Accessibility from the Front Line - A UK Industry Perspective of Web Accessibility

Jon Dodd

Institution:

Technical Director, Bunnyfoot Universality

Email:

jon@bunnyfoot.com

In 1999 when I started my usability consultancy, web accessibility was in a very sorry state. So has anything changed in the intervening five and a bit years?

Well I’m happy to report things are improving steadily, but there is still much to do, particularly if we are to receive the benefits of accessibility above and beyond just ticking the boxes to achieve mere technical compliance. In the following short article I describe some of the progress that has been made, and some of the challenges and potential solutions for the future of web accessibility and web design in general.

Accessibility five years ago

Five years ago I started a new usability consultancy. We needed to find a niche in the market because we were newcomers and had not spawned from an existing agency with the benefit of taking clients with us. We decided to specialise in accessibility, as well as usability, for two main reasons. Firstly, the state of web accessibility was woeful, as even a cursory browse of pretty much any site would testify. Secondly, no UK usability company was doing accessibility, whereas we saw it as a vital factor that would increase in prominence and importance as the industry developed.

Many of the barriers we had to promoting our accessibility services were born out of a widespread ignorance and misunderstanding within the industry of the whole accessibility issue. In the end it boiled down to the fact that clients were not requesting accessibility and therefore agencies were not delivering it.

Increased awareness and education was what was required. We tried to give a more commercial slant to the great efforts of organisations such as the Royal National Institute of the Blind (RNIB - [1]). We ran business focused seminars, generated PR through a concept piece that required good accessibility to work (the UK’s first Internet enabled car), and targeted large organisations through innovative (or so we thought) Braille mailings. In those days though the thing that created the most impact were videos from user testing sessions involving people with disabilities. These videos were viewed with fascination and were often greeted by comments such as “I really didn’t know blind people could use computers or the net”. Clearly we had a long way to go.

The accessibility landscape now

So how far have we come? Well quite far but not far enough.

Thankfully base level awareness is now not such an issue. The videos still go down extremely well but the level of ignorance surrounding the whole issue has declined massively. Evidence for this includes:

• The ubiquitous presence of accessibility in most web site requirement specifications - clients have clearly caught on.
• The increased number of specialised agencies out there to assist with accessibility - this would not have happened if there was not the demand.
• The increased promotion of accessibility by industry media and by design agencies in their marketing - clearly it is being viewed as an issue of importance.

Much of this increased awareness, I think, has been stimulated by threat of penalties or legal action under legislation such as section 508 of the rehabilitation act in the US, or the Disability Discrimination Act in the UK. Although accessibility provides many positive benefits to organisations which implement it, it is still the stick of legal action that appears to be hitting home the most.

Despite the increase in accessibility awareness, misunderstanding and misapplication still appear to be a big issue. The main evidence for this is the still woeful general level of accessibility. For example, in the UK in 2004 a formal investigation of 1000 sites by the Disability Rights Commission revealed that 81% of UK sites failed even the most basic of accessibility checks [3]. The picture is likely to be similar, if not worse, in most other countries. You can test this easily yourself by using a text browser such as lynx, just surf a typical selection of sites and see how you get on. In addition, from personal experience I still frequently encounter remarks such as “it’s OK we have text only”, “better accessibility will just make it boring though”, “we will make it accessible afterwards”.

So some progress then, but still a long way to go.

Some leading lights

Some organisations have produced sites that stand as examples of what you can achieve as long as you tackle the issue seriously and make accessibility an inherent part of the development or redevelopment process. One thing about the best of these sites though is that unlike five years ago, where you could usually spot an accessible site a mile off by the way it looked, these sites have not compromised on brand or aesthetics in order to deliver exemplary accessibility.

My favourite best practice web accessibility example is the UK local search engine site “Yell.com” (see screenshot, Figure 1 - [2]). All pages exceed the double-A accessibility level with many achieving triple-A compliance. Importantly though Yell.com commissioned user testing with a broad range of participants, including users of adaptive technology such as screen readers. The findings from the user testing were used to improve Yell.com’s practical accessibility. So not only does it conform to technical accessibility standards, the site works in practice too - and this is an important distinction.

The best sites do more than just tick the boxes to get a badge of conformance. Inclusive user testing like that performed by Yell.com is the key to this. Not all sites that achieve technical accessibility compliance are actually usable by people with accessibility needs. I have seen triple-A (the highest accessibility conformance level) conformant sites that were practically unusable by anyone, but particularly so for those using adaptive technologies such as screen readers and screen magnifiers.

Conversely some sites that do not achieve technical conformance are actually very accessible in practice. An example of this is evidenced from the above mentioned DRC accessibility investigation [3]. Five of the 1000 sites tested received special praise for being highly accessible as tested by a review panel of users with disabilities. One of these sites was an online only bank, egg [4], that actually failed the technical compliance check. Those responsible for the egg.com site knew of the technical failures which were caused by legacy back end systems but had prioritised fixes to the most important issues in order to make the site accessible in practice. Further, the real accessibility had been checked through user testing.

