Web Site Expert巻頭レポート(英語)

WCAG 2.0: Accessibility Standards Reach Beyond HTML

A couple of summers ago, I went to dinner in downtown Los Angeles with a couple I'd known since college-and got a lesson in wheelchair accessibility. Smart and funny, my friend had survived multiple surgeries related to breast cancer and wasn't going to let wheelchair use stop her from a good night out. I volunteered to push, and as we made our way from the specially equipped van to the restaurant, and from there for a stroll in the LA night, I began to understand what curb cuts, ramps and other accommodations mean to someone who can't walk-which over time, let's face it, could be any of us. A basic curb might not seem like a big deal, that is until you try to push a wheelchair over it. A simple thing like sloping the curb to the street turns out to be a big deal, indeed.

Since 1999, these kind of simple accommodations have been going on in Web design, as well. And just like the building codes, the Web Content Accessibility Guidelines, or WCAG, are a set of specifications for building websites that are as accessible by as many people as possible. It has taken eight years between WCAG 1.0 and WCAG 2.0-a process so lengthy that people were commenting on (and sometimes grousing over) the new standard back in 2006. WCAG 2.0 takes into account the changes to Web technology since HTML, including dynamic pages, multimedia, and even the CAPTCHA (Completely Automated Public Turing Tests to Tell Computers and Humans Apart) tests that try to figure out whether you are a human or a machine.

Behind WCAG 2.0 is a basic premise: that access to information is a basic human right. But it's one thing to agree with the idea that the Web should be equally accessible to all, and quite another to put the WCAG's design principles in practice. WCAG 2.0 is comprised of 12 broad guidelines, including, non-text content, time-based media, keyboard accessibility, and navigability. Each of these is then subdivided into "success criteria." For example, under the general guideline "understanding," success criterion 2.4.7 says any keyboard operable user interface "should have a mode of operation where the keyboard focus indicator is visible." That is, users should be able to clearly see which part of the screen they are interacting with using the keyboard. You might display a vertical bar in the field where text is to be inserted.

That's one success criterion. There are many others. For Web designers, trying to build an accessible Web page by referring to these criteria en masse would be like trying to be a good driver by reading the entire motor vehicle code. Designers need something to distill it, and that will come in the form of training courses, or even better, web authoring tools and other technologies that have the guidelines built in. Indeed, that's already happening. On the heels of the WCAG 2.0 announcment, Deque Systems Worldspace announced compliance-checking software compatible with the new guidelines. Other efforts and building in the spec are sure to follow.

Judy Brewer, who directs the Web Accessibility Initiative at the World Wide Web Consortium, says that WCAG 2.0 is not the end point, but is part of an ongoing process that will, in time, lead to websites that include as many people as possible. WCAG 2.0 represents guidance for accessibility of Web content and applications, and is backed by online resources for working designers. Updated guidance for authoring tools and browsers is still a work in progress. Brewer also stressed that the W3C is not a policy making organization. Rather, it brings people together and gets them to reach consensus. Consensus is seldom easy, and that's the achievement of WCAG 2.0.

Brewer is W3C's chief liaison on accessibility policy and standardization and holds a research appointment at the Massachusetts Institute of Technology's Computer Science and Artificial Intelligence Laboratory. A wheelchair user, she also knows something about accessibility in the physical world. We spoke by phone.

How did the World Wide Web Consortium first get involved with accessibility?

The W3C was started in 1994 by Tim Berners-Lee, the inventor of the Web, as a place to develop Web standards in an open forum. Within a few years, some people in the Web community realized the importance of ensuring that all the standards coming through W3C should help ensure accessibility to people with disabilities-whether visual, auditory, physical or cognitive. The Web Accessibility Initiative started just a few years later, in 1997. One of our mandates was been to look across all of the technologies and technical standards under development and ensure that they could support accessibility. For instance, we looked at the Synchronized Multimedia Integration Language, or SMIL, an XML language for describing multimedia presentations. We wanted to make sure that the SMIL specification included, among other things, captions for people who are deaf or hard of hearing and descriptions of the video for people who can't see what's on a website.

