Don’t break Fix the Web

This week, Fix the Web was launched, and attracted some excellent publicity. It’s an initiative that is a potentially significant step towards empowering people who experience accessibility barriers on line to report them in a way that might increase the chances of someone fixing the problem. One great difficulty in evaluation and repair of the sheer mass of web content we have today is helping people who experience accessibility barriers identify them as such and tell someone who might be able to make a difference.

Not all disabled web users are web developers; it’s unlikely many people will have heard of WCAG, let alone be able to describe a problem with a web page with reference to WCAG success criteria. But Fix the Web gives a voice to someone with a disability to say “hey, I couldn’t add an item to my shopping cart”, and have that problem passed on to volunteer accessibility specialists who can investigate, and back up that real-world problem with a technical description of the underlying code or design defect. And the more people who identify the same problem, the louder the voice.

While Fix the Web may not be able to actually repair the page (though other crowd-sourcing approaches have taken this step, such as IBM’s Social Accessibility project ), it does give web site owners evidence of real customers being unable to use the site for its intended purpose AND informed technical information that might help them fix it.

However, there’s a problem. As pointed out on Twitter today by Eric Eggert (@yatil), Fix the Web ‘s own web site has accessibility issues. This, it’s argued, may lead to a loss of trust by the very people it’s meant to serve – disabled people who think “well, if Fix the Web can’t make their own site accessible, how can they tell other people to fix the problem I report?” This would indeed be a huge problem, if the only way to report barriers was through the site, and some people were prevented from doing so.

But it’s not. The great challenge of Fix the Web is reaching out to people who experience accessibility problems, who have never heard of Fix the Web, never heard of WCAG and who might never have considered reporting barriers. Installing browser toolbars in public access PCs is one approach, encouraging people to email or tweet details of problems is another.

The accessibility of the Fix the Web site should be as good as possible – of course it should, and it needs to be, soon. However, the credibility and viability of the initiative should be based on the quality of the reporting system – the awareness it raises amongst disabled web users, the skills of the volunteer accessibility specialists in following up problems, and the quality of the communication with web site owners.

Casting aspersions on the whole system based on the accessibility of its own site – which is, after all, an intermediary in a wider communication chain – is a waste of energy, particularly by accessibility advocates. Let’s not break Fix the Web before it has a chance to make a difference.

User-centred Design where it’s needed most

On the same day this issue was being discussed on Twitter, I attended a presentation by Simon Judge, an augmentative and alternative communication (AAC) specialist working at Barnsley Hospital in northern England. I knew Simon’s name from various open source accessibility initiatives such as OATSoft and MAAVIS; he was in Dundee as a guest of my colleague Dr Annalu Waller, who leads a research group on AAC.  While a busy day meant I didn’t have a chance to speak to him, it was interesting to hear his work on exploring AAC user needs and preferences (and enjoy some vintage 70s TV – Toby Churchill demonstrating assistive technology on Tomorrow’s World!)

In listening to Simon’s talk about work on the Devices for Dignity project, I was again struck by the relationship of accessibility and communication, and why we haven’t quite got things right yet.

It’s impossible to overestimate the positive impact AAC systems can have on people with severe communication difficulties. Yet, for the importance of such systems,  there have been  long term problems with AAC systems relating to their usability and appropriateness for the person whom they are supposed to be helping.  The problem seems to stem from the medical context in which they’re usually given to people – without real appreciation of the environment or personal contexts in which they might be used. Unlike choosing, say, a mobile phone, AAC users are likely to be limited in their knowledge of what choices are available to them, and normally there will be an intermediary (AAC professional) involved in the selection process. Practical issues such as reliability and durability can reduce trust in a device that ultimately is the difference between being able to communicate and not.

Simon has spent time speaking to AAC users and professionals in order to understand the similarities and differences in their perspectives; generally users are more negative in their views. So clearly there are issues in usability of such technologies. On the surface, the apparent limited use of user-centred design by the AAC and assistive technology field might seem surprising, but traditional UCD techniques can be difficult to apply in the field of AAC design particularly since access to users early in the design stage might be challenging.

This is an ongoing area of study, but work like Simon’s, and that of Annalu, Graham Pullin and colleagues here in Dundee, will hopefully lead to a more user-centred, more context-aware design process in creating genuinely usable and useful assistive technology.

