Joe Clark: Accessibility | Design | Writing

Comments on the Ontario accessibility spec, February 2009 edition

The specification for information and communications accessibility is barely improved over any previous “draft.” It’s unworkable. It conflicts with other laws, chiefly federal copyright law. It forces publishers and libraries to create alternate formats against their will and, in some cases, illegally.

By far the worst sin is vastly exceeding the legal duty of organizations to provide accommodation short of undue hardship. The spec forces organizations to do every crazy thing a remote committee has ever heard about that even tangentially relates to “accessibility” – even if the information or communication is already accessible or something else has already been done to make it thus.

The specification isn’t ready for prime time and has to be sent back to the drawing board. The specification, if passed in anything resembling this form, will result in legal challenges. It would be my pleasure to appear as an expert witness in those cases; this is a standard that needs to be struck down, not rushed through.

I know you had Jutta Treviranus of the ATRC helping you, but I guess you didn’t trust her. I was aghast at the atrocities of earlier drafts (still uncorrected, I see), and I told a committee member the following: “Jutta is incapable of making that many mistakes.” Obviously committee members overruled her. This is what you get when you stack a committee full of non-experts – and, incidentally, when you exclude me from it.

The commenting process is inaccessible

Surely the pièce de résistance was the ministry’s insistence that comments be submitted solely in a file format that has few accessibility features and was proprietary and secret till 2008 – but one that just coincidentally happens to work just fine on the Windows machines that are the only computers the ministry has ever known. Word to the wise, as it were: Asking for comments on an accessibility specification in Microsoft Word shows you learned not a single thing about accessibility during this process.

What’s wrong with it this time?

The same things that were wrong with it the other two times I put in hours of unpaid effort writing critiques:

The committee ignored those critiques. I shouldn’t be surprised, because despite my 30 years of accessibility knowledge, extensive publication history, and actual practice, my application to join the committee was rejected. Had I been on the committee, I could have at least called bullshit on half the nonsense that was proposed, and forced the minutes of meetings to attest that the committee was not unanimous.

The specification:

Five times something is not five times better

Over and over again, the specification insists that content creators – protected by federal copyright law, which in turn includes droit moral – produce multiple versions of everything they create. ¶5.3.1 insists that every scrap of print be reproduced five different ways. Not only do you have to produce some kind of E-text (unspecified; this alone will lead to lawsuits), you have to do the impracticable and create electronic Braille files. (Using what? MS Word?)

You have to produce 5n files for every n documents you create – now and forever, even if you know with certainty no person with a disability will ever need such formats, and even if the original format (e.g., valid HTML) already is accessible to most disabled users or all known disabled users (¶5.3.6).

¶5.3.4 forces creators to transcribe every audio recording. That clearly includes every music CD sold in a store. And you have to invent a system that gives “the ability to amplify, pause, and repeat audio.” I thought I was just making a recording, not a machine to play back that recording.

For video, the insanity heads straight for the horizon line and requires (in ¶5.3.5) captioning and description and a transcript (even though we already have captioning) and one or more sign-language translations. (I choose British Sign Language – is that OK?) Does the committee understand this makes all of YouTube, to name one example, illegal? (Google is an organization operating in Ontario.)

For appointments, we now have to provide for all of sign language, deaf-blind intervenors, notetakers, and real-time captioning. However will all those people fit into my gynecologist’s office for my annual pap smear?

Requiring the invention of new technologies

The spec requires organizations to invent entire new equipment that can display E-text in any number of ways (including a nonsensical adjustment of “spacing”; ¶5.3.6), machines that can understand human speech (¶¶5.4.2&4), and machines or equipment that can understand responses in “text” (¶5.4.1).

We have to provide alternatives to mechanical keypads (¶5.4.5) and the use of mechanical controls as alternatives to touchpads (¶5.4.6).

Your phone system also has to accommodate chat [¶5.5.2(a)]. Your text system has to accommodate voice [¶5.5.3(a)].

Is this specification about accessibility for disabled people or about forcing Ontario organizations to invent a perpetual-motion machine? That’s about as attainable as what the spec actually requires.

Forcing the production of alternate formats

Under the federal Copyright Act, the copyright holder has exclusive first right to create an alternate format, and has the sole right to create a large-print alternate format. Only if the copyright holder fails to create an alternate format other than large print may anyone else do so. And no alternate formats may be made of cinematographic works.

That’s the law, around which the specification does an end run. The spec violates the legal rights of creators by stripping them of their right of first refusal to create most alternate formats and their exclusive right to create large print. It violates creators’ right to control the origination of derivative works, and infringes on the droit moral of authors by violating the integrity of an original work (through its transformation into an unauthorized version with different authorship).

§6.1 sets out an entire flowchart of mechanisms to violate the federally guaranteed rights of copyright holders. This in itself will result in a lawsuit, for which I would be pleased to be lead complainant. I would love to personally take you and your friends down over this provision. I already produce my works in accessible formats, and I aggressively defend my copyrights. Plus I know “information and communications” accessibility inside and out.

You’re going to have a real problem if you try to push this provision through. The most likely outcome, of course, is a reaffirmation that a puny committee of a provincial government cannot single-handedly overturn federal law. This may be an expensive and embarrassing lesson for your committee, but if that’s what it takes, fine.

