Accessibility: carrots or sticks?

I went along to the Disability Rights Commission's launch of the report The Web: Access and Inclusion for Disabled People this week. Billed as the largest ever survey of websites, the DRC's research - in co-operation with Helen Petrie at City University - included a survey of 1,000 UK home pages, 100 more in-depth website studies, interviews with stakeholder groups, a questionnaire, a comparison of accessiblity and usability, and an assessment of results provided by automated testing tools.

The results were not althogether surprising: only 19% of website home pages (192) complied with Web Accessibility Initiative (WAI) Guidelines 1.0 level A, only 0.6% (6) complied with WAI level double-A, and none with triple-A. Government websites fared somewhat better than the private sector, with 32% of home pages level A compliant.

Blind users were found to be the group most disadvantaged by inaccessible websites. For most disabled groups - using assistive technologies - the main effect was the long time taken to complete tasks, rather than failure to complete. The major problem reported by most groups was 'unclear and confusing page layout of pages', with 'confusing and disorienting navigation mechanisms' also making a significant appearance among the main barriers identified.

The DRC also reported that automated testing tools only found around 55% of real user problems. In other words, almost half the real problems reported would not have been uncovered through use of a testing tool. Reliance on tools, or a tick-box approach to accessibility, will not address many real-world accessibility problems, it concludes.

Survey questionnaires revealed that while there is a reasonable awareness of accessibility among website commissioners, especially among larger firms. But taking accessibility into account when developing websites, and conducting testing, both lagged behind awareness. Firms reported a low level of knowledge about how to implement accessibility measures. Website development agencies also reported low levels of in-house expertise in this field, with a high reliance placed on automated testing tools (only 9% had included disabled users in user testing).

The DRC has come out with recommedations that include a kite mark system, and "accreditation and registration of Web developers" (in the words of DRC chairman Bert Massie), and in this I believe it is jumping the gun. In any case, if guideline tick boxes are not the answer in making websites accessible, why should developer tick boxes be expected to do the trick?

Developers certainly show some awareness and a willingness to work towards accessibility. What seems to be lacking is turning that willingness into practical skills.

According to the report, UK government guidelines fared poorly when it came to getting developers up to speed: 5% of developers said they had used them, compared to the 65% who reported referring to WAI Guidelines.

While the research did not assess the level of training in this area provided in educational institutions, it's clear from reviewing today's course programmes that few courses at any level or in any specialist area - digital media, multimedia, computing, Internet managment, e-commerce etc. - incorporate accessibility concerns.

It's small, voluntary efforts, such as Ian Lloyd's Accessify, that are perhaps most effective at present in spreading the word and indicating practical approaches.

Without usable guidelines and adequate training, and without corporate policies and practices that support the acquisition and application of that knowledge, it's difficult to see how developers are to achieve what's now being expected of them. Meanwhile user testing, while necessary, will only be productive when developers have the knowledge to make the best use of findings. Let's see some reasonable carrots put into place before hitting developers over the head with regulatory sticks.