WebAIM - Web Accessibility In Mind

E-mail List Archives

Thread: Reality Check (was RE: accessibility for deaf)

for

Number of posts in this thread: 9 (In chronological order)

From: John Foliot - WATS.ca
Date: Sun, Apr 23 2006 8:10AM
Subject: Reality Check (was RE: accessibility for deaf)
No previous message | Next message →

Christian Heilmann wrote:
> Is there an assistive software that could translate text into sign
> language? While that could only have the grace and usefulness of
> babelfish (as language translation is a human thing and just cannot
> be automated) it could be helpful.


WAIT A SECOND!!!

Sign "language" is just that, a language - a form of immediate direct
communication between two people similar to non-deaf speaking. While
there exists TTF fonts that can provide "spell-out" signing, the
variations being discussed in this thread (American Sign Language,
British Sign Language, French and Quebecois Sign Language) exist because
there are "short-form" signs that represent concepts (words), rather
than having each word spelled out letter-by-letter.

I am not completely clear here on how this discussion could be relevant
on a web site:

1) Declare a "signing" font in your CSS (?) - this won't work as it
implies the end user has that particular font installed - a real
stretch. (BTW, I *do* have a TTF file of ASL "letters" - contact me off
list if you want a copy - I have a Braille TTF as well)

2) Post images of hand signing to spell out your text? (What happens
when images are not supported?)

3) Similar to Television transmition that provides signed translation
(often seen with political speeches for example) - an embedded <object>
(media stream) that provides simultaneous signed translation? (what
happens if the system does not support the media stream?)

Step back folks... Attempting to provide any form of sign language on
your web site will probably be less than fruitful 99.999 % of the time -
it's overkill of the magnitude akin to <table summary="This table is
used to lay out the page"> and/or <a href="" title="Link to the foobar">
(declaring a self evident link is pointless IMHO).

In any given territory, deaf users that are functioning in their
particular society have already developed coping mechanisms to deal with
the printed form - their level of reading comprehension may not be at
the highest level (thus WCAG #14.1 "Use the clearest and simplest
language appropriate for a site's content") however we don't see product
packaging in sign language, nor newspapers, magazines, junk mail or any
other form of written communication - why should the web be different?

Just my Canadian $0.02...

JF
--
John Foliot = EMAIL ADDRESS REMOVED =
Web Accessibility Specialist
WATS.ca - Web Accessibility Testing and Services
http://www.wats.ca
Phone: 1-613-482-7053







From: zara
Date: Sun, Apr 23 2006 10:00AM
Subject: RE: Reality Check (was RE: accessibility for deaf)
← Previous message | Next message →


> In any given territory, deaf users that are functioning in their
> particular society have already developed coping mechanisms to deal with
> the printed form - their level of reading comprehension may not be at
> the highest level (thus WCAG #14.1 "Use the clearest and simplest
> language appropriate for a site's content") however we don't see product
> packaging in sign language, nor newspapers, magazines, junk mail or any
> other form of written communication - why should the web be different?


Actually, I was not speaking specifically of text to sign language translation, I was speaking of providing sign language for appropriate Web media. However, while sign language and other languages like English and French, etc., are different in structure, some have looked into the issue of providing sign language for text before, whether to aid comprehension of written text or as a learning tool for the deaf. I know of Vcom3D which had developed signing avatar software that could sign text content. I am unaware of what became of this project but I remember seeing this product demonstrated at CSUN a few years ago and deaf people who attended the session were pleasantly surprised at how well it was able to sign and how understandable it was, despite certain kinks still to work out. It should be noted that this kind of product needed a powerful computer to run as well. If someone from Trace is member of this list, perhaps they could tell us more about it as, if I recall correctly, they were a partner of this project at the time.

Although many deaf people have indeed developed coping mechanisms to deal with written languages, the fact remains that it still represents a great challenge for many. Illiteracy is still an important problem among this segment of the disability community and so of course, continued efforts need to be applied to help overcome this issue as well. I feel that the deaf are often short-changed in Web accessibility, if only when it comes to applying the very few WCAG guidelines that concern them. In my mind, just the fact that we are talking about their needs, something that I rarely see on these lists, can only be a good thing.



Catherine



--
Catherine Roy, consultante

www.catherine-roy.net
514.525.9490








From: Paul R. Bohman
Date: Sun, Apr 23 2006 10:10AM
Subject: Re: Reality Check (was RE: accessibility for deaf)
← Previous message | Next message →

On 4/23/06, John Foliot - WATS.ca < = EMAIL ADDRESS REMOVED = > wrote:
> WAIT A SECOND!!!
> Step back folks... Attempting to provide any form of sign language on
> your web site will probably be less than fruitful 99.999 % of the time -

You're right that implementations of sign language by web developers
on their sites would be burdensome, impractical, and most likely
ineffective. However, on a theoretical level, a text-to-sign-language
"screen reader" could potentially be of some benefit to deaf users,
assuming that users buy/install the program on their end (or use it as
a web service). As such, the software would be an assistive technology
rather than a web design technique.

