97% of websites still inaccessible

United Nations Global Audit of Web Accessibility, conducted by accessibility agency Nomensa on behalf of the United Nations, shows that 97 percent of websites fail to meet the most basic accessibility requirements. One hundred websites from twenty different countries and five different sectors, including travel, finance, media, politics, and retail, were tested. Only three of those one hundred sites passed all Priority 1 checkpoints of WCAG 1.0. Three.

A story on the BBC News website, ‘Most websites’ failing disabled, notes that 93 percent did not provide alternative text for all images, 73 percent relied on JavaScript for important functionality, and 98 percent of the sites did not use valid markup.

Sadly, the results do not come as a surprise to me. In fact, they were highly expected.

It is a sad fact that most Web “professionals” still do not even try to make the sites they build accessible. I have performed similar surveys, focusing on Swedish websites, with similar results. Public sector sites are generally slightly better than those in the private sector, but only slightly. A couple of years ago I believed that by now, the majority of people who make their living from designing or developing websites would have realised the importance of accessibility. I was wrong.

Unfortunately the majority are simply not interested in creating high quality work as long as they can produce junk and still get paid.

Posted on December 15, 2006 in Accessibility, Quicklinks

Comments

  1. That is so sad and disappointing.

    Sometimes, personally, I get lazy with websites I know the customer doesn’t care is accessible or not. But, by doing this, I suppose I’m adding to the glut of inaccessibility that is out there.

    sigh Time to stop being lazy, I guess. Lead by example.

  2. Unfortunately the majority are simply not interested in creating high quality work as long as they can produce junk and still get paid.

    That is the most succinct way to put it. I think that is the most direct, poignant way to chastise people who aren’t making an effort to design with web standards. Thanks for saying it, as I will probably be using some form of it in the future.

  3. Forget WCAG 1.0 60% of the prominent sites in India just don’t work. I mean half the time they are crashing. The ads are ill placed. The text is hard to read. Some of them will have text box so small that your age will barely fit there and they want you to put in your full name there.

    Almost all of them will not render nicely in firefox.

    Not having alt for images is least of my problem. I stopped visiting them since they don’t play nice with firefox. Now I get my news about India from google news.

  4. December 15, 2006 by Alejandro Moreno

    It’s very sad. I work for the public sector (in Canada), and there’s nothing that says we should build accessible web pages. So of course, we don’t.

    I develop mostly intranet apps, and right now the attitude is “they need this app, so they better be using IE, with JS and CSS enabled; if they don’t, screw them.”

    =o(

  5. That is really pathetic. But, as you stated, as long as they can keep putting out crap and still get paid - where is the motivation to do better?

    I think the same holds true for agencies/design firms around here - they put out crap and get paid. Client doesn’t know any different, and there is no punishment for doing things wrong (in their eyes).

    I know we have been down this road before, but what is the best option? I dont think we need to continually educate our clients (some dont have the time or care to even know) - yet I dont think its right that people get paid for garbage. But, again, what is the motivation to do better?

  6. Roger. Delete this comment if you think I’m over the top or off base or simply because it’s your site.

    Countless times I see you write “reader’s digest” style of articles on standards and related material. I see articles here that recap what you have said in 10 other previous articles with a different page title. While you are contributing to the community with your findings, and I’m sure it is much appreciated by many, you are not moving up the notch in really educating the people. The reason is; I see you preaching to the choir by writing articles on how great standards is and how unfortunate it is that people are not using them widely.

    What I would like to know is, if you are so concerned about the Web, why don’t you simply come down to IRC Efnet #CSS and volunteer by helping the developers everyday in real-time? I warn you though, there are no ads or SERPs to think about.

    This is not meant to be an attack on you personally but how frustrating it is to see the same type of ‘noise’ day after day on the Web.

    With warm regards, -Sarven

  7. Yes it is sad that so many poor developers refer to themselves as “professionals” when those select few of us strive to do things the right way. I work for the US Government managing the web presence of one of the agencies. For my agency, valid code has been made mandatory. I was in the position to write the policy for it, so I did. From here on out — as long as I or the other key members of my group — are here, any work that is put forth will be done the right way, accessibility included.

    Instead of choosing the easy way, we’ve all chosen the right way — a choice I wish other web professionals would make more often.

    Re: Sarven – I barely have time for the list groups and MeetUps I am a part of, let alone helping people out in real-time on Efnet. Roger makes excellent information available to the public. Instead of complaining to Roger, why not use csarven.ca to set the notch?

  8. December 15, 2006 by Chokdee

    I can relate to Sean McGee’s comments about being lazy. A long time ago I didn’t fully comprehend how important the online world is to visually impaired folks until I married my wife and met her kid sister who is 17, in high school, and is legally blind.

    She can see color and light and is able to read font size 72pt, but mostly her online world is seen thru JAWS. She navigates like a champ - even on sites that are inaccessible. She is truly inspiring. She understands that “sighted” people cannot fully comprehend her world. Needless to say, after meeting her, I vowed to make sites I design as accessible to everyone as possible.

    BTW, was there any further news on the lawsuit against Target’s website concerning its inaccessibility?

  9. December 15, 2006 by Roger Johansson (Author comment)

    Sarven:

    What I would like to know is, if you are so concerned about the Web, why don’t you simply come down to IRC Efnet #CSS and volunteer by helping the developers everyday in real-time?

