Accessibility: Panning for gold

I've been meaning to write about my current thoughts on accessibility for a while now and an email I received on the subject spurred me on.

I've been meaning to write more about my current thoughts on accessibility for a while now. In the last week I received (among many) some interesting emails on the subject and this has spurred me into writing. Now I'm not usually in the habit of making public any private correspondance, but these messages made me think about what I have been calling privately, a mature approach to accessibility.

One email was from a designer whose progress I have watched with admiration over the last few years. It read,

Hi Andy, just a quick email to say how much I like the new site. A few questions if I may? I see the Malarkey website does not pass the Online Bobby AAA test and that you do not mention the Web Accessibility Initiative and Bobby AAA accessibility rating at the footer. I'm just curious, is this for a reason or has it slipped the net so to speak?

This email made me consider the issues he raised carefully, and although there are two questions, he makes three points relating to Bobby and ratings.

Tools not standards

We have many conversations with potential clients about accessibility and many requests for proposals where Bobby compliance is required. While I am genuinely happy that accessibility is being raised, I often find myself correcting the misconception that Bobby and similar tools are standards. Like Photoshop, Dreamweaver or Visual Studio, these tools are simply developed to be part of a web design tool-kit, they are not standards.

Accessibility, a part of the process

I hope that it is well known by now my view that web accessibility should not be an issue, but should be a fundamental part of the design process, equal to, but no greater in importance than usability, good copy writing or visual appeal. Designing and developing any product should be a careful balancing act, and tackling any project with a holistic approach should balance the needs of all concerned. Including,

I also believe that a mature approach to accessibility is about the decisions which we make within the context in which we make them. Sometimes these decisions are more straight-forward (what are appropriate alt attributes or should we use empty alt strings (alt="")?). Other times these decisions are more complicated. Often what is misunderstood is that in accessibility there are few absolutes. Disability is relative and therefore by definition, accessibility is also relative. What matters are the judgements that we make within the context of balancing the needs of all concerned.

What concerned me about the email was that this person equated accessibility with Bobby and with ratings, rather than seeing accessibility as an integral part of a whole process. I am finding that this is not an uncommon view.

Another site from a local firm proudly proclaims,

We are developing websites with full Bobby (Watchfire) compliance.

And an unsolicited email offered me a free online Bobby check ;) accompanied by,

To see if your site is up to the current accessibility standards, type in the URL of your site into the box below and submit it to the Watchfire analyzer.

Panning for gold

I am concerned here about two issues. The first is that I am seeing a trend where designers see accessibility as a matter for compliance rather than as a series of judgements to be taken in the spirit of providing all parties with the best possible solution. I fear that this may promote a negative view of accessibility where it becomes seen as a limitation rather than an opportunity.

Secondly I fear that compliance (particularly in relation to the misguided view that Bobby is a standard, ) is becoming a dubious sales tool for those panning for gold in the Accessibility Klondike.

Back to the original question

So back to the original questions posed in the email.

...I see the Malarkey website does not pass the Online Bobby AAA test and that you do not mention the Web Accessibility Initiative and Bobby AAA accessibility rating at the footer...

We used Bobby to help us spot issues that might otherwise have been overlooked during a reasonably fast build. Do I care that about Bobby's rating of the site? No. I care about the needs of my customers and others visiting the site, and of course I care about my business.

Did I forget to mention WAI, Bobby or AAA? No. People with disabilities know when a site is better accessible by using it, they don't need telling. I have written before about badges and I don't need any more to pin on my parka ;)

But my real issue here is not simply whether or not to display badges, I've been there before. My problem is with the way that compliance is now often both perceived and used. And here I don't yet have a simple answer.

Over to you.


  1. #1 On November 30, 2004 01:13 AM brothercake said:

    I can't add anything really - I agree completely.

    Perversely, Bobby badges are increasingly becoming marks of ignorance - you know which companies to avoid, because they're the ones who tout Bobby compliance.

  2. #2 On November 30, 2004 01:39 AM YoungHistorians said:

    I guess it's become the equivalent of Alexa ratings for websites, or even the Golden Web Awards?

    Just shake your head and look away.

  3. #3 On November 30, 2004 02:03 AM Jonathan Snook said:

    I'm the same way when it comes to site 'badges'. In fact, I use none. No mention of XHTML or CSS or AAA or other TLA's in my left/right/top nav or footer. I've left that information for my About page.

    And you're also bang on about looking at Bobby/Watchfire as just a tool. It can help lead us onto a more enlightened path but it isn't the destination in itself.

  4. #4 On November 30, 2004 02:03 AM Jules said:

    Good article and for the most part, I can't argue with it. However, if you need to get an oil change for your car and one business advertises "$20 oil change" and another one advertises "$20 oil change, light inspection, fluid inspection, and belt check", which one are you going to go to? Pretty obvious decision isn't it except that the other business offers all the same services and options but, for whatever reason, doesn't advertise it: their loss.