I do know that there are such programs--I've seen them demonstrated at
conferences and such--but as has been mentioned, their effectiveness
on a practical level will always be limited by the fact that the
process is a true translation from one language into another (e.g.
American English into American Sign Language or British English into
British Sign Language, etc.) and not just a transfer from one medium
(e.g. text) into another (e.g. sound, in the case of JAWS, Window
Eyes, etc.).

I suppose you could more easily translate from written text into
something like Signed English, but this is not the same thing as
full-blown sign languages such as American Sign Language or British
Sign Language.

Still, on a theoretical and experimental level, I see no problem with
the idea of having a text-to-sign-language "screen reader". If someone
can get it to work effectively (which is dependent on the success of
technologies for making any kind of automated language translations
effective), it could be a great asset for people whose first language
or preferred language is sign language.

--
Paul R. Bohman
Technology Coordinator
Kellar Institute for Human disAbilities
George Mason University




From: Sandra Andrews
Date: Sun, Apr 23 2006 10:20AM
Subject: Re: Reality Check (was RE: accessibility for deaf)
← Previous message | Next message →

As for clear and simple writing: we are thinking about podcast accessibility
over here. Should we be presenting, along with the podcast, an expanded
outline, rather than just a word for word transcription? The word for word
transcription is hardly clear and simple.

Sandy Andrews


On 4/23/06, zara < = EMAIL ADDRESS REMOVED = > wrote:
>
>
> > In any given territory, deaf users that are functioning in their
> > particular society have already developed coping mechanisms to deal with
> > the printed form - their level of reading comprehension may not be at
> > the highest level (thus WCAG #14.1 "Use the clearest and simplest
> > language appropriate for a site's content") however we don't see product
> > packaging in sign language, nor newspapers, magazines, junk mail or any
> > other form of written communication - why should the web be different?
>
>
> Actually, I was not speaking specifically of text to sign language
> translation, I was speaking of providing sign language for appropriate Web
> media. However, while sign language and other languages like English and
> French, etc., are different in structure, some have looked into the issue of
> providing sign language for text before, whether to aid comprehension of
> written text or as a learning tool for the deaf. I know of Vcom3D which had
> developed signing avatar software that could sign text content. I am
> unaware of what became of this project but I remember seeing this product
> demonstrated at CSUN a few years ago and deaf people who attended the
> session were pleasantly surprised at how well it was able to sign and how
> understandable it was, despite certain kinks still to work out. It should be
> noted that this kind of product needed a powerful computer to run as well.
> If someone from Trace is member of this list, perhaps they could tell us
> more about it as, if I recall correctly, they were a partner of this project
> at the time.
>
> Although many deaf people have indeed developed coping mechanisms to deal
> with written languages, the fact remains that it still represents a great
> challenge for many. Illiteracy is still an important problem among this
> segment of the disability community and so of course, continued efforts need
> to be applied to help overcome this issue as well. I feel that the deaf are
> often short-changed in Web accessibility, if only when it comes to applying
> the very few WCAG guidelines that concern them. In my mind, just the fact
> that we are talking about their needs, something that I rarely see on these
> lists, can only be a good thing.
>
>
>
> Catherine
>
>
>
> --
> Catherine Roy, consultante
>
> www.catherine-roy.net
> 514.525.9490
>
>
>
>
>
>
>
>



--
Sandra Sutton Andrews, PhD
http://ideal.azed.gov
University Technology Office
Arizona State University



From: zara
Date: Sun, Apr 23 2006 1:10PM
Subject: RE: Reality Check (was RE: accessibility for deaf)
← Previous message | Next message →

> As for clear and simple writing: we are thinking about podcast
> accessibility over here. Should we be presenting, along with the podcast,
> an expanded outline, rather than just a word for word transcription? The
> word for word transcription is hardly clear and simple.


I had to look this term up but do you mean by "expanded outline", a summary of the content with links provided to more complete information should one wish or need to go further ?

WCAG requires providing a text equivalent for every non-text element (including audio and video). And WCAG defines text equivalents as "written so that they convey all essential content".

So your accommodation would need to provide equivalent information (as in equal value). I would hazard to say that a word for word transcript that is difficult to understand would probably not be of much value to anyone.