    Three reasons:

    1. I didn’t know about that IRC channel.
    2. I don’t have the time.
    3. Even if I did have the time, I think I can help more people by writing articles that can be read by anybody, than by answering the questions of a handful of people (or a couple of hundred, I don’t know how many people are normally in #CSS).

    I don’t subscribe to any Web related mailing lists anymore for the same reason - I find them extremely time consuming.

    And I’m sorry that you think I am repeating myself. To some extent I am, but when I do it is generally intentional. Most people don’t read everything I write.

    Chokdee:

    BTW, was there any further news on the lawsuit against Target’s website concerning its inaccessibility?

    The last thing I heard was that Target requested that the case be dismissed but the judge turned it down.

  10. December 15, 2006 by Collin Miller

    Resolved, that accessibility is, and will continue to be a joke until the big boys look up and realize we’re working on something bigger than their pocket book here.

    Don’t just blame developers. Many of them can and should be doing a lot more to aid in accessibility. Valid syntax, correct markup and the like go a long way toward accessibility.

    Having standards that are actually standard can deal the largest blow to the behemoth.

    When everything behaves the same way everywhere accessibility presents itself as a logical conclusion, not a rebellion worth fighting for. The standards-conscious developer could rest well knowing their work fits into a well-formed world.

  11. Sarven

    I see articles here that recap what you have said in 10 other previous articles with a different page title.

    If you don’t say the same things time and again, it’s likely that new readers will never hear you say them. As with most philosophies, different people play different roles (I’d say religion, but I’d hardly call web standards a religion). There are the missionaries, the evangelists, the preachers, etc. Missionaries would go into the wilderness and build churches and help people (or get on IRC channels and help developers with code). Evangelists would tour around the country giving pre-written speeches about a certain facet or new area emerging in the philosophy. Preachers stay in their church / blog and say the same thing over and over again so that people never forget what the fundamental point of the philosophy is.

    No one cog is better than any other. We need each one. We even need lay-people and non-believers. We all work for the greater good. Lets not feel self-important and start dissension among the ranks. Lets support each other for the role we play.

    Amen

    :)

  12. 73 percent relied on JavaScript for important functionality

    My goodness. The other stats are expected, but this is a real shock. I didn’t even know that 73% of websites even used JavaScript, let alone depended on it. This just emphasizes the need to educate web developers about Unobtrusive JavaScript.

  13. Sarven - perhaps Roger is following the standard marketing rule about exposure; it takes several times for people to even notice a message, and many more before they actually really notice it and understand it.

    Given that not everyone subscribes to the 456 RSS feed then not everyone is going to know about every article on web standards compliance, accessibility and usability that Roger writes.

    Therefore, do the math and you’ll see that Roger has to write this message into hundreds of articles before a large number of developers actually see, read, understand and are sold on the idea of working towards the greater good by developing standards and better practice compliant websites.

    As for browsers and user agents - the more standards compliant sites web professionals produce, the more incentive web browser developers will have to develop browers angled towards leveraging the possibilities of these sites and start making obsolete old HTML 3 sites; progress here has to be parallel from both sides.

  14. @Ryan Christie and Nathanael:

    I’m speaking of reducing the “noise” that’s on the Web with the whole standards movement. There are plenty of praises already (perhaps no need to stop as Robert also suggests), but what I find lacking is the transition for old-fashioned developers or the beginners to adapt to these practices easier. If you can’t convert them into doing something the right way, by explaining and showing ‘how to’, it doesn’t matter how great something sounds. At the end of the day, people want to know how to do foo, bar, baz.

    @Roger Johansson:

    There are plenty of IRC channels both on EFnet and Freenode that deal with Web development; from #CSS to #PHP. #CSS alone has close to 200 people everyday. There are plenty of questions and answers on real-world problems and why following a set of standards is good for them. Plenty of good debates on practices and problems that arise as well. You are welcome to join and contribute.

    It is indeed time consuming to join mailing lists, or help in an IRC channel, or even post your findings to a centralized location (i.e. Wikipedia). In the end, however, it is that bit of contribution that makes all the difference.

    @Robert:

    You’ve gained my utmost respect with your comment. I will pay a visit to your site now.

  15. This number is complete BS. When you write “97% of websites are inaccessible” what you should have written is “97% of websites are inaccessible to the .001% of our population who are blind, deaf, crippled, and using an out-of-date version of lynx.”

    Firefox + IE5+ = 90% of the browser market. How can you claim for those 90% of the people that the sites that use javascript for navigation don’t work? They don’t care! It works!

  16. December 16, 2006 by tripleshift

    i read in the article and in many comments a sense of frustration towards laziness of webdevelopers not operating with standards and accessibility in mind.

    this frustration is legitimate and i understand the general concern, but i think that the reason why so many websites are indeed faulty in different ways, is not to be found in the laziness of webdevelopers, but instead in their ignorance.

    if you know how to make a website, how to write proper, semantic html markup and how to use css to get the presentation results you desire, you do it as naturally as drinking water. its not an extra effort anymore like when you have to learn those things for the first time. i mean, it is like riding a bicycle, once you’ve learned, you never forget. sure you need to practice and keep yourself updated on new issues and techniques, but you don’t just forget what you know because you’re lazy.

    so when a webdeveloper builds a crappy website we shouldn’t blame his laziness, instead we should blame his incompetence or ignorance.

    many so-called professional webdevelopers are actually lazy in learning semantic, standards and accessibility as long as the client still pays, and in this aspect, yes, we can blame laziness.