We also work on developing guidelines for the accessibility of content for browsers and media players, and for the authoring tools used to create Web content. We want to make sure both that people with disabilities can use those tools, and that the tools help support production of accessible content. A third layer is the evaluation of tools that assess Web content accessibility, checking for conformance to the Web content accessibility guidelines. There are about a hundred of these evaluation tools for different aspects of Web accessibility, all of which also require human judgment. For instance, you can check automatically to see whether alternative text is available for an image, but you can't actually tell if it's the right alternative text. If a picture of an airplane is described by the alternative as a 'cow' an automatic checker wouldn't know the difference.

So artificial intelligence is not yet a factor.

The evaluation tools are getting smarter. For example, color contrast is important for some forms of color blindness, which affects up to eight percent or more of the male population, depending on the country. Some recent tools now check for something called "luminosity contrast ratios." That's not exactly AI, but the technology gives a more accurate assessment than what a person can do.

It sounds like WCAG is the Web equivalent of the accessibility standards in the physical world.

That's a good analogy-and it applies to retrofitting, as well. In the world of architecture and construction, buildings without ramps and street corners without curb cuts have had to be ripped apart and retrofitted to build accessibility in. It's cheaper to build accessibility in from the beginning, and that's what we advocate for the online world, as well. We want to get ahead of deployment and build in the support from the beginning.

As Web content has become "essential reading," are more groups paying attention to accessibility?

We have seen a huge increase in interest from government and industry. And the interest has been global, from the U.S., Europe, Canada, Australia, and of course Japan. China now has Web accessibility standards based on our work, as does India.

What's going on in Japan?

One of the key players is the Japanese Industry Standards Association, which has been working on Web accessibility for about four or five years. They felt that our earlier version of the Web Content Accessibility Guidelines were not completely adequate for their needs, so they formed an access working group several years ago to develop some of their own standards. That has also been true elsewhere. But we feel that because the Internet is an international phenomenon, there's a huge benefit to having a unified set of Web accessibility standards used internationally. If you use the same international standards for Web content accessibility, then authoring tool developers can focus on product support to help meet those standards..

So the Japanese Industry Standards Association took us up on our invitation to get involved directly in our working group for the 2.0 version of the Web Content Accessibility Guidelines. A very dedicated group of people in Japan participated over several years and gave a lot of feedback into the specifications. They translated drafts into Japanese and got feedback from industry and the disability community. We've been very pleased with their cooperation and contributions, and we've heard from our colleagues in Japan that they're pleased at the changes reflected in the new standards.

Eight years have passed between WCAG 1.0 and 2.0. What did you learned in between?

Perhaps the most important thing was that WCAG 1.0 didn't address the rapid advance of Web technologies. For instance, 1.0 mainly addressed HTML-and we actually cautioned against the use of scripting with regard to accessibility. WCAG 2.0 acknowledges the importance that scripting now plays in Web design and addresses how to program in accessibility. Similarly, we've addressed multimedia more precisely; for instance, videos need to have some accompanying accessibility information that describes that video. And media players need to communicate via an API with assistive technology, such as a screen reader that redirects displayed text displayed on to speech synthesis. The goal here is to make all kinds of Web content work for people with disabilities.

We learned other things, as well, that are reflected in WCAG 2.0. People wanted more supporting material to help implement the guidelines, and the guidelines themselves needed to be more understandable. The guidelines needed to be more precisely testable so that people can clearly know if they've achieved the level of accessibility we describe. In some cases with 1.0, there were not clear testing procedures. With 2.0, we've done that.

And we also addressed the use of CAPTCHA [Completely Automated Public Turing test to tell Computers and Humans Apart] tests: those annoying little graphics of almost indecipherable letters that were designed to foil spam bots. The trouble is that if somebody is using a screen reader, they aren't going to be able read what's in that little image and you can't put alternative text behind it because the spam bot itself would be able to read it. So in 2.0, we say that you need to think about other screening tests, as well, such as a logic test, so that people with disabilities aren't unfairly excluded.

What are the guiding principles of WCAG 2.0?

