Is web accessibility inaccessible?

Posted in Sandi's blog on 4 October 2011 | 4 Comments

I gave the keynote presentation at the recent A11yLDN event, where I laid out the stall for what I believe is the future of web accessibility. I called my talk, "Does anyone know the way to Web Accessibility utopia?". I worried about being contentious, well, not really, when I asserted that accessibility utopia does not exist, that the WCAG is out of date and that accessibility is a subset of usability. I pontificated about inclusive design, about being reasonable and that no one creating a website has limitless resources.

Much to my surprise, instead of the expected AA, DDA and alt text rebuttals, it was like the room heaved a huge sigh of relief. There was an overwhelmingly positive response, particularly from mainstream tech folk who are desperate to engage but need something "real". The trouble is that accessibility is actually quite inaccessible.

Accessibility is subjective, immeasurable and unattainable, yet it's heralded as being absolute, determinable and easy to achieve if you follow the rules. Where are the rules? En route to Accessibility Utopia and that's where they belong. Of course, I want equality for all, but I am also aware that this is an ideal whether I like it or not.

Accessibility is about everyone participating in civil society, where access to information is essential. All people should be able to access, use and interact with online content, but the misconception that inalienable human rights are immutable in law makes folk think it's all or nothing. But human rights are where it all began and most importantly it's about human beings.

People are all unique, yet we all share the same human characteristics and abilities in a multitude of configurations, so understanding the range of human behaviours and abilities - in all of their complexity and unpredictability - is the way forward. Segmenting a person's needs based on a single disability without considering any other factors is absurd. When I registered blind and was assessed by Access to Work, they gave me JAWS. Of course, all blind people use JAWS. Right? Guess they didn't get the memo about the fact that 97% of folk who are registered blind have some residual vision? At this point I did not need a Screenreader, but even if I did, JAWS is incompatible with my ADHD brain and I nearly threw the computer out of the third story window.

Accessibility cannot be measured by technical conformance or policy compliance. It can only be measured by disabled people using websites as intended for their non-disabled peers. In practice, this means being able to complete specific tasks, such as registering with an online service, finding an organisation's contact information or commenting on a blog.

Léonie Watson did a lightening talk about AlphaGov and GovUK. Léonie is a long standing accessibility advocate and has the unenviable position of fighting the accessibility corner for GovUK. There was outrage when, after spending a reported £261k, AlphaGov launched a website that many folk found to be inaccessible and Léonie was brought in to put things right. However, the project has not adequately invested in research and I just don't think that randomly asking people to provide feedback on AlphaGov is an appropriate substitution. A project of this magnitude must integrate diversity into the thinking, starting with high quality quantitative and qualitative research, if they have any hope of getting this right. I really wish our industry would start thinking of websites as products that people interact with and approach design accordingly.

In stark contrast was the event the following day in the Helen Hamlyn Centre at the Royal College of Art, ironically titled "The Problem Comes First". The inspiring grown-up professional show demonstrated how proper product design is done - by clearly identifying problems and conducting extensive user centred research to find solutions, long before scoping or even dreaming of developing a prototype. It felt like a homecoming; I cannot count high enough to know how many times I have said that technical innovation and inclusive design are not mutually exclusive, but I don't need to tell the Helen Hamlyn Professor of Design Jeremy Myerson that. He's been named by Wired as one of Britain’s 100 most influential people in technology. Geeks take note.

Sandi Wassmer smiling

About Sandi

Businesswoman Sandi Wassmer registered blind in 2008. In her blog, she shares with us the 'shenanigans of visual impairment'. Find Sandi on Google+

Read Sandi's full profile | List all Sandi's posts
Sandi's RSS feed | Sandi's Twitter account

Comments

  • "Accessibility cannot be measured by technical conformance or policy compliance."

    Sandi, while I don't disagree with your over-arching sentiment, you haven't addressed a key problem on the landscape that *must* be addressed: the Lawyers.

