Access Matters

Dennis in Web Axe Episode 24 mentions”

If the tab orders are different than the actual flow of the content on your page, it’s going to be really disorienting for the (screen reader) user.

Really? We know that this sort of disorder will be apparent in regular browsers. That happens with screen readers? Please try a test case where I have intentionally made the tabindex values different that their order in the source. Use your favorite screen reader to tab from link to link and tell us what sequence you hear.

Recently, Bruce Lawson talked about receiving a copy of his new book, a collaboration with a number of well known names in the accessibility field:
Web Accessibility: Web Standards and Regulatory Compliance by Andrew Kirkpatrick, Bruce Lawson, Richard Rutter, Christian Heilman, Jim Tatcher and Cynthia Waddell. While engaging in a bit of unwarranted self-flagellation, Bruce mentioned an “Accessibility Old Wives Tale” that I thought still worth exploring. Bruce says:

…in the introduction I commit a cardinal sin that I only recently berated Andy Budd for:

One of my biggest bugbears, the Accessibility Old Wives’ Tale™, is to be found on page 130: “.. many screenreaders will ignore text between form elements, unless they are enclosed in a label.” So, which screenreaders are those? … So this is a plea to all authors, not just Budd: if you make statements like “some browsers” or “many screenreaders”, please identify the culprits.

I believe this is really a problem of modality. In straight through reading mode, one probably hears everything spoken. Then, in forms interaction mode, only field labels are spoken. Let’s find out. Please try a very simple test case and let us know if the extra paragraph between the form fields is spoken while you are interacting with the form. Since we’re trying to be more precise that “many screenreaders,” make sure you tell us which reader and version.

When I put this site into hiatus late last year, I set a couple of goals for when I would bring it back to activity. First, I wanted to do a proper design job and move away from a quickly adapted Wordpress theme. Second, I planned to ditch the quiz numbering scheme based on WCAG 2.0 while it was still in development stages. The number scheme has changed since I started using them, and WCAG 2.0 itself hasn’t turned out so well.

OK. So much for good intentions. You’ll see that the design hasn’t changed yet, and those silly numbers are still scattered throughout the content. I’m foregoing both so that I can publish a couple of test cases for situations I’ve read about recently.

By the way, the reasons for the hiatus, notable projects for my day job, have gone splendidly. Since these projects sit inside IBM’s firewall and are targeted to our 300,000+ employees, I will mention only brief details here. The first project is a forums facility that offers threaded discussions on about 1200 topics to over 40,000 participants in both web and netnews formats. This is the third generation of our forums in 25 years. It launched well early this year and the forums are flourishing with ever increasing use and participation. The other project is a collaborative help facility that has not yet launched, but will be the first to bring end user collaboration onto a set of Information Technology help resources that were previously very tightly edited and controlled by a central team. We’re bringing that application out of the “command and control” bunkers of the Web 1.0 era.

Next, a couple of quizzes (your participation expected) regarding (1) another subtlety of how forms are read by screen readers, and (2) a claim that screen readers are sometimes confused by unruly tabindex sequences.

Great work by James Edwards brings this blog briefly out of hiatus. James has completed the testing and results compilation for our JavaScript test case regarding form elements.

Quite some time ago, we published our second quiz about JavaScript accessibility. This one focused on form elements and all of the ways that JavaScript might interact with them.

As with the first set of results, they are extensive and fill a wide table. You will find them published on a wider page for better presentation.

Tell us what you think of them.

while two very intersting projects at my day job consume all available energy. I hope to resume here early in 2006.

The previous quiz question asked about supplementing ALT text with TITLE text, and how it might affect assistive technology. As always, we find variations in how assistive technology handles the various combinations. I’ll leave the recap for later. Now, we’ll turn to a more comprehensive question and test case about TITLE text. As Joe Clark rightly pointed out, the previous test case was too simple. OK, Let’s try a more complete test. TITLE text can be applied to almost any HTML element, so that’s what I’ve done.

The Titles Everywhere test case has TITLE text applied to every element. I’ve used almost all the elements, even those which are containers for others. A TITLE on an UL? What good is that? I don’t know, but the specs allow it. It’s interesting to see what appears on hover for each of these, and it might be interesting to see what assistive technology does with them. Try the test case and report any interesting findings.

Since every quiz needs a question…

Q. For what elements do you normally use TITLE text?

• A usually only images
• B images and … itemize