    that is why roger’s articles (and i give my humble answer to serven) are indeed more important than his eventual help in irc and newsgroups. when roger writes about not using frames or about the correct use of doctype and so on, he does way more good then 1000 times the time he could spend on irc.

    as roger, i’m not surprised at all by these numbers and as a final note to this comment, i’d like to say something in regards to elliott’s comment.

    accessibility is a form of freedom. take freedom of speech for instance, when this freedom is denied to one person or category, freedom ceases to exist. so is accessibility. when a website is denied to one, it is denied to all.

    accessibility is only when is for all.

    bye

    tripleshift

  17. Elliot Back

    The official estimates for disabled people is some 600 million worldwide. That’s some 10% of the total population, of which around 80% are believed to live in developed countries.
    I think the above indicates that a higher number than “.001% of our population” may be in need of improved access to information and services on web sites.

    One doesn’t have to be “blind, deaf, crippled, and using an out-of-date version of lynx” in order to have a limited set of choices in today’s societies, and there’s no real need to set, or keep, accessibility-barriers on the web. The web is after all one of the few options left for many disabled people, who want to be part of our societies.

    Regarding reliance on Javascript, and the (around) 90% it works for…

    They don’t care! It works!

    You’re probably right - they don’t care, and that’s probably the main reason why the problem isn’t addressed and solved on such a high number of sites.

    I don’t put too much emphasis on numbers either, but the fact remains: the negative numbers are too high when it comes to accessibility.

    I am not “blind, deaf, crippled, and using an out-of-date version of Lynx”. Instead I’m just mildly affected by aging, and I’m using the latest version of Lynx alongside the latest versions of other good browsers. That’s enough to make a large number of otherwise useful sites pretty inaccessible, so I can imagine what the more severely affected, or disabled, web surfers may find.

  18. I spend time on forums and mailing lists helping people out with CSS problems so I can understand where Sarven is coming from but the thing is, people need some prodding in the web standards direction for them to end up in these places asking their questions in the first place.

    I can give them advice on how to solve their problems or do things in a better way, but something has to have sparked their interest in this area to start with and that’s where people like Roger come in.

    So I completely agree with Robert in that we need all types of people to help promote web standards.

  19. December 16, 2006 by Benjamin Hawkes-Lewis

    Elliott Back:

    97% of websites are inaccessible to the .001% of our population who are blind, deaf, crippled, and using an out-of-date version of lynx.

    This is deeply misguided. ‘Accessibility’ (Wikipedia) is being used here in a perfectly normal way to refer primarily to disabled access. There is a major debate in web development about whether ‘accessibility’ is:

    1. About disability, pure and simple.

    2. About interoperability too because that’s required for disabled accessibility (because of the variety and expense of assistive technlogies).

    3. About interoperability too because that’s the only way to sell disabled accessibility to people who can’t grasp that making things accessible to the ‘blind’, ‘deaf’, and ‘crippled’ is a decent thing to do.

    The 97% figure comes out of a context where ‘accessibility’ is primarily about disability, not interoperability.

    Before attacking someone else’s statistics, you might like to double-check the feasibility of your own between pulling them out of the air and setting them down as evidence. The systematic collection of global disability statistics is still in a planning stage, but even from the limited figures we have, and even taking account the major difficulties of concretely defining categories like ‘blind’, ‘deaf’, and ‘crippled’, it’s obvious that 0.001% is a wild underestimate.

    Of the 6,224 million people in the world in 2002, the World Health Organization estimated that 37 million (0.59%) were blind and an additional 124 million (1.99%) had low vision (PDF). (Note all figures for world and US population are taken from the US Census Bureau.)

    Out of 5,916 million in 1998, the World Health Organization made the conservative estimate that 120 million (2%) people worldwide cannot hear a normal conversational voice with their better ear (PDF). In 2005, out of a worldwide population of 6,451, the World Health organization estimates 278 million (4.3%) suffered from moderate to profound hearing loss.

    I can’t find any statistics for the ‘crippled’ (I suspect they exist only in disaggregated form), but including only the ‘blind’ and ‘deaf’ gives a combined total of 2.59%. Yes, that’s 2,590 times more than your guestimate.

    Of course, disability is about more than your narrow pigeonholes of ‘blind’, ‘deaf’, and ‘crippled’. As Georg adumbrated, in a recent paper, WHO estimated that ‘10% of the world’, population — approximately 600 million people, of which 200 million are children — experience some form of physical, mental, or intellectual disability’. The World Wide Web Consortium’s Web Accessibility Initiative aims to address the needs of all of them, and it was the Initiative’s Web Content Accessibility Guidelines that Nomensa used as the basis for this United Nations survey.

    Incidentally, in case you’re thinking that these figures are inflated by the poor health regimes of developing countries and are unrepresentative of the comparatively wealthy population with easy internet access, estimates from the United States seem if anything to be higher (wealthier populations can afford to collect more extensive statistics I guess). For instance, estimates from the Survey of Income and Program Participation indicate that out of 287,716,000 Americans, around 1 million (0.35%) are deaf and 10 million (3.5%) are hard of hearing. In 1994 the National Health Interview Survey found 15% of the population were disabled by its definitions. 38% of those over 60 were disabled, so there’s a good chance you and I will be joining them eventually.