    Badges may not mean much to those of us in the know but for others who are looking at our web sites to see if we can create an accessible site for them, the badge may be the foot in the door to an enquiry about a contract. The reverse also is true: not having a badge may be enough for a potential client to move past our Web site to another. Furthermore, if the badge, as many are, is linked to the validation site, then, whether or not the potential client can fully appreciate the details of the tests, if the potential client can see a passing grade against a battery of tests, again, it may be another reason for them to contact you.

    Those who know, don't care about the badges or the accuracy of the so-called tests: those who don't know, have no other means to evaluate claims of accessibility if either a badge or the results of a test aren't available to them.

    It is my opinion that because accessibility tests cannot be fully automated, like HTML and CSS validations can be, accessibility badges are a damned-if-you-do, damned-if-you-don't scenario. It would be much simpler scenario if accessibility tests could be automated but that is not, and likely will not ever be, possible.

  5. #5 On November 30, 2004 09:16 AM Tim Parkin said:

    Perhaps we should recruit a set of volunteer users who can provide *real* tests of a sites accessibility. Then we can add 'Duncan Smithfield' approved and 'Dorothy Whitehead tested' etc. Who could argue with thowing a little money directly into the disabled community and soliciting their feedback at the same time? Perhaps a website could be set up to match users up with companies who wish to make use of their services in assessing their websites. It does seem that these users are the real people who are being targeted and seem to never really get mentioned, let alone included, in the process.

  6. #6 On November 30, 2004 09:46 AM Seb Duggan said:

    brothercake is right - the onese who tout their Bobby compliance are usually the worst:

    " has a new text based film times service available. The text based times can be viewed from a variety of browsers and has been designed for use with Bobby and Wave accessibility tools."

    And this is the most god-awful site I;'ve ever seen. Unfortunately, since last time I visited they've taken down the link to the web design company who did their site, who "specialise" in making sites accessible; a pity, because their site was almost as bad...

  7. #7 On November 30, 2004 09:58 AM Matt Lindop said:

    I love Tim's idea of a service to match up skilled disabled users with companies who need their input. I remember Julie Howell saying years ago that we should regard testers as consultants and pay them as such. The challenge for us as developers is to find the right people (and not just people with visual impairments).

    The beginnings of a new community project? Blind Eye for the Geek Guy?!!

  8. #8 On November 30, 2004 09:59 AM Robert Wellock said:

    Fair enough, how much do you want to pay me to review your sites then; I have what is considered an official disability that does affect how I myself interpret data.

    Precisely I don't charge because I am part of the Horse's Mouth itself not a checklist, robot, a second-hand opinion or pseudo-interpretations devised by some web-development outfits.

  9. #9 On November 30, 2004 10:24 AM Patrick said:

    The problem lies not only with web authors, but with the "standards" bodies. Bobby shouldn't offer a badge of compliance, in my opinion, but even worse is the use of WAI badges. Like you say, these are guides and tools, not rules and there is no black-and-white like there is in XHTML validation, for example. I don't claim "AAA" because a lot of it is bollocks and the _guidelines_ are vague. And I don't take any notice of any other's claims of "AAA".

  10. #10 On November 30, 2004 02:08 PM Rob McMichael said:

    I'm not sure I agree with you guys here.
    Firstly to get the web more accessible the idea needs to be promoted as much as possible. The more badges that are seen but company heads etc the more they will request them when they commission sites. If at this point you want to guide and inform them of better practice then so be it. (after all I first got into standards through curiosity over badges)

    I also feel that the guidelines checked by bobby (WAI and 508) are necessary in creating accessible websites, take the naming of frames of alt text requirements for example. They are required to get a AAA cert, they cant check the value of the text, after all it may be a * or just blank but without it you have an inaccessible site. Therefore why not get the full cert and then add to it with your knowledge.

    It is however a problem as I'm sure you could make an inaccessible site which validate in the guidelines and vice versa.

  11. #11 On November 30, 2004 02:30 PM Tommy said:

    "The more badges that are seen but company heads etc the more they will request them when they commission sites."

    The problem with this is that the badges have no correlation to the actual accessibility of the site. Badges will only encourage badge hunters; people who don't give a d**n about accessibility, but who like to collect pretty badges. In the long run, this will give accessibility a bad name.

  12. #12 On November 30, 2004 03:26 PM Rob McMichael said:

    Opss... Double post (blame NTL)
    (Ed: Don't worry Rob, I'll tidy around after you ;) )

    I agree with what you are saying, but the most inaccessible site which gets a bobby AAA rating is better than the worst site without. I it isn't likely that they are going to purposely fill in all the ALT text with blank values...

