Reflections on Web Adaptability and Techshare 2009

Last week I gave a talk entitled “From Web Accessibility To Web Adaptability” at the RNIB’s Techshare 2009 conference. I have already posted about this talk and described how I had created a slidecast of a rehearsal of the talk (containing an audio track synched with the slides) in order to (a) check the timings for the talk and (b) allow the co-authors of the paper on which the talk is based to see how I intend to present our work. An additional benefit is that the talk is more accessible to people who attended one of the parallel sessions at the conference or who couldn’t attend the conference. In addition people who could attend the talk will be able to revisit the ideas and share them with colleagues.

In addition to the slidecast of the rehearsal I also brought a Flip video recorder with me, together with a tripod and recorded my live talk. This 30 minute talk is now available on Vimeo.com (and a master copy is also held on the UKOLN Web site).

It should be noted that there are some differences between the rehearsal and the live talk. In part this is due to the delayed start of the talk (due to technical difficulties) which meant I had to skip a couple of my slides. But in addition on the evening before the conference I met up with a number of conference participants, including Lisa Herrod (one of the co-authors of the paper) and Joshue O Connor, who is a member of the W3C WAI Protocol and Formats WCAG 2.0 and WAI-ARIA Working Group.

The chat I had with Joshue provided me with a fresh insight of my criticisms of the WAI model. I’ve argued previously (initially in a paper on “Forcing Standardization or Accommodating Diversity? A Framework for Applying the WCAG in the Real World” published in 2005) that expecting a combination of best practices for accessible Web content (WCAG), Web authoring tools (ATAG) and Web user agents (UAAG) to provide rich accessibility is naive. And, in addition, focussing on this model fails to provide any assistance on what content creators should be doing in a world of flawed browsers and a rich diversity of ways of creating Web content.

The valuable discussion I had made me realise that the flaws aren’t in the model itself. Rather it’s with the user community’s acceptance of the model as the approach which should be accepted in the real world. The WAI model is valuable in managing WAI’s development activities and clarifying different areas of responsibilities (how the content can be described; how tools can be used to create and manage that content and how user agents – browsers, automated agents; aggregators, etc. can then access and render such information). But this isn’t a model which we need to use ourselves when we are developing institutional policies for our approaches to enhancing the accessibility and usability of our services or when legislators are writing laws describing the legal responsibilities organisations have in providing accessible services.

Following my talk, Joshue and I had a brief chat. Despite the concerns I’d raised it seems that we had similar views. The difficulties, I feel, is in how the WAI approach is being adopted in the real world. So whilst I appreciate WAI’s advocacy in promoting take-up of their guidelines, I now have a better appreciation that their hands are tied when it comes to real world deployment challenges. WAI aren’t in a position to advise on how we should prioritise our (increasingly scarce) resources – such as the example I gave in my final slide on how higher educational institutions should go about enhancing the accessibility of PDFs in institutional repositories.

But perhaps WAI could help by openly stating that decisions on how WAI guidelines should be deployed is up to individual organisations to decide. We do need to remember that there are ‘accessibility fundamentalists’ who bought wholesale into the WCAG 1.0 vision and who may now be finding it difficult to come to terms with a more flexible approach. Let’s use the release of WCAG 2.0 to promote a more flexible approach to accessibility in the real world. And let’s also not forget that the UK Government’s blunt approach of “The minimum standard of accessibility for all public sector websites is Level Double-A … Websites owned by central government departments must be Double-A conformant by December 2009” . This policy fails to recognise the low penetration of UAAG-conformant browsers in the Government sector, the resources needed to implement this policy, the reduced level of funding which government departments will be faced with and the likelihood that risk-averse decisions-makers in government departments will use the policy as an excuse to deploy innovative Web-based services.

The slidecast and video of my talk at Techshare 2009 gives another illustration of how providing a diversity of resources might enhance the accessibility of a resource (my talk and the related ideas) which is, to my mind, preferable to not making these resources available as they aren’t universally accessible. And this view appeared to be shared by a number of people at the conference who couldn’t attend my talk but were interested in listening to what I had said.