    I also wanted to pick you up on your crude reduction of interoperability-for-accessibility to working with ‘an out-of-date version of lynx’. I think you’ll find few people test with a seriously out-of-date version of Lynx, because very few people use a seriously out-of-date version of Lynx, and it’s relatively easy to upgrade. Debian is notorious for being the slow coach when it comes to adopting the latest versions of Linux software. The version of Lynx in Debian stable is 2.8.5, which is the current release. The only browsers where people regularly seem to test old versions seem to be Internet Explorer, Firefox, Opera, and Safari (in roughly that order). The other browsers are lucky to get tested at all. Web Content Accessibility Guidelines 1.0 was certainly designed to make web content available in text browsers, but in practice the most commonly considered problems of disabled interoperability involve the major graphical browsers and assistive technology. Accessibility-orientated developers tend to be asking whether something works with the JAWS screen reader and Internet Explorer or Window-Eyes and Firefox, because whether it works in Lynx is a pretty poor test of that.

    Firefox + IE5+ = 90% of the browser market. How can you claim for those 90% of the people that the sites that use javascript for navigation don’t work? They don’t care! It works!

    Web accessibility is about the entire potential audience, not ‘90% of the people’. Also, even if we accept your 90% statistic, it doesn’t follow that JavaScript works for that 90%. First, many people no longer test in Internet Explorer 5. Second, many people use Firefox or Internet Explorer but have JavaScript either restricted or completely disabled for security reasons. (And many of those are on a corporate network and cannot change their own settings, even if they know how.) Third, the vast majority of disabled people are using either Internet Explorer or Firefox, so your 90% includes people who have often have trouble with JavaScript navigation. Fourth, a lot of JavaScript navigation is poorly designed, so that people find it hard to use because it works contrary to their expectations.

  20. December 16, 2006 by Benjamin Hawkes-Lewis

    Hmm. Sorry if that’s hard to read; Markdown seems to have stripped out my curly quotes and apostrophes.

  21. December 16, 2006 by Roger Johansson (Author comment)

    Benjamin: Sorry about that. I’ll take a look at the settings for the various plugins I use to see if I can stop that from happening again.

  22. December 16, 2006 by Jeroen Zaalberg

    Unfortunately the majority are simply not interested in creating high quality work as long as they can produce junk and still get paid.

    Like Robert, I think the key lies in this paragraph but for a different reason. It’s creating versus producing. Taking pride in doing your job the right way.

    Accessibility in the narrowed form of screenreaders, which we all are talking about as the reason for standards, is really not what the standards are about. Accessibilty is just a deritive of doing your job the right way. All to often we take accessibility as a syncronym to the standards we should be and have to work with.

    When we “repair” our kitchen cabinet with some tape, we now we’re not doing it the right way. When a mechanic welds your new exhaust pipe on your engine instead of bolting it on, we are not happy. Both “solutions” will do the job but in the wrong way. Just like tables in a web page.

    When we set our eyes to another field: Builders and electricians. We normally think about them as labourers. But in reality they’re not. They’re every bit of a professional creator as we are because they (well, should) work according to standards. There should be such number of wall-sockets on a wall, placed at such a height. In this case, accessibility is again a deritive. What’s more important is that they have a passion to be up to standards and are interested in their job. They are creating instead of producing.

    Taking the screenreader as the reason and point of discussion for web standards is irritating. It should not be our focus for convincing someone to take up the standards for web design. In fact, it belittles the discussion. I am not saying that we should forget this group of people but focussing on screenreaders as we are doing is really using it as a scapegoat for implementing standards. It’s like ending sentences for someone who stutters or taking everything out of hand from somebody who is disabled for a crusade, and should be avoided. By taking screenreaders as focus we are forgetting about the real reasons for standards.

    Think a moment about real life and what it would look like if everything was “accessible”. Nice, right and it’s coming. Look at shops with special elevators for disabled etc. But again, only a deritive of the standards which we accumelate them on, year after year. Revisions of standards to make things better. And in web design we are only at the beginning of these steps. Funny, if you think about that we should be high-tech but in reality are just in our infancies.

    For accessibility, we should first become serious about our jobs. Working with standards is the right and only direction. Other ways are just “duck-taping” all the pieces in place, so it holds. Creating versus just producing. But if you want to talk seriously about accessibility (and I am not talking about semantics, here) we should first take up standards for the right reason: doing our job right and be passionate about it. Real accessibility will come along the way. Talking now about accessibility is like building a new garage to address your car problem for theft or weather conditions when not even your house is ready yet and I’ve got 97% of the internet to prove it. That 97% is not a number of accessibility but of compliance to the current web standards.

    Jeroen

  23. December 16, 2006 by Frank Taillandier

    For those who want to monitor the quality of their websites, some standardistas have gathered a list of best practices, to improve on-line services quality. It has been lately translated to english : Opquast

  24. 97% is a large number, and it’s a shame, but I’m not surprised as it does seem a lot of developers are getting away with not doing right by the client and accessible web development isn’t exactly in demand. In other words we have to be self-motivating and content with doing it because it’s the right thing to do. Clients want that the site looks nice on their browser and nothing else seems to matter. We end having to be educators as well as creators.

    As far as repeating oneself, I feel it’s bears repeating. The numbers support that.

  25. Elliot Black

    When you write “97% of websites are inaccessible” what you should have written is “97% of websites are inaccessible to the .001% of our population who are blind, deaf, crippled, and using an out-of-date version of lynx.”

