "see" hotspot regions. I suggest that text links be provided
just in case.
(2) Links are an integral part of the WWW and for that reason (I don't have
any actual experience with ATs), I believe that ATs would announce in some
means or another to the user that a link has been encountered. Other
circumstantial evidence is based on the current practice of creating link
text. In the past, many of us wrote "Click here to go my resume" with "click
here" as the link text. The phrase "click here" is mouse centric and many
accessibility professionals recommend dropping any reference to the mouse
because not everyone uses a mouse. Furthermore, "click here" does not
identify the content of the destination page. With those recommendations,
links are now being rewritten as "For more information, read my resume"
where "my resume" is link text. Given that, if links were not announced to
an AT user, they would read right past "my resume" without any indiciation
that they just went past a link. Therefore, it is my belief that links are
identified to an AT user. Furethermore, since AT users are provided with
that information when they encounter a link, I don't feel it is necessary to
use the text "link" in Alt or Title text, they know it is a link. If a
sighted user can understand from the blue underlined text that "my resume"
is a link to my resume, then don't suggest that people with little to no
sight are any less intelligent - alt text on a link graphic should provide
the same text information as in the graphic.
Jules
---------------------------------------------------------
Julian Rickards
Digital Publications Distribution Coordinator
Publications Services Section
Ontario Ministry of Northern Development and Mines
Phone: (705) 670-5608
Fax: (705) 670-5690
>
From: Rachel Tanenhaus
Date: Tue, Dec 16 2003 2:00PM
Subject: RE: What assistive technology does (and doesn't do)
← Previous message | Next message →
Hello!
Thanks so much for the prompt response! I too am all in favor of
including text links and alt text with the hot spots. And I'm certainly
willing to agree that there's plenty of evidence that pretty much all AT
recognizes links on web pages. I think the question the designers had
was more along the lines of, "Well, if there's ONE behavior that we can
attribute to all AT, are there others? Is there a list of such
behaviors? And if so, at what point would a behavior make a list - if
80% of the available technologies perform the behavior, if one
technology with 90% of the market performs the behavior, etc.?"
I do very much appreciate your response, and I'm going to save it and
refer to it when I provide trainings and technical assistance. Thank
you!
-Rachel
Rachel H. Tanenhaus, MPH
Information Specialist
New England ADA & Accessible IT Center
374 Congress Street, Suite 301
Boston, MA 02210
Phone: (617) 695-0085 (v/tty)
or (800) 949-4232 (v/tty) (in New England)
Fax: (617) 482-8099
E-mail: = EMAIL ADDRESS REMOVED =
URL: www.NewEnglandADA.org
From: Tim Harshbarger
Date: Tue, Dec 16 2003 3:08PM
Subject: RE: What assistive technology does (and doesn't do)
← Previous message | Next message →
Jules,
Could the possible conflict be in the differences between server-side and
client-side image maps?
>
From: Derek Featherstone
Date: Tue, Dec 16 2003 2:20PM
Subject: RE: What assistive technology does (and doesn't do)
← Previous message | Next message →
Rachel Tanenhaus wrote:
> I think the question the designers had was
> more along the lines of, "Well, if there's ONE behavior that
> we can attribute to all AT, are there others? Is there a
> list of such behaviors? And if so, at what point would a
> behavior make a list - if 80% of the available technologies
> perform the behavior, if one technology with 90% of the
> market performs the behavior, etc.?"
Hi Rachel -- perhaps this may be more along the lines of what you are
looking for:
The Assistive Device Behaviour Chart, maintained by Gez Lemon over at
JuicyStudio
http://juicystudio.com/assistivedeviceschart.asp
It isn't completely exhaustive, but it is a great start. As we all learn
more about AT behaviours, we are encouraged to contribute to the chart so
that we all benefit.
Best regards,
Derek.
--
Derek Featherstone = EMAIL ADDRESS REMOVED =
Web Accessibility Specialist / Co-founder of WATS.ca
Web Accessibility Testing and Services
http://www.wats.ca 1.866.932.4878 (North America)
----
To subscribe, unsubscribe, suspend, or view list archives,
visit http://www.webaim.org/discussion/
From: Rachel Tanenhaus
Date: Tue, Dec 16 2003 2:32PM
Subject: RE: What assistive technology does (and doesn't do)
← Previous message | Next message →
Hi, Derek,
This is definitely along those lines; thank you! How completely nifty!
I guess my next step is to fire off a letter to the WAI folks asking
them at what point they determine that enough AT have adopted a given
behavior that it can be assumed to be more or less universal - and
therefore taken into considering when designing accessibility
guidelines.
(And Gez, I don't know if you're on this list - I've seen you on others
- but thanks for making the chart!)
