Validity and Accessibility

Why is the WCAG working group lowering the importance of validity by moving it? Gez Lemon has taken a look at the minutes from a recent WCAG working group meeting and found some very disturbing info. Read Validity and Accessibility for his comments.

Unless I’m completely misunderstanding something, this sounds like a really stupid move. One of the reasons given is this:

Current WYSIWYG and CMS tools do not necessarily generate valid code, making it difficult or impossible for many authors to meet this SC. (We cannot force authors to do manual coding to conform to WCAG.)

Hello? It’s the tools that need to be fixed and the developers and authors that need to upgrade their skills, not the guidelines that should be dumbed down.

Update: Matt May explains why this was done in A principled argument. I agree with the reasons, but I am still worried about the consequences.

Posted on June 20, 2005 in Accessibility, Quicklinks, Web Standards


  1. This is indeed a sad day. We’re finally getting somewhere with standards and then they do this!

    Of course the tools have to be fixed. Of course valid web standards code is the way to go.

  2. Will wonders never cease.

    As much as something like this horrifies me… why am I not as surprised as I might be?

    Roll on WCAG 2…

  3. I don’t understand what they expect to achieve with this move. OK, so more sites will become ‘accessible’ to a higher degree, but only because they’ve changed the ruler by which they do the measuring. Those websites will still be broken.

    This seems like fixing the issue of motorists speeding by allowing motorists to have a broken speedometer. ‘Sorry officer, but my speedo tells me I was doing 20mph, not 60mph’ ‘Oh, my apologies sir. On your way now…’

  4. I agree—it’s like fixing speeding by allowing a broken speedometer. Of course they can’t force people to hand code to meet standards, but these are standards that people should want to meet. People can complain to the companies producing the products they use and eventually those companies will be forced to improve or move on.

    I realize that this is highly idealistic and probably won’t work, but when people begin to realize that they can’t produce valid code with the software package that they use, they will probably change to something else that will (if they really care about producing valid code).

    The WCAG to, in my opinion, wimped out on the issue of valid code and it’s more than a little disappointing. Not all code will always be valid. It’s a goal, to be sure, but sometimes it can’t be met. When that goal is met, I think that a little higher recognition is necessary. A gold standard, if you will, as opposed to the bronze standard of stuff that loads and sometimes even looks cool.

  5. This is a turn for the worst…I’m really disappointed by this. Validity IS achievable, and if the software is not up to scratch - then the developers should try harder and strive for the refinement that is desperately needed across the web. Some people even hand-code in valid markup (such as myself)…tools should be good enough by now to ‘assist’ shouldn’t they?

    #Laughs #@ Matt! Yeah - so true!

  6. Without valid code, DOM scripting doesn’t work. So I hope they at least “reccomend” sticking to the standards. Offcourse big players like microsoft don’t do open standards, so they would lobby to have their broken code validate.

    They said they were going to make their generated code accessible. Now I know how they planned to do this exactly. =(

  7. June 21, 2005 by Wayne Godfrey

    Is this coming out of Washington, DC? Sure seems like the American way. Dumbing down has been an American institution for decades now. Why do they take this approach?

    Money talks..all else walks.

    Knowledge costs and that affects the bottom line. Unfortunate but true.

  8. I am sorry but there are an awful lot of web folks in the world and 95% depend on the WSYIWYG tools they have to use. Sure they don’t eat, sleep, and poo W3 - so what? Valid code should not be a requirement to make a site mildly accessible, especially given that the massive majority of people that maintain content do not care about code and couldn’t possibly keep a site valid 24/7.

    I think the whole valid thing site wide is a great goal but you need to think practically. You can a make a site accessible without re-writing the code to standards and just cause you have standard code doesn’t mean your site is accessible.

    The decision to downplay valid code is could be in response to the fact assistive technologies do not require the code to be perfectly valid to work fine. In fact many ‘valid’ sites that pass bobby are in-accessible. But what would happen if you could get sued or have a human rights complaint against your business simply because you claimed priority 1 compliance but you had a <br /> wasn’t a <br /> ? or even against the government?

    Think about how massive government sites are. In Canada the Federal and Provincial governments strive for priority 2 - which in WCAG 2 will mean the code needs to be valid right? Well you have millions of pages, tens of thousands of maintainers to teach, and a lot of templates to upgrade. Good use of tax dollars if their sites are accessible without being valid?

  9. June 23, 2005 by Roger Johansson (Author comment)

    Jesse: What you are saying is correct. What I’m worried about is the message that this sends. There is a fairly large risk that people who don’t really care see this as a go-ahead to skip validation. That could then lead to sites being accessible to users of assistive devices that rely on IE/Win’s HTML parsing but inaccessible (or accessible but unusable) to those who use Firefox or Safari or any other browser that follows web standards more strictly than IE/Win.

    Yes, the current WAI guidelines also have validation as a priority 2 item. I would have liked to see it made priority 1, but after reading Matt May’s response (thanks, Dave) I can see why they decided to move it back to priority 2. I also agree with the reasons Matt gives for doing it, but I am worried about the consequences.

    Yes, web accessibility is first of all about making it possible for disabled people to access the web. But my opinion is that providing that access should be done without potentially breaking the web for others.

  10. What I would like to see is that they phase in some requirements so that the technology can be brought up to speed. There is no point having valid code if there are no advantages for the assistive technology users… but if they starting saying it would and had their technology take advantage of it then that would be great.

    I think what is required is that they don’t rely on what the browser renders and instead renders its own page - become their own type of browser. A media type for CSS would be great for that, or XML even… something.

    I do wish they would have a sort of ‘upgrade path’ for the WCAG rather than just ‘here it is, we may update it someday, add some things, you know whatever.’

  11. I code my page to all the relevent standards, using access keys, title and alt attributes, css layout, etc. Miss a quote mark or an unencoded ampersand, and all of a sudden its not accessible?

    …um no.

    accessibility is about methodology, not standards, validation, in terms of accessibility, is a should have, not a must have.

  12. July 31, 2005 by Roger Johansson (Author comment)

    Jim Depends on the context of that quote mark or ampersand. In most cases nothing serious will happen, but it is possible for unencoded ampersands to cause real problems for everybody, not just those who use assistive technology: Ampersands and validation.

  13. Our website contains numerous accessibility guidelines and might be of interest to you:

    There is plenty of free advice about how to quickly create accessible content for websites.

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.