  13. #13 On November 30, 2004 03:42 PM Malarkey said:

    Is Flash in itself bad? I don't think so, it's what designers do with it that can make for poor sites. Is Bobby bad? No I don't think so either. But my concern goes wider than badges and ratings.

    I think that the issues highlighted in my examples illustrate that many designers think that Bobby is accessibility and think that because a site 'passes Bobby' that is therefore accessible, job done...

    This mis-guided view is not solely held by designers. It is often held by clients in public sector and private sector organisations. This is great news for Watchfire who are rapidly becoming the Hoover of accessibility, but poor news for the implementation of mature accessibility and the people who matter, the users.

  14. #14 On November 30, 2004 04:23 PM Rob McMichael said:

    So rather than saying that you disagree with Bobby or the fact it rates sites on build rather than semantics, you disagree with people who will manipulate the ratings. I think we can all agree on this and the fact that bobby and other checkers of accessibility, check the markup rather than the content. Just as a CSS valuator will make sure you have valid CSS but not good design.

    This is something I have known about for a while, but haven't really considered as much of a problem as perhaps it is. I guess there will always be people will bend the rules or just do botch jobs. I mean the Labor party site has the bobby badge and it doesn't even pass... On reporting this to bobby they didn't seem to care:

    "Hello Rob,

    Thanks for contacting Watchfire support.

    Unfortunately Watchfire does not actively monitor a client's use of the "Bobby Approved" logo on their web page to ensure they meet the stated compliance. It is ultimately the responsibility of the web administrator for a given web page to ensure their site continues to meet compliance."


  15. #15 On November 30, 2004 06:18 PM Roger Johansson said:

    Agreed about the automated tools. They can be useful, but can't be used as a substitute for manual checks.

    I'm also seeing more and more less than fully accessible (to put it mildly) sites claiming to follow WAI guidelines. In most cases, I believe they are doing so out of ignorance - the developers don't understand what accessibility is.

    And while we're on the subject of accessibility: ever considered running your text and background colours through the Colour Contrast Analyser, Andy? ;-)

  16. #16 On November 30, 2004 08:14 PM Malarkey said:

    @ Roger:

    Cheeky bugger! ;)

    In all seriousness though, Jez's colour/contrast tools are very useful indeed and we do use them on a regular basis. I'm hoping that if 'precious' designers like me use low contrast colour schemes, that Invasion of the Body Switchers will provide a useful tool in offering a high contrast alternative as we have done on the new Stuff and Nonsense site.

  17. #17 On November 30, 2004 08:21 PM Mike Pepper said:

    Hmm. Badges again, eh.

    I badge and will continue to do so on most every page of my sites until we reach critical mass and 'standards'-compliant accessible development becomes the norm.

    'Woz all this, then?' Click ... Hmm, that's cool. Gonna get me some of that. I explored standards after hitting a compliance level button.

    Web design can be a confusing and daunting rite of passage, akin to being dumped at the centre of a complex maze with numerous dead ends, some of which take many days or even months of fumbling before the futility of a particular development cul-de-sac is realised.

    I want to encourage wannabe developers to set off on the right foot because it was only once I'd discovered there were indeed guidelines available that I was able to settle into the business. I found a framework within which to nurture an understanding of development best practice.

    Besides, W3C badges are really no different to blog referrals: they're links to recommended reading and until we develop formal standards, W3C guidelines are the best we've got.

  18. #18 On December 1, 2004 09:40 AM Small Paul said:

    The WAI Guidelines sure are vague. And sometimes, seemingly self-contradictory. Hopefully version 2 will tighten things up.

    The tricky thing is that businesses are often going to want badges. I know it's a bit depressing, but businesses can and do get taken to court if they don't comply with accessibility legislation. Compliance-style behaviour will always exist when there are bottom lines like this, because compliance is measurable, whereas careful balancing acts are more difficult to measure.

    I entirely agree that the benefits of following accessibility guidelines (i.e. your site will be better! Genuinely more useful for everyone!) should be emphasised, but clear standards with attendant badges are an important tool as well - it's the only tway to get uninterested businesses to play along.

  19. #19 On December 1, 2004 11:45 AM Robert Wellock said:

    I'd agree half-way with Mike I had a plethora of webmasters and more normal users give me positive feedback of the display of labels like ICRA and valid mark-up icons; though I don't collect them.

    When I was analysing a real-life user today whom has SPLd surf the web she mentioned that some colour contrasts were easier on the eye. Albeit she actually found it counterproductive that some sites allowed her to change the contrasts and colour leading to more confusion.

    I'd agree the contrast on this site is hardly that strong razing the question; do two horse's mouths speaker louder than one. If the horse's mouth could speak, his days would turn to weeks and you'd be bored with all that you know.

