Every website / web application should conform to accessibility standards and adhere to good usability practices, right?

Wrong.

Though it's an admirable target to aim for, achieving accessibility standards compliance with optimal usability is not always viable. The two don't always play nicely together. Sure enough, well-formed, semantic HTML output and well though out content structure will do most of the work for you in hitting the twin targets of accessibility and usability but occasionally there is a see-saw effect. If you want to use rich interactive elements to enhance the user experience and improve the potential usability of your design, you may find that you are also making the design less accessible. What are the questions that we should ask ourselves when planning a new feature for a website or web application? The following are some of mine.

"Will it impact accessibility?"

We need to determine if our new feature will have a negative impact on the accessibility of the content. This could be due to user interactions e.g. clicking on a link, causing the page to fetch new content from the server and inserting it into the middle of some existing content. Non-visual browsers may not register the fact that this new content exists.

"Do we REALLY need this feature?"

If you can't justify the interactive element as absolutely essential, then there is little point in wasting time and effort on building something that will actually degrade the value of your content. In many cases you may feel that a feature is not strictly essential but gives your content a competitive edge by making it stand out from the crowd. This is fine; I've poured enough effort over the years into doing stuff just because it was 'cool' to know that stagnation occurs if you don't try to push the boundaries from time to time.

"Can we make it accessible easily?"

Are there any existing techniques that will allow us to build the new feature and still retain the required degree of accessibility? Looking for the path of least resistance is always a good exercise. If you find a solution, you may have to compromise your feature's functionality or the degree of accessibility slightly, but the savings in time and effort may be worth it.

"How accessible does it need to be?"

The target audience may affect how much effort, if any, you need to put into accessibility. From a purist point of view, this is blasphemy; all web content should be accessible. From a business point of view, there may be a good case for not conforming to accessibility guidelines. A friend of mine worked on some web-marketing material for a large car manufacturer. He told me that their position on web content accessibility was "Blind people don't buy cars". This seems callous at first glance, but when you think about it, the work required to ensure that all of their web-content is accessible to the blind is probably not worth the return they will get in terms on non-sighted customers. It becomes unviable from a business perspective.

It's a rather short-sighted view (if you'll pardon the pun) as accessibility could extend to users accessing the content from mobile devices. Web capable mobile phones and PDAs are commonplace these days but if your content does not display properly on a small screen, you could be shutting out more potential users than just those with a disability.

"Do we need to do it properly?"

If you are concerned only with passing automated accessibility tests then your work won't be that difficult as the validation software cannot detect whether or not your dynamically generated div with XMLHTTP imported content is vital to using the website. A NOSCRIPT tag in the page will probably see that your page validates, but that's not really in the spirit of the thing is it?

What's the final score then?

In an article written for Digital Web magazine, P-P Koch writes:

"The delicate balance between accessibility and usability needs more thought. At the moment I don't see any answers, only a few questions, one possible rule, and a potential danger. The rule is "Accessibility should not restrict usability"."

As a possible rule, this is a nice starting point but is a little too neat and tidy considering the mass of complications and assumptions that it represents. The issues of accessibility and usability in web design are, in my opinion, still too immature to strap down with rules of thumb. Until we no longer have to make compromises between our usability enhancements and accessibility guidelines then we should evaluate each piece of content on it's own merits.

For now at least, I declare a draw.

David Kinsella is currently the content strategist for KMP Interactive Marketing & Technology and also runs his blog; Web Developer 2.0 which concerns web development trends and technologies.