-Rachel
Rachel H. Tanenhaus, MPH
Information Specialist
New England ADA & Accessible IT Center
374 Congress Street, Suite 301
Boston, MA 02210
Phone: (617) 695-0085 (v/tty)
or (800) 949-4232 (v/tty) (in New England)
Fax: (617) 482-8099
E-mail: = EMAIL ADDRESS REMOVED =
URL: www.NewEnglandADA.org
From: James Gagnier
Date: Tue, Dec 16 2003 2:44PM
Subject: Re: What assistive technology does (and doesn't do)
← Previous message | Next message →
That's exactly it. Client side image maps can be read by screen reading
technologies but server side image maps can't. Case in point,
www.mapquest.com is inaccessible to screenreaders.
James
----- Original Message -----
From: "Tim Harshbarger" < = EMAIL ADDRESS REMOVED = >
To: < = EMAIL ADDRESS REMOVED = >
Sent: Tuesday, December 16, 2003 4:08 PM
Subject: RE: What assistive technology does (and doesn't do)
Jules,
Could the possible conflict be in the differences between server-side and
client-side image maps?
>
From: Mary Martinson
Date: Tue, Dec 16 2003 3:05PM
Subject: RE: What assistive technology does (and doesn't do)
← Previous message | Next message →
Another reason to avoid link text such as "click here" is that some screen
reader users bring up a list of links and navigate from the list. So link
text has to make sense out of context. "Click here" or "more information"
doesn't mean anything without the rest of the sentence.
There is also a difference between server-side and client-side image maps.
It is my understanding that screen readers don't pick up information that's
served up by the server. So a server-side image map would have to have
redundant text links, but a client-side image map could use alt text. Could
be mistaken on this, but that's my understanding.
Mary
From: Mary Martinson
Date: Tue, Dec 16 2003 3:05PM
Subject: RE: What assistive technology does (and doesn't do)
← Previous message | Next message →
You might want to look at
http://www.juicystudio.com/assistivedeviceschart.asp
From: julian.rickards
Date: Wed, Dec 17 2003 8:03AM
Subject: RE: What assistive technology does (and doesn't do)
← Previous message | Next message →
It appears that this discussion went on for quite a bit after I had left
work and I am going through the messages - this one may have been answered
already.
Regarding your question, I just checked both check points 1.1 and 1.5 and
1.5 specifically refers to client side but 1.1 does not although it refers
to image map regions. Of the two check points, I would have suspected that
1.5 (Until user agents render text equivalents for client-side image map
links, provide redundant text links for each active region of a client-side
image map.) would have been more geared to server side but in fact it is
more clearly refering to client side than 1.1.
However, 1.5 is a priority 3 and may simply be there to ask developers to
accomodate older ATs that don't support client-side image maps. I haven't
used Lynx but if Opera's text-only mode is representative of Lynx, then
image maps are not visible to Lynx users at all other than perhaps the alt
text of the image as a whole.
Jules
---------------------------------------------------------
Julian Rickards
Digital Publications Distribution Coordinator
Publications Services Section
Ontario Ministry of Northern Development and Mines
Phone: (705) 670-5608
Fax: (705) 670-5690
>
From: Rachel Tanenhaus
Date: Wed, Dec 17 2003 8:28AM
Subject: RE: What assistive technology does (and doesn't do)
← Previous message | No next message
Tim gave me this fantabulous response:
<stuff deleted>
> I would say that if the information is not encoded in the mark-up, it
> should somehow be explicitly indicated in the page content.
That's a really useful rule! Thank you so much! (Even better, I can
explain it to designers, and they'll *get* it!)
> This is what I mean.
> <img src="picture.jpg">
> That mark-up tells you that the content is an image. It also
indicates
> where the image is located. It does not tell you what information the
> image may contain. Therefore, the alt attribute for the image does
not
> need to include information such as that this content is an image or
where > it is located (if that is important to the user.) It does need
to contain > other information that is not part of the mark-up for the
content.
> This is why it is so important to separate presentation from
structure.
> When the mark-up is serving double duty, it is no longer clear whether
> something like the h1 element is being used to indicate a level 1
header
> or just that someone wanted to use larger font sizes or whatever.
Indeed. I'll point that out to them too.
> Rachel, I personally have concerns about the approach of designing to
> common functionality. Which AT are we letting set the common
> functionality standard? Are we looking at just the most current
versions > or which versions are owned by what percentage of the user
population?
> When I answer those questions, I also seem to find serious flaws or
> concerns with the answers.
Exactly. And when I see that debate on some of these lists, I've
noticed that it's turned into something of a religious war - no one's
really going to convince anyone else anyway. :)
> I sometimes wonder if the best approach is just to design with
standards,
> test with real users (every chance possible,) remain open to
suggestions
> from customers, and use what you learn to improve upon future efforts.
La. I'm the choir. Preach on. :)
Unfortunately, so many people design web sites these days relying on
their authoring tools rather than knowledge of HTML that "design with
standards" is a new concept for them. I do emphasize in my trainings
that designers should involve real, honest-to-God people with
disabilities in evaluating their sites.
Thanks again for a useful and thoughtful response.
-Rachel
>