Let’s talk about accessibility

August 25, 2006    Accessibility

The other day, Jeff posted some thoughts about accessibility which were, for the most part, well-received. Some people apparently didn’t get the message, though, and went after him in the comments, so he had to post a follow-up clarifying that, no, he’s not an evil corporate monster who hates handicapped people.

Jeff’s first post was somewhat kicked off by Roger Johansson’s post on Monday, which had comments that got so bad he had to close them for a while, then re-open them with a note:

Comments are open again. You are free to disagree with my personal opinion on this matter. However, do keep it polite and constructive. Name-calling, trolling, and personal attacks will not be tolerated.

When looking at some of the nastier comments left for Jeff and Roger, I wonder: what the hell is wrong with these people? Why can’t we have real discussions about accessibility?

Welcome to Accessibility Club

The first rule of Accessibility Club is: talk about accessibility. John Foliot probably thought he was being oh-so-funny when he pointed out some unescaped ampersands in Jeff’s HTML, but really he was just making himself look like an asshole. What do ampersands have to do with accessibility? Absolutely nothing, as he even admitted in his comment, so leave off it and actually talk about accessibility.

The second rule of Accessibility Club is: talk about accessibility. I repeat myself here because this bears repeating. For example: making petty swipes about designers “painting their pretty pictures” has absolutely nothing to do with accessibility, and just cements your place in the asshole camp. Yeah, John, I’m talking about you again.

The third rule of Accessibility Club is: leave the spin at home. If someone says something you don’t agree with, building a straw man that’s easier to attack isn’t going to win you the argument. Grant Broome, this one’s for you.

The fourth rule of Accessibility Club is: Accessibility Club is not Fight Club. Now, there are people in this world who don’t honestly care about accessibility, and yelling at them isn’t going to change their minds. Those people just need to be kicked in the balls.

But there are people in this world who do care about accessibility, even if they’re not always able to do everything you, personally, would like them to do. Yelling at them isn’t going to make them start doing the things you want them to do, and runs a real risk of alienating them entirely. People who are making good faith efforts but suffering under real-world constraints are not your enemies.

What accessibility is not

Now that we’ve covered the ground rules of discussion, let’s take a look at a few important concepts which need to be kept in mind while we talk about accessibility:

• Accessibility is not a binary toggle. HTML has clear rules on what’s allowed and what’s not, and figuring out whether something is valid HTML or not is a matter of checking the HTML against the rules; either it follows the rules or it doesn’t. There is no equivalent of this for accessibility, because accessibility means different things to different people. Even the oft-maligned W3C Web Content Accessibility Guidelines acknowledge this, by introducing the concept of “levels” of accessibility and by being called guidelines.
• Accessibility does not exist in a vacuum. It’s impossible to talk about accessibility without considering it in some particular context; “accessibility” as a Platonic ideal divorced from any context just doesn’t exist. This follows logically from the non-binary nature of accessibility: the question “is this accessible or not?” is useless. The useful questions are specific situational ones, like “would adding skip links to this site make the content easier to access for users of screen readers?” and “would setting these navigation links to display as block make it easier for users with muscular/coordination impairments to reach and activate them?”
• Accessibility cannot be decided by rules. This follows logically from the above. Yes, there’s WCAG 1 and the under-construction WCAG 2, and there’s some hand-waving in US federal law and some other countries have more or less adopted bits of WCAG 1 as law. But because accessibility isn’t a binary thing and has to be considered in context, any attempt to codify it into specific rules is doomed to failure — there will never be any set of rules that’s good enough to judge “this is accessible” versus “this is not accessible” for all contexts.

I would trust that these things are obvious, but repeated experience has taught me that such trust would be misplaced.

What accessibility is

And now the flip side:

• Accessibility is a continuum. For a colorblind person, a site which takes care to provide contrast between the color of foreground text and the background color it displays against is probably going to be more accessible than one that doesn’t. For that same person, a site that also takes care to distinguish links by underlining them is probably going to be more accessible than one that doesn’t, and a site which carefully chooses the colors of its visited and unvisited links is probably even more accessible. There are lots of things you can do to improve a site’s accessibility in a given context, and how accessible your site will be in that context depends on the exact combination you implement.
• Accessibility is an ongoing process. There’s no point at which you can step back and say, “OK, our accessibility work is done”. Working your way along the continuum means paying attention to your users and to available accessibility techniques, and making changes over time.
• Accessibility involves trade-offs. As with any feature, accessibility requires knowledge, time and effort; in most contexts, this means it’s also going to cost money. Making the right choices to improve a site’s accessibility means looking at the benefits a particular technique offers and weighing them against the cost of implementation.

With Jeff’s posts that last one seems to be where the controversy centers, so let’s look at a couple of examples:

• Providing alt text for images doesn’t require much time or effort, so it’s a low-cost improvement. And the accessibility benefits for visually-impaired users are significant; without that text, an image may well be useless to them. So this is almost always a good trade-off to make.
• Providing transcripts of videos (or embedded captions, though that’s a messy realm of technology) may be easy if you’re producing the videos yourself and they’re scripted (for example, a TV news broadcast is almost always read from a prepared script). And the accessibility benefits for hearing-impaired users are on a par with the benefits of alt text for visually-impaired users. But if you’re not producing the video or if it’s unscripted, it can have much higher costs — you’ll have to have someone transcribe the video, and not everyone who puts video online can afford that. Whether you provide transcripts for videos, then, depends on the resources you have available.

And just to hammer it home, because I know there are people who will ignore the rules of Accessibility Club and try to shove words into my mouth: when I say it “depends on the resources you have available”, that doesn’t mean “it’s OK not to care about it”. That doesn’t mean “don’t worry about it”. That means “it depends on the resources you have available”. For example, if you’ve got transcripts for your videos or if you’ve got the ability to produce them, and you know it would provide a significant accessibility benefit for some of your site’s users, then you should provide transcripts. But if you don’t have them and if producing them would be prohibitively expensive, then don’t provide them, and focus your resources where you know they can do good. In general, I think it’s better to reach the people you can reach than to reach nobody at all.

Closing thoughts

Web accessibility is a deep and complex subject, and — pardon the pun — not always a very accessible one. There are lots of things which can impair someone’s use of the web, in lots of different ways, and some are more common than others. A lot of accessibility guidelines have been propagated over the years based on anecdotal evidence or on hypothetical considerations (usability, which is arguably a superset of accessibility, suffers from the same problem). What solid research there is on the topic can be hard to find at times, and hard to understand when you do find it. Many of the laws regarding web accessibility are horrendously vague. And no matter how hard you try to get it right, there’s probably going to be an asshole somewhere who will be all too happy to crucify you over a particular trade-off you decided not to make.

But that doesn’t mean we should give up. It means we should be willing to adapt, willing to share what we know in a civilized manner, willing to make the trade-offs we can afford to make, willing to ignore the inevitable rogue assholes and, above all, willing to learn.