    To your merit, there is probably less than .001% of the population that are blind, deaf, crippled, and using lynx. It’s the portion of the population that are blind, deaf, crippled, or are browsing alternative devices that we do accessibility for, which includes the person you defined in your made up statistic. ;)

    People with disabilities, as was said before, account for at least 10% of the population. I have no statistic for how many people browse on alternative devices like cell phones while they are out-and-about. I imagine it is quite a bit more than you suspect.

    And, thanks, Benjamin, for the informative reply.

  26. tripleshift - you’re right, it is ignorance rather than laziness. The extent of my laziness is that I might have too many nested DIVs in a page, so it’s not entirely 100% pure semantic but I just might not have time to optimise the code.

    If it takes 5 nested DIVs to implement a layout with rounded corners then that’s what I’ll go with. It still works, it’s still accessible, and although it might make standardistas raise an eyebrow it is standards compliant, readable by screen readers, search engine spiders and so on.

    I don’t think standards compliance is about meeting every guideline, rule and better practice to the letter with 100% semantic coding and all that. It’s inevitable you’re going to forget to encode an ampersand, or use a strong where you should have used a h4 or something. That doesn’t make it non-compliant. But not fixing up every little thing can be attributed to a forgivable laziness.

    Yes, it is ignorance that’s the killer. And I don’t think one on one hand holding is the answer. What’s the ration of aware developers and designers to ignorant ones? 1:100? Whatever the number, too many to help in addition to doing our jobs. Plus, what about all the ones that don’t ask for help on these IRC channels or other forums? The ones that don’t even know about CSS?

    It’s been driving me nuts the last few weeks having to continually pull BR tags out of a new under-development website and replace with margins or padding - but at the end of the day realistically even using BRs for spacing is forgivable. It doesn’t make the site inaccessible. It still works.

  27. December 16, 2006 by Benjamin Hawkes-Lewis

    Nathanael,

    I don’t think standards compliance is about meeting every guideline, rule and better practice to the letter with 100% semantic coding and all that.

    I don’t see how standards compliance can be about anything other than meeting every rule. Best practice could conceivably transcend compliance however, not least because the standards could be inconsistent, inadequate, unrealistic, or otherwise flawed.

    It’s inevitable you’re going to forget to encode an ampersand, or use a strong where you should have used a h4 or something. That doesn’t make it non-compliant.

    Well, actually it does, or ‘compliance’ is meaningless.

  28. The results are sad - but yes, I agree Roger - not surprising at all.

    I’ve said this elsewhere on the web recently - but I really feel that some parts of the Web Standards community have also been turning their back on bringing Web Standards (and Accessibility) to the attention of more people - not just designers.

    There’s a bit of the ‘who holds the purse-strings’ about all this though: Web Designers are already aware of Accessibility - but their employers and fee-paying clients don’t care for it because they don’t see it as being of importance. I don’t think it’s the designers who need educating about better web practices as much as the customers/clients. If the so-called ‘Web professionals’ who churn-out these poorly-made sites suddenly stopped what they were doing and refused to make anything less than an accessible, well-made site with good, clean mark-up…then their employers/clients would simply go and find someone else to do the job for them!

    We need to educate the Customers/Clients…(AS WELL as the designers!) - This should be the ‘real’ focus of the Web Standardistas for 2007!!

  29. December 17, 2006 by tripleshift

    @ Nathanael

    what i worry about the most is ignorance as i stated. you seem to agree but indeed, some of the things you said you seldom do, prove me wrong in a certain way.

    everybody was blaming laziness more or less to the same extend or, at least, of the same kind of the one you declare for yourself, and even though it is surely less dangerous then total unawareness of standards and accessibility issues, that kind of laziness can still endanger the struggle towards global accessibility.

    benjamin pointed out very well the problem here by saying

    I don’t see how standards compliance can be about anything other than meeting every rule.

    if is not so, if we let even the less non-semantic h4 with strong slip through, we create vicious precedents to redo this sort of errors and do even bigger ones. from strong as h4 to p for lists to divs for everything and so on.

    the bad attitude of compromising with important rules, makes the rules less important and therefor us more likely to brake them, either for ignorance or laziness.

    @ Matt Robin,

    interesting point that of educating clients, it is sure part of the problem, but i don’t think it is the most important and primary thing to do.

    you know that clients are on average way more appealed by quantity rather than quality. there wouldn’t be lame silly commercials on radio and tv or trash show where people throw furniture to each other if there was always a search for quality.

    lets face the fact that most final consumers are not as fine and delicate as graphic designers or art directors (well at least in theory :) ) are, let alone prepared and able to understand the technical mumbo-jumbo that is behind a well formed, accessible and semantic website.

    those targets are always making clients to lower their research for quality since most would be wasted. that is a shame, but is indeed the reason why things are the way they are. that plus incompetent web professionals who sell their work for quality when is barely quantity.

    in this light, i see that clients shall be on the receiving side of a change. they should find themselves in a world where good web-making, respectful of standards and accessible, is in fact THE standard and not an extraordinary work which (for the clients) is often not worth the hassle to pay for.

    i am not throwing away your whole idea, it is still important to make clients aware of the benefits (both for the user and the client) of standard compliance and accessibility in web applications, but that can’t be as important as:

    • all web professionals are made aware of the rules and learn those rules

    • every web professional is not lazy and keeps up with the rules evolution

    • governments make serious laws about accessibility of websites (not just theirs, if they were)

    unless (by law and for real) it will be impossible for a bank or an online store or any other website that offers any kind of service, to exist if not accessible, clients will always care little about paying more dough to be accessible.

    i know this is utopia, i get back to my fairy tale dream.

    sorry for the long comment.

    bye

    tripleshift

  30. Which part of the Canadian public sector doesn’t have accessibility guidelines? The Regina city government?

  31. notes that 93 percent did not provide alternative text for all images

    I hope this is a misprint. According to the accessibility sites I’ve read, and the experience I’ve had, you shouldn’t have alternative text for all images!

