Re: Request to publish HTML5: Techniques for providing useful text alternatives as a first public working draft

>We could define that the @alt content attribute contains replacement
>text for the image, and that its presence indicates that AT users
>should be presented with the alt text and not be told about the
>existence of the image.

I don't think this is a good idea, I suggest the user agent whatever
it is, should provide a user with options to be not notified or
notified of the presence of an image along with the presentaion of the
text alternative.  The usefulness of knowing an image is present
depends very much on the user.

There are a number of situations where denying a user the information
that an image is present reduces the users ability to understand and
interact with the content.
For example, users with images disabled in Firefox, currently have no
way to identify where images are on the page, so they they can
load/view/copy  individual images if they want to. i contrast the lynx
text browser,  provides a number of modes 1 which provides text
alternatives without indication of an image, another where the image
is linked with the alt text as the link text. This enables users to
download or view the image in an associated graphics viewer.

I think you are correct that many instances of text alternatives are
less than optimal in relation to the ideal of what a text alternative
should be. But in many cases the less than optimal text alternative is
better than the alternative of no alt.

> 1. All these documents are declared incorrect (and ideally someone
> should spend time evangelizing and educating these authors)

I believe this is currently the case in hixie's version of what
constitutes a conforming alt attribute value in HTML5.
The document I am editing provides a different set of conformance
criteria based on providing the best information to users within the
constraints of user agent behaviour.
There has been a lot of envangalizing about what should be considered
correct text alternatives in any given context, there is not always
agreement on what is a correct text alternative in some contexts.
Rather than attempt to mandate the concept of one correct way, i am
attempting to provide authors with some informed choices about what
text alternatives they provide and what methods they use to rpovide
them.

regards
stevef
On 7 May 2010 10:56, Jonas Sicking <jonas@sicking.cc> wrote:
> On Fri, May 7, 2010 at 2:08 AM, Steven Faulkner
> <faulkner.steve@gmail.com> wrote:
>> Hi Jonas,
>> I am unclear about what you are saying here:
>>
>> "My concern is that with this definition of alt, we'd instantly not
>> only make a large body of content inaccessible. We'd also do AT users
>> a disservice by telling AT software to hide the fact that there is an
>> image there since the alt attribute was used."
>>
>> which definition of alt?
>> In what context would we instantly make a large body of content inaccessible?
>
> Sorry, I meant to say "incorrect" instead of "inaccessible".
>
> We could define that the @alt content attribute contains replacement
> text for the image, and that its presence indicates that AT users
> should be presented with the alt text and not be told about the
> existence of the image. And that putting a description (rather than
> alternative text) would be incorrect.
>
> However for existing documents out there that contain descriptions in
> the alt, then this results in two problems:
>
> 1. All these documents are declared incorrect (and ideally someone
> should spend time evangelizing and educating these authors)
> 2. Until the document is fixed, users of AT tools will be at a
> disadvantage because they aren't being told about the existence of the
> image.
>
> / Jonas
>



-- 
with regards

Steve Faulkner
Technical Director - TPG Europe
Director - Web Accessibility Tools Consortium

www.paciellogroup.com | www.wat-c.org
Web Accessibility Toolbar -
http://www.paciellogroup.com/resources/wat-ie-about.html

Received on Friday, 7 May 2010 10:32:56 UTC