Web accessibility advocates may recognise parallels with screen reading technology too – so be forgiving of people who are not intimately familiar with their AT, and be conservative in your assumptions of the role a user’s AT plays in enhancing the quality of their web interactions.

As a web developer, there’s little I dislike more than building sites to be accessible.

That’s some confession. Now, cards on the table – I don’t know Drew, though I do appreciate his efforts in promoting web standards adoption. And I’m not here to suggest he is arguing against accessible design – on the contrary, from his blog it’s something that he feels obliged to do, so he does it:

So as much as I find it an unpleasant chore, I’m firmly committed to building sites that can be accessible as I can make them.

What I do take from the blog, though, is that this is an example borne of frustration that accessibility considerations are stifling innovation in web application design and development. As I’m not yet familiar with the technologies discussed in the blog, I’ll leave the details of the argument surrounding those technologies to others.  I will say that much of Drew’s ‘accessibility’ objections seem to focus on the avoidance of dependency on JavaScript – which, since publication of WCAG 2, is no longer considered a disability accessibility showstopper (in other words, the very use of JavaScript does not immediately exclude a particular group of users on account of a disability).

I wrote a paper a couple of years ago that commented on how important it was for accessibility that the Web Standards movement embraced WCAG, after accessibility advocates argued that standards-based web design naturally encompassed accessible design. But are we now seeing signs of the same movement become tired of dealing with the human messiness that is accessibility?

Accessibility is no longer seen as purely an exercise in WCAG validation and a nice technical string to a developer’s bow, where all user needs are met in their entirety. Now we know that assistive technologies don’t always take advantage of HTML’s accessibility support in the way we might expect; there is a heated debate over whether current HTML attributes aimed at supporting accessibility must have evidence of usefulness for their inclusion in HTML 5, and whether non-use or abuse means rejection from the spec., regardless of potential usefulness.

And we know that users themselves are non-conformant, in that they cannot be assumed to have the appropriate accessibility solutions available for their needs. To me, the problem is this. It’s quite possible to legitimately argue that old browsers that poorly support standards are holding back evolution of the web. It’s also probably legitimate to say that the assistive technology industry has not yet met its side of the bargain in terms of supporting standards (judging by comments screen reader developers are not participating in the HTML 5 specification process). But for people who have no choice in the matter in terms of the assistive technology they use, is it fair to accuse them of holding back innovation?

I accept that accessible design can in some cases be challenging, frustrating and time consuming, and this is all done under a – occasionally unreasonably applied – pressure to ‘do the right thing’. But the danger of comments from respected figures in the web standards movement that accessibility is a drag on progress may be interpreted by those who are less informed as evidence that accessibility is something that’s no longer worth striving for. Railing at the shortcomings of screen readers doesn’t help people who have to use them. I don’t know the politics of the apparent lack of engagement of screen reader vendors in HTML 5 and other Web standards projects. But suggestions that accessibility advocates work on educating AT vendors instead of encouraging content providers to think about accessibility could easily be distorted and interpreted as the sort of attitude to disability the appalling Daily Mail stirs up amongst the political-correctness-gone-mad brigade.

Let’s not try to blame accessibility on unreasonably holding us back. Let’s look instead to innovation through inclusive design – the Ford Focus, Oxo Good Grips and several other examples of where taking into account ‘extraordinary users’ has led to a better user experience for all. Can WAI-ARIA lead to more examples being added to that list?

But, when I said disabled web users had ‘no choice’ above, of course that’s not quite true. For those who need a dedicated assistive technology, there increasingly is a choice, especially in the number of quality open source alternatives emerging. The problem is that lack of awareness of what’s available usually leads to a reduced choice of the most well-known (and ironically expensive) commercial products, and is something that activists like OATS and the guys behind AccessApps are trying to address.

So, just as increasing rejection of Internet Explorer in favour of other browsers has helped drive standards-based development, perhaps the same can happen in the assistive technology scene. Poorly implemented AT is a problem, but ATs that take advantage of web standards, and increased awareness of their existence amongst potential users, can help ensure that accessibility doesn’t take the blame for holding back innovation on the web.