    You should always have alt attributes (tags) but only populate them with text if the images represent text themselves.

    For example, you shouldn’t use them for “Picture of Tree” as it’s not important for the user to use the site and it just takes longer for things like screen readers to complete their task.

  32. would be nice to know how they collected their random sample - did they only use large companies/organisations in the categories, or did they google for some keywords and pick at random?

  33. Unfortunately, such reports could be seen as justification to small business not to bother. “If 97% of websites aren’t accessible it won’t matter if mine also isn’t”.

  34. December 18, 2006 by Benjamin Hawkes-Lewis

    David G. Paul,

    If you follow the link Roger provided at the beginning of his post, Nomensa actually explain roughly how they choose the sample. (No, it wasn’t a random sample: sites were selected to represent a spread of sectors and countries.)

    Nice Paul,

    Or small businesses could sniff out an opportunity to serve customers that bigger rivals ignore.

  35. December 19, 2006 by John Serrao

    That figure seems misleading. With the crazy number of browsers and their various requirements, you wouldnt be able to forward web technology without leaving a few of the old people out. I mean, if we take accessibility to the letter, wouldn’t this disqualify most of Google and Yahoo, with their extensive use of AJAX?

    Im definitely with you on getting tables out of everyone’s lives and getting the web up to spec in the CSS dept. Unfortunately that isnt going to happen until CSS is adopted (and I mean fully adopted so we all dont have to hack our hacks). Maybe accessibility should be a bit more accessible.

  36. Benjamin,

    Many models of compliance, certification and assessment across many industries that I’ve had experience with allow for a certain margin of non-compliance which with certain conditions can still be labelled as compliant.

    Usually the conditions are “You should fix or address these things as soon as possible”.

    So in that context, I believe a web site can still be deemed “standards compliant” even if it doesn’t validate because a content author didn’t encode an ampersand, or a user submitted a comment and used a <br> tag in their markup which goes against the “CSS should be used for presentation” guideline.

    I’m not making excuses for laziness of any form, but I think it’s important to have a realistic and practical approach to this issue with feasible outcomes rather than a 100% hardline approach.

    Plus - personally I’d rather focus on designing for optimal user experience, performance, reliability and security rather than stressing over the minutiae of rare and relatively insignificant validation errors.

    I’m working with you guys here ok, so please don’t have the impression I’m at odds with the Improve The Web Movement. I’m just suggesting that sometimes people are too quick to pull out the microscope. Yes I’m an advocate for accessibility - both schools, web standards, standards compliance, XHTML, CSS, semantic markup, separating content and presentation and will continue to be.

    Just suggesting we set realistic goals for other designers and developers; don’t make an elite exclusive club.

    Did you get 100% on all your exams at college? Did you still graduate?

  37. Validation aside, it’s a shame that many developers don’t even follow simple basic standards and best practices. I mean take a look at my posted example about BostonPizza.com. JavaScript required for the navigation? How wrong is that!

  38. ^ Yes THAT is important; that’s a mandatory must-fix (or rather never-do).

  39. December 19, 2006 by Stevie D

    It’s important to keep the “inaccessibility” problems in perspective.

    Sure, there may be 97% of websites that, in one may or another, do not validate and tick every box on the accessibility list. But how many of those are serious problems?

    Missing or completely inappropriate alt text is a big problem. Alt text that we would consider to be not perfect, but not harmful, is not a big problem.

    Specifying a tiny font in pixels is a big problem. Specifying colour and contrast ratios that are just outside the recommended range is not a big problem.

    Failure to specify a doctype or character encoding can be a big problem. Failure to encode every ampersand in every URL is unlikely to be a big problem.

    Layout tables nested five deep will probably cause big problems. [div class=”left”] and [span class=”red”] may give the author headaches at a later date but won’t cause any huge access problems.

    And so on…

    Yes, we should aim for perfection, but we must be realistic in how long it takes us to get there. I agree with Nice Paul that

    Unfortunately, such reports could be seen as justification to small business not to bother. “If 97% of websites aren’t accessible it won’t matter if mine also isn’t”.

    We need to break the problem down into bite-size chunks. Start off by saying that, eg, 30% of websites will be difficult or unusable for people with poor eyesight. 25% of websites are useless for people with screen-readers. And we need to always make it sound as though the non-conforming websites are in a minority, or we’re giving the naysayers a big headstart!

  40. tripleshift: That’s a well-considered reply to my comment (I appreciate that). I feel I need to further clarify that all the points you’ve mentioned (in particular: ‘all web professionals are made aware of the rules and learn those rules’ and ‘every web professional is not lazy and keeps up with the rules evolution’) - all these are already underway in some form or another (although I agree there needs to be far more of it!) while customers are seemingly none the wiser and not demanding Web Standards through sheer ignorance. The points you’ve raised are definitely important - but we’ve got that covered - the Web Standards Community already takes these points to Web Professionals all the time…the next step is to get demand for Web Standards on the rise (only Customers themselves can makes these demands) - forcing lazy, (psuedo)-Web Professionals to get their skills up to date or go the way of the dodo. Web Professionals themselves certainly have no excuses for not using Web Standards in all of their work, you’re right - it should be the standard - but with no demand for such a standard to be in place it just won’t happen at all…(and that is regardless of how many web professionals know Web Standards or not).