  20. #20 On December 1, 2004 01:33 PM Malarkey said:

    If the horse's mouth could speak, his days would turn to weeks and you'd be bored with all that you know.

    Errr, you've lost me there Robert. ;)

  21. #21 On December 1, 2004 03:17 PM Andy Budd said:

    There used to be a usability "rule" I'd see quoted in web design magazines all the time which basically said that no page on a site should be more than 3 clicks away from another page. This stemmed from a fairly sensible usability heuristic which about site navigation which required a certain degree of expert knowledge and professional judgement.

    However to help the layman this heuristic was boiled down into an oversimplified rule which, while it may have been true for 90% of sites, definitely wasn't true for all sites. Because of it's simplicity it caught on and I had several clients telling me that their sites needed to be no more than 3 click deep.

    I think the same is true of things like the WCAG and Bobby badges. Web accessibility is a complicated thing that requires a certain amount of professional judgement. A bobby or WCAG badge however is a very simple thing that clients can get their heads round and request off web designers.

    In some ways I think this is a good thing. For professional web designers like ourselves who understand the issues, it can be a little frustrating. However there are a lot of web designers out there who know nothing about accessibility so it's good that their clients can ask for accessible sites in a language they feel comfortable with.

    The problem is that like usability, web accessibility isn't a black or white issue. However for clients who's main jobs aren't web design, they don't have the time or inclination to learn about there issues, so need an easy way of communicating their needs to their developers and knowing that their developers have done a good job.

  22. #22 On December 1, 2004 06:57 PM Robert Wellock said:

    Since I have undivided attention I'll explain the above; within the passage you mentioned:

    People with disabilities know when a site is better accessible by using it, they don't need telling.

    I assume you meant 'more' accessible instead of 'better' or even 'aptly' don't quote me on that my command of English grammar is shaky.

    I then pointed out what the "Horse" was, or should I say what basically it represents. Yes, perhaps it was a little abstract but the point I was razing referred to: "How many actual real-life disabled users do you test your web designs with?"

    Since it appeared that you were inferring that's what you did during the development and reviewing phase. If this isn't the case on what premise can you possibly know that the site is accessible. I mean you could ask the so-called guru Joe Clark but he isn't disabled so all he could give you a straight answer but some recommendations from his past studies.

    I cannot give you the answer either because I am not the only "horse". Basically what I was asking for was more clarity on how you performed 'real-world testing with such user-groups'.

    As I said today, one met someone who really was not impressed by various customisable JavaScript Style changer I was showing her today as I mentioned she has a common form of SPLd.

    Don't worry I am not being argumentative I am just trying to make sense of the facts presented.

  23. #23 On December 1, 2004 08:45 PM Malarkey said:

    #22 Robert: I cannot give you the answer either because I am not the only horse

    You hit the nail on the head here. and reflects what I said way back in the original column,

    Disability is relative and therefore by definition, accessibility is also relative. What matters are the judgements that we make within the context of balancing the needs of all concerned.

    The question you raised about people with disabilities being involved in testing is an important one. It can definately be a problem finding such testing resources and whilst we don't 'test' every design with a wide range of users with disabilities, our experience over the years of working with people with a wide range of disabilities gives us some insights which we carry forward into every project. We then look at the target audience for a site and invoke the relevant testers.

    Could it be that after all this time, you and I finally agree on something? ;)

  24. #24 On December 2, 2004 05:05 PM Grant Broome said:

    Badges, accessibility, disability user testing.
    A couple of points, if I may:

    Badges are percieved to be of little worth by "us" because they are normally based on some form of self-assessment. We know that. Clients however still seem to think there's some sort of magic in having a Bobby logo, and, as frustrating as it is, it is still perceived as a measure of accessibilty.
    Damn you Bobby!
    And here's the double edged sword...
    Bobby helped to raise awareness of web accessibility by creating an interest in, and beginning to inform visitors of the potential issues.
    "Luke, I am your father"

    It seems that the Shaw Trust Web Accreditation service is in demand, but not well enough known.
    You can have your site tested by people with varying disabilities, and display a badge (if you want. The value of a badge like this is more obvious) if they find your site to be accessible. The main idea of the service however is to offer the kind of testing that lots of people here are talking about.

  25. #25 On December 5, 2004 06:36 AM Norty Pig said:

    I agree with what you say about Bobby's AAA rating and it being a tool for developers not a set in stone standard. A lot of people think their site's accessible because Bobby said so though and it is becoming a general propaganda being thrown about. Good marketing by Watchfire. Personally I think aiming for pleasing AAA on every site in every situation is a little bit obsessive. I've noticed quite a few get a life obsessives on these issues. Like your work, something to aspire to. Ciao.