There are four main principles. Perceivability, which covers things like making sure there are text alternatives for images, captions for audio, and that there is sufficient color contrast. The content needs to be operable, which includes guidelines for keyboard access, timing of input-so the session doesn't time out too quickly if you have difficulty moving your hand. The content needs to be understandable, so there are guidelines for readability, predictability and input assistance. The final principle is that the material needs to be robust, which includes a guideline for compatibility: the Web content has to work smoothly with different technologies, including some of the specialized technologies for people with disabilities.

It sounds like in all of these, authoring tools will play a big role.

Absolutely: they have become the cornerstone. Most people aren't going to take the trouble to sit down and read through the guidelines, no matter how elegant and thoroughly documented. They are going to want the software to do it for them. Version 2.0 of the Authoring Tool Accessibility Guidelines, which is now in draft form, will give guidance to developers of software for producing Web content and, in turn, help Web designers achieve conformance with the Web Content Accessibility Guidelines. So the quantum leap will come when mainstream authoring tools support the production of accessible content. That is, when they have this capability built right in.

When that happens, the tools themselves will encourage designers working on all kinds of things-management systems, social networking, and interactive Web features-to think about accessibility. The tools will remind them to use alternative text images, and that a table needs a summary, descriptive column or row headers. The tools should then walk you through the process of creating accessible content. That, I think, will do the most towards driving up the percentage of accessible content on the Web.

Sight impairment and hearing impairment are so different. Have you had to work on two sides of the house to address both needs?

The challenge is actually broader than that. For example, somebody might have a photosensitive seizure disorder, so that flashing content could trigger a seizure. We've built that into our guidelines. In Japan, an aging population is a huge issue. You have a lot of people who have subtle changes not just in vision and hearing, but dexterity, and short-term memory. The last two are functional impairments, but not necessarily what you think of first when you think of disabilities.

Who is driving the move toward Web accessibility standards?

The guidelines have been driven from several different directions. On a practical level, as the Web has taken on enormous importance around the world, people with disabilities don't want to be excluded. We've also seen an increasing awareness from governments around the world, some of which reflects on the accessibility gains made in the physical world. For example, I use a wheelchair, and whenever I come out to the San Francisco Bay Area, I'm in heaven because of the decades of work that have gone into making the physical environment work well there. Most of that didn't happen on a purely voluntary basis. Even though there are strong business and civil cases for accessibility, that's often not enough to get people to do the right thing. So there's been a fair amount of "encouragement" from governments in different parts of the world. In Japan, accessibility remains largely voluntary, and we'll see if that remains the case in the future. We've also seen some companies take leadership positions on accessibility to show they are ahead of the pack or simply as part of their corporate social responsibility philosophy.

There's a newer driving factor worth mentioning: the United Nation's Convention on Rights of Persons with Disabilities. This is a treaty with an almost unprecedented amount of support: 137 signatories including Japan, though not, so far, the U.S. It was in development over two years and went into effect last May. The treaty completely reframes the issue of Web accessibility, making the case that accessibility isn't just a nice thing to do for business reasons, or maybe because your government requires it. The treaty says that access to information is an essential human right. There are some concrete ways to go about ensuring this right and most countries that are taking this up are looking at the W3C for technical guidance.

Now that WCAG 2.0 is in place, how will implementation proceed?

We've had a group of Web developers using the guidelines on a trial basis for several months, trying them across many different types of Web content and software technologies. Now that 2.0 has been formalized, we expect that that broader adoption will accelerate. For instance, in Japan, the guidelines will be included in the Japanese industry standards for Web accessibility, thereby getting attention from both industry and government sites there. In Europe, we have supporting testimonies from two different commissioners, which constitutes a pretty high level of endorsement.

On the individual level, a lot of people already trained on use of the earlier Web Content Accessibility Guidelines will now get training for WCAG 2.0. But we think the transition will be smooth. The principles behind 2.0 are very consistent with 1.0. So we've found that most websites that already conform to WCAG 1.0 won't require significant changes. Of course, there will be some exceptions.

How much of WCAG 2.0 and its supporting materials are already translated into Japanese?

Currently, the WCAG 2.0 proposed recommendation, which includes the main specification, are translated. I believe that organizations are now working on a more formal translation of the final version. I'm not sure of the schedule for translating the supporting materials, but there is a very committed group of people working on Web accessibility in Japan and they have been coordinating closely with us.

おすすめ記事

記事・ニュース一覧