    Or put another way: we could have all the Web Professionals on the Planet capable of using Web Standards, making brilliant, accessible web sites…and customers will still not want them for perceiving them to be a costly waste of time and effort…the true benefits of Web Standards need to be brought to the attention of the consumer/customer - the ones dictating how their sites are made. I just don’t see any effort in that direction by the Web Industry at all - none, zero, nada!

    Of course we also need to get Web Professionals up to standard too… I think we’re agreed on that my friend :)

  41. December 20, 2006 by Benjamin Hawkes-Lewis

    Nathanael:

    A specification may of course specify a margin of error allowed for compliance. Can you find any evidence of such a margin in the (X)HTML specifications? Such a margin doesn’t seem appropriate for a machine-readable document format.

    Your analogies of clubs and exams seem to be conflate how we assess web designers with how we judge whether a given document conforms. Even good web developers make mistakes; redefining those mistakes as good work doesn’t make any sense to me.

    You say you would prefer to “focus on designing for optimal user experience, performance, reliability and security”, but since I view conformance as a prerequisite for all four I don’t see any contradiction.

    Establishing that a given instance of non-conformance has no effect on usability is more time-consuming than just producing conformant documents in the first place. Developers’ assumptions about what doesn’t matter tend to be mistaken. For example, if you fail to encode an ampersand in XHTML, then Firefox and Opera will stop processing the document. If you use a STRONG where you should have used an H4, then screen reader users cannot navigate to that section by heading and document mappers cannot construct accurate tables of contents. Forbidden presentational markup like FONT can create new obstacles for partially sighted and colorblind people viewing the page.

    You assert that producing valid documents is an unreasonable expectation. Given the ready availability of validators and linters, I don’t see why that’s true. It is the case that many content management systems, while marketing themselves as standards-based, are actually tag-soup-in and tag-soup-out systems that guarantee markup corruption, “Valid XHTML” badges not withstanding. That’s an reprehensible flaw in those tools however.

  42. ACCESSIBILITY is NOT about DISABILITY, it’s about FLEXIBILITY!

    A website that contains really small text can be just as inaccessible for a person who wears eye glasses, as for a person who has extremely limited vision. The solution in this scenario is not to make the text bigger and bigger; but to make the text more FLEXIBLE, so that an online visitor can change the content to suit their individual needs, either by amending the browser text size or by using an adaptive device such as a Speech or Braille Reader.

    As an IT & Web Professional with nearly 8 years experience, I fully support accessibility, web standards, and the need for extensive manual testing. Sadly, many people (designers and developers included) still seem to be missing the bigger picture.

    So in support of equality, I’ve decided to take a more proactive approach to FLEXIBLE and ACCESSIBLE websites. As of January 1st 2007, all websites produced by myself will be automatically held to WAI & W3C standards (done in majority already, but compulsory from 2007 onwards).

    As a further incentive, I will also be including a ‘2-year compliancy warranty’ with web packages; ensuring that websites maintain their quality with evolving standards.

    This was not written as an advertisement to my own business, so I have not included my own URL in this post.

  43. ACCESSIBILITY is NOT about DISABILITY, it’s about FLEXIBILITY!

    A website that contains really small text can be just as inaccessible for a person who wears eye glasses, as for a person who has extremely limited vision. The solution in this scenario is not to make the text bigger and bigger; but to make the text more FLEXIBLE, so that an online visitor can change the content to suit their individual needs, either by amending the browser text size or by using an adaptive device such as a Speech or Braille Reader.

    As an IT & Web Professional with nearly 8 years experience, I fully support accessibility, web standards, and the need for extensive manual testing. Sadly, many people (designers and developers included) still seem to be missing the bigger picture.

    So in support of equality, I’ve decided to take a more proactive approach to FLEXIBLE and ACCESSIBLE websites. As of January 1st 2007, all websites produced by myself will be automatically held to WAI & W3C standards (done in majority already, but compulsory from 2007 onwards).

    As a further incentive, I will also be including a ‘2-year compliancy warranty’ with web packages; ensuring that websites maintain their quality with evolving standards.

    This was not written as an advertisement to my own business, so I have not included my own URL in this post.

  44. December 20, 2006 by tripleshift

    @Matt Robin

    i agree we agree, even though i don’t feel as optimistic as you, or even better, i feel a little bit less optimistic than you.

    i know most of my points are already underway, even the law part which i think is the most important and sadly the less improved so far.

    i simply think that a lot shall be still done in the evangelism field (roger himself pointed out that in an article he wrote a few months ago) and that we shall not consider web-professionals are even halfway through the learning process, let alone the fact that still most don’t know anything about accessibility and standards.

    in this i know we’re both on the same line, as most if not all the people who enjoy reading roger’s.

    nevertheless i felt and still feel we cannot rely on clients to demand for accessible websites, as much as we don’t expect constructors to build accessible slopes and lifts, unless there’s a law that forces them to do so. not just the high principle of offering accessibility. but this is just my point of view.

    we could have all the Web Professionals on the Planet capable of using Web Standards, making brilliant, accessible web sites…and customers will still not want them for perceiving them to be a costly waste of time and effort.

