Ah yes, the ongoing saga of whether or not WCAG 2.0 should include validity in there somewhere.

Well, according to the lastest minutes of the Working Group, they have finally got around to looking at it, and they have decided to refer the issue back to Committee.

They had previously dropped validity from their criteria, simply instead requiring “unambiguous parsing”, but given the popular support for retaining validity in there have referred this back to Committee. I’m pleased with that. I’d like validity to be re-included, but what was also important to me was the thought that the idea would be treated fairly and with due accord, rather than simply written off because they don’t like the sound of it.

They certainly seem to be considering is seriously, and I thank them for that. I just hope that they arrive at the correct decision…

It might be worth explaining why I feel validity deserves to be included in the first place…

Firstly, validity is important as a strong indicator of accessibility. If a page or a site is valid, this shows that the developer has gone to some degree of effort to ensure that it has been put together properly and this is an indicator that the page is more likely to be accessible.

However, I find myself — despite being one of the “universalists” he so despises — agreeing with Isofarro that invalidity does not equate to inaccessibility.

You can spot a universalist wearing an accessibility disguise by listening to their rationale as to why a website is not accessible — this includes whoppers like “Flash isn’t accessible because it doesn’t work in Lynx”, and “its inaccessible because it doesn’t validate”.Isofarro

I really don’t want to rehash the argument with Isofarro about how to define accessibility, other than to point out that I’m one of the “universalists” he decries, and that I agree with him wholeheartedly as regards those two “whoppers” statements. And I suspect most of the other “universalists” would agree.

I don’t have a problem with accessible Flash. I’ve seen a very good ‘Report A Housing Repair’ service that uses Flash, appears to be accessible (although my experience of testing Flash for accessibility is limited), and because you’ve got nice clear interactive diagrams, is likely to be of much more benefit to someone with cognitive impairments than the plain-Jane HTML equivalent would have been.

This service also had a non-Flash method for reporting a repair which wasn’t as good. That’s what we call Graceful Degradation or Progressive Enhancement. And that’s fine. Great site, great piece of accessible Flash, with a non-Flash fallback. Perfect.

I do therefore feel as a “universalist” that I’ve been mis-represented somewhat, but that’s by the by — I don’t want to dig over old ground, I want to look at the issue of why I believe validity needs to be included in WCAG 2.0 despite agreeing with the sentiment that an invalid site is not necessarily an inaccessible site.

If a site is valid, it has been built to a set of formal rules, and follows them, meaning that the user agent — assuming that it interprets those rules correctly — will not have problems in rendering the site to the user.

Most current browsers do better than this. They’ll take code which is all over the place, doesn’t follow the rules, has badly-nested or unclosed tags, and will manage to translate this into something usable which can then be rendered to the user. Well, hurrah to them.

The accessibility guidelines WCAG 1.0 included validation as a level 2 criterion, meaning that if a site wasn’t valid, it wouldn’t be impossible for a disabled user to use, but, according to the WAI, this would mean:

A Web content developer should satisfy this checkpoint. Otherwise, one or more groups will find it difficult to access information in the document.WCAG 1.0

So it appears that under WCAG 1.0, the WAI seemed to consider that either validity was part of accessibility, or that universality was part of accessibility; after all, these are the Web Content Accessibility Guidelines we’re talking about, as produced by the Web Accessibility Initiative.

So what is going on? Is validity a part of making websites accessible to disabled people or isn’t it? You want a simple answer? Well in that case, it is and it isn’t. Validity is a concept related to universality which may impact upon what access by users with disabilities (and others).

Validity relates to universality in the sense that a valid document should be treated the same way by whichever user agent you have. Validity relates to accessibility for disabled users in the sense that an invalid document may behave and be rendered in an unpredictable and possibly unusable fashion, and just because current mainstream browsers are able to extract sense from tag soup, doesn’t mean that every browser that is ever popular will be able to do the same.

…and when you consider one of the guiding principles of WCAG 2.0 is:

Principle 4: Content should be robust enough to work with current and future user agents (including assistive technologies)WCAG 2.0

… then it is my contention that in order to ensure that your content is robust enough to work with unknown future user agents, it must be constructed in a predictable manner so that these unknown future user agents can render it predictably, otherwise someone using the as-yet-unknown user agent could encounter problems that would impact upon whether or not users with disabilities were able to use the site.

Did I mention the user agents would be unknown?

Validity is a concept that, whilst related to universality, if not achieved may result in accessibility problems with a particular user agent. If you want to ensure — as the WAI do with WCAG 2.0 — that content has to be robust enough to work with unknown future technologies, then you simply must tie down the structure of that content to an agreed specification of some sort, so that the future user agent can understand how it is meant to work with it.

So we need validity included. Maybe not at level 1, but it needs to be there.

Although if you’re building or commissioning a website, you need to consider this. If you site is not universal enough to work on a particular user’s configuration, then it doesn’t matter how accessible it is, they won’t be able to use it. If it’s universal but not accessible, then usability is irrelevant as disabled users won’t be able to get to it anyway. If it is universal, accessible and not usable then no-one’s going to bother with it because it’s too difficult to use.

Remember The Holy Trinity: Universality, Accessibility, Usability.