Friday, May 5th, 2006

AJAX and Screenreaders: When Can it Work?

Category: Accessibility, Ajax

One of the largest complaints that people have about Ajax these days is the lack of standards support in the sites developed with it. Since all of the data is dynamic, there’s problems caused with several different types of applications used for “alternative browsing” – such as screen readers. Developers know this and are trying to take steps to improve the situation, but it’s a slow road so far. In this new article on SitePoint, you can check in on one developer’s quest to get a simple Ajax script to be correctly read in a common screen reading application.

We’ve all heard a great deal of buzz about AJAX in the last few months, and with this talk has come a legion of articles, tips, presentations and practical APIs designed to explore the possibilities and try to arrive at best-practice techniques. But, for all of the excitement and hype, still very little has been said on the subject of AJAX and accessibility.

Wherever we look, from discussions at AccessifyForum, to popular blogs like those of Derek Featherstone, and Peter Paul-Koch, the one thing we can all agree on is that we need more information.

And that’s why I’ve written this article: to present some of the data and analysis I’ve compiled, and see if it points to a useful conclusion.

He gives a little background first, noting that he’s been at this for a while, working for a few months before on how screen-readers parse through a web site and do their thing. He’s found methods to get the applications to read certain parts of the site better than just the default, but the real problem is the update of the page via the DOM. There’s just not a reliable way he’s seen to let the software know the page has been changed.

That doesn’t stop him from trying, though, and he goes through seven variations of the same page of information, trying to send something back to the software to let it know there’s something different. Unfortunately, in the end, there just wasn’t a reliable way that he found to get the job done. Not even a Javascript alert() box worked in one of the applications. Even more unfortunately, this problem seems to lie with the screen-reading applications themselves, and not with the content of the page. They’re just not made to handle this sort of page alterations yet.

Posted by Chris Cornutt at 7:08 pm
7 Comments

+++--
3.5 rating from 51 votes

7 Comments »

Comments feed TrackBack URI

Anyone know if most screenreaders support window.open() and immediately read the contents? If so, pop a window, document.write in your update message and include a close() link? Since it would be unnecessary, it could be a site pref…

Comment by Stephen Clay — May 5, 2006

I don’t think screen readers have any sort of scripting abilities at all. That’s one of the most critical reasons I can see for making sure that all the functionality you’re providing with AJAX has a non-AJAX variant. The whole of the website should be accessible without CSS and with Javascript disabled.

Comment by Mike Ritchie — May 6, 2006

Agree with Mike- it’s generally not worth it to make a dynamic application that works with screen readers. In fact, any site with sophisticated design that users javascript/css is probably going to be mangled by the screen readers. I look at the accessible version of a site as being roughly equivalent to the mobile version of a site- in fact- an xhtml site for phones/pdas is probably the best kind of site at which to aim your screen reader…until the mobile browsers start supporting heavy JS.

Comment by matt m — May 6, 2006

the problems lie with the screen readers.

Comment by ajaxmeoff — May 7, 2006

Here’s what I learned from Freedom Scientific, makers of the popular JAWS screen reader:

From: Freedom Scientific Technical Support [support@freedomscientific.com]
Sent: Friday, April 28, 2006 7:39 AM
To: Kevin Hakman, TIBCO Software Inc., TIBCO General Interface
Subject: Re: AJAX?

Dear Mr. Hakman:

Thank you for contacting Freedom Scientific Technical Support. The JAWS screen reader is scheduled to support the AJAX style of web site design in our up coming release of JAWS 7.10.

If you have any additional questions regarding this or any other issue, please don’t hesitate to contact us.

If replying to this message, Be sure to include all previous correspondence pertaining to this matter so that we might better assist you.

Sincerely,

Dennis G. Godin
Technical Support Specialist
Freedom Scientific Blind/Low Vision Group

http://www.freedomscientific.com
Phone support: 727 803 8600

Our Mission:
To develop, manufacture and market innovative technology-based products and services that those with vision impairments and learning disabilities use to change their world.

Comment by Kevin Hakman — May 8, 2006

I wonder if he thinks his “tests” are universal. Most of the people I know use higher versions of JAWS. Here at work we run 7.0

Comment by RyanB — May 9, 2006

JAWS 7.0 behaves identically to JAWS 6 in respect of these tests (I have subsequently run the tests in v7).

The significant change is between 5.0 and earlier versions, and 6.0 and later versions, at which points some kinds of dynamic changes stop working (ie, the later versions have less dynamic capability than the earlier, which the test reults note).

It’s pretty clear that AJAX apps are not going to work in the current crop of screenreaders, but that’s not the significant point of the article – the point is that these devices fall through the net of progressive enhancement; they are script capable browsers, and as far as object testing reveals, they’re just a vanilla version of IE or Firefox, with the same capabilities; yet they’re not, and their users won’t necssarily know what’s wrong.

AJAX applications open a whole new world of access problems for blind users, and others with specific access requirements. If anything can be the interface trigger, and anything could change in response, how does a screenreader user know where to start? How can they make sense of such an app?

It’s easy to say that the screenreaders are broken, and it’s true in a sense, but it doesn’t help – it’s just passing the buck. We hope the technology will step up to the challenge – perhaps we’ll have to make it – but in the meantime we can’t just shrug and forget about it.

Comment by brothercake — May 21, 2006

Leave a comment

You must be logged in to post a comment.