Libraries have no choice

Much worse is the requirement for “educational libraries,” clearly including all university libraries and the Toronto Reference Library, to create alternate formats on demand. There is no limitation whatsoever on this requirement. Libraries do not have the legal right to create alternate formats in the first place, but we know that already.

The specific issue here involves the following scenario. A disabled person enters a library and says to the first person they meet “I would like your entire collection made available in accessible format.” The library must do exactly as requested, despite the profligacy and vexatiousness of the request.

Publishers have no choice

Almost as bad is the illegal infringement on publishers’ Charter-guaranteed free-speech rights and federally-mandated copyrights. Publishers must produce accessible formats, specifically for multimedia (or, as the spec writes it, “multi-media”) material. But really, the intent is to force every publisher operating in Ontario to produce every single item in multiple alternate formats.

The committee doesn’t know the first thing about document production and is not afraid to display its ignorance. “Committee Comments Section 6” (sic) falsely states that “publishers have... materials in electronic format prior to production of hardcopies of textbooks.”

Publishers may have nothing but camera-ready copy, or scanned image files. The latter are “electronic”; what they are not is accessible or usable for blind people, the chief (and usually only) beneficiaries of this provision. Publishers may have entire textbooks “in electronic format,” but their rights to use such books may be contractually limited.

Most importantly, committee members are so unfamiliar with computers – everybody uses MS Word, right? – that they seem to think every computer file can be instantly used by disabled people. Desktop-publishing documents are not accessible documents. If I have the rights to do so and if I don’t mind having a gun put to my head (that’s what the specification does), I can hand you my InDesign or Quark files. Now what are you going to do with them? (They’re “accessible,” aren’t they? Because they’re “electronic”?)

But I reiterate: This portion of the specification requires publishers to do something against their will. It forces them to generate speech they do not otherwise wish to generate. That’s an infringement on their rights in the first place, but the premise is false. The fact that a “file” is “electronic” means nothing in accessibility terms. (Are Braille books electronic? Is large print? Aren’t they accessible too?)

The specification takes over the election process

In the guise of making “information and communications” accessible to disabled people, the spec hijacks the entire election process, requiring all-candidates meetings equal in number to the number of political parties running. This has nothing to do with accessibility, of course.

Nobody authorized you to require WCAG 2

Committee members are locked in a Windows universe in which their sole window on the Web, you might say, is a blue e. With this level of acumen, I am not surprised that committee members are apparently unaware that WCAG 2 is not the only accessibility standard in existence.

There are two WCAG versions – WCAG 1.0, published in 1999, and WCAG 2.0, finalized on 2008.12.11. Both versions of WCAG have different conformance levels (three levels each).

There’s also an independently-developed set of errata for WCAG 1.0, the WCAG Samurai Errata, that bring WCAG 1.0 reasonably up to date with contemporary development methods. In WCAG compliance, developers choose to conform to:

There are no other options. The spec eliminates the first option; that is a fair goal of an accessibility standard. But it is unreasonable to require the use of WCAG 2 when WCAG 1 has not been cancelled and is perfectly suitable for most Web sites that do not use JavaScript, Flash, or other interactivity. Nor does the committee have any right to dictate the level of WCAG compliance to be met.

The standard is atrociously written

Produced by non-writers with little discernible command of English grammar and punctuation, the specification reads like the output of a committee. Many definitions are simply wrong or irrelevant and are, individually and together, reason to sue the Ministry to stop the implementation of the spec.

Take it from the top

The fact that this version of the specification was even allowed into public view with the endorsement of the committee is troubling, as the spec is a total mess.

The committee needs to start over again, preferably by firing a few members from the committee and drafting in a few more experts, very much including me. The committee has a history of ignoring outside advice; notably, this version is barely different from previous “drafts” of the specification and contains most of the same mistakes. It seems logical to assume that the committee wants to rush this version into production as a final standard without so much as a deleted comma. You’ve probably been given a deadline, in fact.

Well, you’re not going to make your deadline. This thing is a disaster: It won’t actually achieve accessibility for people with disabilities (and do nothing more), it overreaches into areas of federal law, it requires onerously redundant alternate formats, and, quite simply, it violates the constitutional rights of content creators and others.

Really, this stuff isn’t that hard. Deaf people need captioning and interpreters, blind people need accessible documents and audio description. That solves 80% of the problems right there. The remaining 20% require effort and finesse; all the committee has displayed is effort. And it isn’t good enough.

You absolutely do not have consensus on many of the areas of the draft standard. Even if you thought you had it, you don’t have it now.

Try pushing this through and not only will people sue your asses off, you’ll get nothing but bad press – including bad press from dedicated advocates of accessibility. I assume you would like to avoid either of these nightmare scenarios. Now’s your big chance; you’ve certainly blown every other chance.

Posted: 2009.02.06 14:10

You were here: HomepageAccessibility; → CRTC filings and other interventionsAODA specification → The Ontario accessibility spec, February 2009 edition

Homepage: Joe Clark Homepage: Joe Clark Media access (captioning, Web accessibility, etc.) Graphic and industrial design Journalism, articles, book