    Shakespeare may have written "The first thing we do, let's kill all the lawyers", but the reality is that we can't do that either. We need a measurable standard, whatever that might be, so that we have our equivalent of the Rosetta stone. It won't be perfect (what is?) but it will be a common reference that sets out for all some basic shared understandings of requirements and accountability. Without that shared touch-stone, those same mainstream tech folk that heaved a sigh of relief will be left with being their own judges of what does and does not constitute "accessible"; and what is accessible to George may very likely not be accessible to Susan, but then, what is "accessible"? These are the types of questions that lawyers get paid obscene sums of money to argue. For large (and today even mid-sized and smaller) web-site owners, this is a real issue: with advocacy groups such as the U.S. based NFB (National Federation for the Blind) using their bully pulpit and aggressive lawyers, they are going after web content owners who have inaccessible web content. No matter which side of that discussion you are on, how do you measure accessible versus ‘inaccessible’ in a court of law? How do you meet a community measure of conformance without guidelines and conformance checking? How do you guide developers and outside contractors to your end goal without policies and technical requirements? Big business simply cannot take the risk or not having those types of tools at their disposal, and suggesting that they all simply grok what each and every user requires and deliver on that in a humanistic fashion is…, well, it ain’t gonna happen in our lifetimes.

    It is well and good to say that every project should include disabled testers in their entire start-to-finish publishing process to ensure that the end users can register on-line, or find an organizations contact number or leave a blog comment, but the other reality is that there *is* an additional cost to bringing in additional testers, and frankly given the explosive growth that exists on the web, even today, not every project is going to have that level of testing or resources at its disposal. How do we help those developers succeed? The sad truth is, we give them some guidelines and basic rules that they should understand and follow, and then rely on their (hopefully) good sense to not be literal in the interpretation of those rules. What we can’t do is tell them, “hey, it’s about the users, so understand their needs and do what’s best”, because for many users they have no yardstick to measure that against.

    It’s not “all or nothing”, but it has to be “something”, and it has to be scalable to fit the needs of the larger web – which means publishing standards and using conformance tools to measure against those standards. Wishing for otherwise is an even worse kind of Utopian dream.

    Posted by John Foliot, 5 October 2011 (2 years ago)

  • I agree mostly with John. I understand the big picture you're getting at but there has to be something to measure accessibility by. the WCAG may be out of date but at least it's some kind of guideline. without a basic set of guidelines or rules developers will have good intensions of making content accessible but might not know where to start or they might not even think about accessibility at all because they simply don't know about it or don't think about it. accessibility is tricky as you pointed out because what may be accessible for you may not be accessible for me but at least guidelines give developers things to incorporate and make it so they don't have to try to figure things out on their own. I don't think we'll ever reach accessibility utopia but we can combine research and user experience and try to come up with the most comprehensive set of guidelines possible and of course keep them up to date. it's an ongoing process but I think if we all work together we can inform and make the web a better place.

    Posted by Jessica Rathwell, 5 October 2011 (2 years ago)

  • Note to the ever articulate Mr Foliot,

    I agree with you completely. Web Standards are the foundation and the guidelines, techniques and best practices for accessibility (as well as all of the other disciplines that go into making great web stuff) are essential.

    However, the ultimate test of whether a website is accessible or not is the ability for folk to use it. Its about task completion.

    I was simply trying to get thinking beyond technical conformance, as so many people see this as the holy grail. There must be a middle way, but never at the expense of good technology.

    Sandi

    Posted by Sandi Wassmer, 5 October 2011 (2 years ago)

  • @Jessica, See, I disagree with both you and Sandi that WCAG is out of date.

    WCAG 1 is out of date, but WCAG 2 went to great pains to not be a shopping list of tick boxes, but instead has at its foundation POUR - perceivable operable, understandable and robust. It then goes on to illustrate common issues and problems, and supplies examples of what success criteria might look like, but if you have a better way to skin the cat, and the outcome(s) meets the POUR requirements then have at it.

    How can something as abstract as "your web content must be perceivable, operable, understandable and robust" be "out of date"?

    Posted by John Foliot, 6 October 2011 (2 years ago)

Post your comment

Leave a comment