In my experience the approach that egg took is quite rare; the majority of site owners are more concerned with conformance than reality of use. This is understandable when conformance itself is the most “accessible” accessibility benchmark a site can have. Alas there is still currently a problem with even the notion of conformance.

The difficulty organisations face over knowing what to do

There is a plethora of information available that pertains to web accessibility. Some of it is very technical and not easy for non-techies to evaluate, much of it is unofficial and difficult for non-specialists to determine the validity of the advice and techniques. Organisations have difficulty knowing what to do, even if they are highly motivated in addressing the accessibility of their sites.

The most “official” global web accessibility guidelines are the Web Content Accessibility Guidelines version 1.0 (WCAG 1.0). These guidelines were produced by the Web Accessibility Initiative (WAI - [5]) part of the World Wide Web Consortium (W3C). The problem with these guidelines is that they are out of date (1999), are long and unwieldy and can be difficult to understand and interpret. Furthermore they are difficult to test against objectively - even so called accessibility experts disagree on some of the finer points, sometimes heatedly. Most other current standards are based heavily on WCAG 1.0.

Some potential solutions

In the absence of any clear guidance some large organisations have undertaken to produce their own best practice standards and processes for ensuring web accessibility. For those with which I have been involved they have lead to clear long term accessibility improvements, but have required a significant investment of time and money. Unfortunately most organisations are unable to support this level of investment.

The update to the WAI standard, WCAG 2.0, is in draft form and indeed has been for a number of years. The intention for this version is that it should increase the clarity and ease of application of the accessibility guidance, and make the job of conformance testing easier and more objective. Unfortunately, it is still unclear when WCAG 2.0 will be finally published as an official recommendation, and indeed how much of an improvement it will prove to be for organisations wishing to seriously tackle accessibility at a practical level.

In the UK the Disability Rights Commission has commissioned the British Standard Institute to produce a new guidance document aimed at informing website commissioners and developers of their obligations and of good practice for web accessibility [6]. This guidance takes the form of a Publicly Available Specification (PAS). A PAS is not a full British Standard but is developed using the same rigorous processes. The advantage of a PAS is that it can be introduced more quickly than a British Standard and that it can be updated frequently. The PAS is due to be published in autumn 2005 and will be updated approximately every 2 years. It is certainly hoped the PAS will provide organisations with a source of authoritative, practical and up to date guidance.

A new European initiative has begun developing a European Quality Mark for Web Accessibility. The Quality Mark will be awarded by accredited assessment bodies that will use a common testing methodology which will be based on assessments against current WAI guidelines. One of the key aims is to reduce possible fragmentation that may be introduced if many different localised standards begin to emerge. If successful this is likely to stimulate the take up of web accessibility, and also simplify the task of web site owners in ensuring accessibility conformance.

I would certainly hope that some if not all of these authoritative and soon to be publicly available standards will actually start to bring about real improvements in web accessibility. Accessibility can be a complex business but there are often clear and unambiguous solutions - as long as people know what to do.

The next stage - great design through accessibility

I believe that accessibility from a technical perspective will relatively soon cease to become much of an issue. There are now plenty of talented designers and coders who have the skills to implement web pages that are well coded and conform to standards. Hopefully this will become the norm rather than the exception.

If my company ceases to be needed to provide advice on these technical issues I will be very happy. I for one would much rather devote time and energy to ensuring great design and great usability rather than spotting trivial coding omissions. The next challenge is to stretch our understanding of what constitutes great design. Accessibility considerations can help here.

Testing an interface with people with different perceptual needs can reveal key insights that can bring about better design. Testing a site with someone who relies on screen magnification, a screen reader or who has dyslexia reveals usability issues that may be obscured by more typical users. Fixing the issues typically leads an increase in effectiveness and efficiency for all. In my experience “accessibility user testing” is a sharp usability tool.

Conclusion

Accessibility has improved over the last five years, but not nearly enough. With any luck technical accessibility will soon cease to be such a big an issue as it is currently. Up to date authoritative accessibility guidance materials and standards will help with this, and they look like coming fairly soon.

In five years time I hope to be reporting that people with accessibility needs are typically considered at all stages in the production cycle, and routinely involved as participants of website usability testing. I hope we will all be more focussed on great inclusive design rather than the trivialities of things such as missing alt text.

Reference links

1. http://www.rnib.org.uk - select “good design” for information on their campaign for more accessible design.
2. http://www.yell.com - a great example of maintaining brand and great usability whilst delivering very high accessibility.
3. http://www.drc-gb.org/publicationsandreports/report.asp - the UK Disability Rights Commission formal investigation into the accessibility of 1000 sites.
4. http://www.egg.com - an on-line bank
5. http://www.w3.org/TR/WAI-WEBCONTENT - Web Content Accessibility Guidelines 1.0
6. http://www.bsi-global.com/PSS/press_releases.xalter/pas78.xalter - press release announcing the UK Publicly Available Specification (PAS) for web accessibility in the UK.