That being said, I do not personally have problems with audio and video content (though my computer often does). It would be nice to know what users potentially benefiting from this adaptation think.


Catherine



--
Catherine Roy, consultante

www.catherine-roy.net
514.525.9490







From: Tim Beadle
Date: Mon, Apr 24 2006 1:50AM
Subject: Re: Reality Check (was RE: accessibility for deaf)
← Previous message | Next message →

On 23/04/06, Sandra Andrews < = EMAIL ADDRESS REMOVED = > wrote:
> As for clear and simple writing: we are thinking about podcast accessibility
> over here. Should we be presenting, along with the podcast, an expanded
> outline, rather than just a word for word transcription? The word for word
> transcription is hardly clear and simple.

Jeremy Keith addresses the podcast transcription issue in this blog post:
http://adactio.com/journal/1120

He mentions a ("wetware", ie human) service that will transcribe audio for you:
http://castingwords.com/

Regards,

Tim




From: Joe Clark
Date: Wed, May 03 2006 12:10PM
Subject: Re: Reality Check (was RE: accessibility for deaf)
← Previous message | Next message →

>From: "Sandra Andrews" < = EMAIL ADDRESS REMOVED = >
>
>As for clear and simple writing: we are thinking about podcast
>accessibility over here. Should we be presenting, along with the
>podcast, an expanded outline, rather than just a word for word
>transcription? The word for word transcription is hardly clear and
>simple.

If you're doing actual transcription of podcasts, you'll pretty much
be the first. And no, you shouldn't be summarizing per se. While
utterances are not copyrightable, recorded utterances are
copyrightable and indeed are copyrighted upon creation. Who owns the
copyright can be a complex issue, but it is not automatic that
permission to transcribe is also permission to rewrite and condense.
(Check the concept of moral rights or droit moral and also the
concept of derivative works.)

HTML has poor semantics for transcription, another of the many facts
that doom podcasting from an accessibility perspective, but if you
are least using headings here and there it would help.

--

Joe Clark | = EMAIL ADDRESS REMOVED =
Accessibility <http://joeclark.org/access/>;
Expect criticism if you top-post




From: Joe Clark
Date: Wed, May 03 2006 12:20PM
Subject: Re: Reality Check (was RE: accessibility for deaf)
← Previous message | Next message →

>In any given territory, deaf users that are functioning in their
>particular society have already developed coping mechanisms to deal
>with [print]... however, we don't see product packaging in sign
>language, nor newspapers, magazines, junk mail or any other form of
>written communication - why should the web be different?

I do not support wholesale use of sign language as a Web
accessibility method, but it certainly has its place. You don't see
sign language on cereal boxes because it's impossible. It *is*
possible on the Web.

The BBC accessibility study that is much forgotten now
<http://joeclark.org/axxlog/2003/2003a.html#bbci>; found that deaf
subjects would have preferred more sign language. Something weaker
came out of the Disability Rights Commission study (*still* not
online in HTML
<http://www.drc-gb.org/publicationsandreports/report.asp>;; I still
have my version up <http://joeclark.org/dossiers/DRC-GB.html>;). There
are a few deaf sites that attempt to use sign language
<http://blog.fawny.org/2005/05/23/deaf/>;.

So there is some modest demand. But indeed, if this were a huge
problem we'd already have heard about it. And *requiring* translation
into one or more sign languages opens up a large can of worms. In the
example I keep giving, nothing would stop Ukrainian-speakers at that
point from demanding Ukrainian-language sites for "accessibility."

--

Joe Clark | = EMAIL ADDRESS REMOVED =
Accessibility <http://joeclark.org/access/>;
Expect criticism if you top-post




From: Joe Clark
Date: Wed, May 03 2006 12:40PM
Subject: RE: Reality Check (was RE: accessibility for deaf)
← Previous message | No next message

>From: "zara" < = EMAIL ADDRESS REMOVED = >
>[...]
>WCAG requires providing a text equivalent for every non-text element
>(including audio and video). And WCAG defines text equivalents as
>"written so that they convey all essential content".

I.e., a transcript (or captioning for video).

>So your accommodation would need to provide equivalent information
>(as in equal value). I would hazard to say that a word for word
>transcript that is difficult to understand would probably not be of
>much value to anyone.

I would hazard to say there are scarcely any such examples. If we
can't understand it in writing, we won't be able to understand it in
audio, either. In that case, both groups are equally disadvanteged;
more likely, the podcaster will have enough functioning brain cells
to realize the recording is not fit for broadcast.

--

Joe Clark | = EMAIL ADDRESS REMOVED =
Accessibility <http://joeclark.org/access/>;
Expect criticism if you top-post