    indeed is true, clients have to be willing to pay quality, but in my opinion, to do so they need to be forced by law to offer an accessible service. this will cut the “so-called professionals” off the market and will make the web better.

    in saying that, i still want to re-point out the fact that when a webdeveloper knows the rules, he/she will always code accordingly. i mean, if a client comes to roger and asks for a website, but doesn’t care for accessibility, he will get an accessible site anyway, because roger does so naturally. i don’t think he’ll go with tables or font tags just because he feels lazy or the client doesn’t know or care about the value of standards.

    of course such expertise is not as cheap as a 15yo with dreamweaver, but there’s no talking about the value and quality in accessibility and standard compliance with people who doesn’t care or is not forced to go for it.

    clients will go for accessibility out of their own decision only in small numbers, the big rest needs a kick. not a talk, a kick! :D

    bye

    tripleshift

  45. Like some others I also work for the public domain maintaining a website (as well as an Intranet) part time while I work for myself and the shoddy code that gets passed off is unbelievable. The home page itself is up to 180 warnings and errors…

  46. December 22, 2006 by Chokdee

    ACCESSIBILITY is NOT about DISABILITY, it’s about FLEXIBILITY!

    Very, very true. “Legally Blind” does not mean completely blind. It also covers people with low-vision (like my wife who sets her IE text size to Large). Not everyone who is low-vision uses a screen reader.

    One area that ticks me off is inflexible navigation on a page either thru the use of images instead of text or px instead of something more IE-friendly.

    What I don’t get is that it is easier to code a navigation bar in standards than cutting up all those images and using javascript for rollover effects. Yes, you might have to compromise some graphic design factors but you would gain a wider viewing public. Seems like a no-brainer to me.

    Look at boston.com’s newly redesigned navigation. Nasty and unnecessary (also, a good case of “divitis”).

  47. shows that 97 percent of websites fail to meet the most basic accessibility requirements.

    Last year (2005) a similary test was done for german sites. Only 5 percent (more than 100.000 are checked automatically) are valid in german.

    In the beginning of WWW only text was used. In 1995 and above, Flash, Images and animated design came more and more to most sites. So, new questions are:

    Is the design the problem? Remove the design and we have a accessible site? Or is it the content?

    The W3C tell Web Designers to separate design and content. But will this work with rounded corners, shadow under the images, videos or flash? I don’t think so. What a site have more visitors?

    Text only in black and white or highend multi media site?

  48. December 29, 2006 by Roger Johansson (Author comment)

    The W3C tell Web Designers to separate design and content. But will this work with rounded corners, shadow under the images, videos or flash?

    Yes, why wouldn’t it work? That’s why we have CSS and JavaScript.

  49. I’d say 99.9% actually - considering that I assume:

    1. It’s not weighted towards more accessed/popular sites with real using value for common people. Most people really almost only use a few of the top 20 websites in their country, and spend 90%+ of their online time on these sites.

    2. It’s probably mostly homepages that’s been evaluated and they are usually better conforming than old resources and less mainstreamed content that end up on sub and sub-sub pages.

    Semi-related trivia. I’ve made some investigations on this from time to time the last 7 years, and I still don’t think any university/college in Sweden that conforms to w3.org’s core web recommendations: xhtml+css+wcag - not even the ones that teach web design! To me, that’s a total disgrace for higher Swedish education, and if they don’t prove me wrong soon I promise I’ll start a national boycott campaign for a thousand dollars to show people how behind, and ignorant, they are.

    As an example, just check this post on Karlstad University forums about redundant CSS and what they’ve done about it in the last 20 months. (That is, nothing.) And still they dare say they have “high academic quality” in multimedia, it-design, and innovation/design!

    My final words: CSS is celebrating 10 years, my god! Get it right - or get lost!!

  50. Anyway, I can see in this comment thread, that this “zeldmanian” community is still making the analogy design=style. (For example in comment 47, 48 - Matthias Mauch and Roger Johansson.) Stop doing it!

    [To] style = The way something is [being] presented. (Verbal, textual, physical, whatever.)

    [To] design = Something [being] designated - for someone, for some purpose. (A speech, a website, this sentence, whatever.)

    Thus, style in itself can partly be the content - and you can not seperate design and content, because crafted content is design/designed.

    Also, to make it even more complex, style is essentially designed (as with crafted content) - as opposed to being spontaneous, eternal or natural.

    Or, as Einstein wrote, “make everything as simple as possible, but not simpler.”

  51. January 4, 2007 by Max Bode

    Actually I think the problem is not that people don’t want to follow the rules…it’s more that they don’t know they exist.

    By accident i hit a webstandards page half a year ago…i had never even heard a single word of it before.

    Well it’s not that I never visited websites about (x)html and design etc. it’s more that the topic is just mentioned on really few sites. Or at least, at that time it was mentioned rarely.

  52. January 9, 2007 by Jon D

    Interestingly, I read a piece in Revolution! magazine this morning saying that Powergen (a UK electricity and Gas supply company) have decided to make their consumer website at www.powergen.co.uk as accessible and web standards based as possible.

    Apparently they see it as a key way to differentiate themselves from their competition.

    They’re certainly talking the talk, it will be interesting to see if they deliver, or it turns out to be PR hot air.

Comments are disabled for this post (read why), but if you have spotted an error or have additional info that you think should be in this post, feel free to contact me.