The Accessibility for Ontarians with Disabilities Act (AODA) is a rather weak piece of legislation that gives businesses and other organizations in Ontario until the year 2025 to become accessible to people with disabilities. One of the tasks of the Accessibility Directorate of Ontario is the creation of a standard for information and communications.
That sounds like trouble right there, doesn’t it? A government agency setting down rules for how you communicate? These things have a tendency not to end well.
I applied for membership on the committee and was turned down. The membership list is filled with unknowns (except for Geoff Eden, a known blind activist). That isn’t so bad, of course, but here “unknown” could be synonymous with “no existing knowledge of accessible communications.”
When forming the committee, the ministry involved did at least resist the temptation to stack it with functionaries from known organizations, though the CNIB is in there, an important point we will return to shortly.
I have an initial “draft” of the committee’s standard, still very much unfinished and dated 2007.11.06, and have certain comments to give.
The accessible information and communications standard will recognize and consider other similar standards, regulations and legislation, including standards developed under the Accessibility for Ontarians with Disabilities Act... and the Ontario Human Rights Code.
“Recognize and consider” may be the goal, but in practice, the “draft” simply rubber-stamps other standards and gives them the force of law. It does so even when the standards really aren’t standards, were never user-tested, and have been rejected by the open market. That describes several of the standards mentioned in the draft.
Captioning: Requirements for captioning shall meet or exceed the requirements established in:
- The Canadian Association of Broadcasters Closed Captioning Standards and Protocol for Canadian English Language Broadcasters, Second Edition, February 2004;
- Federal Communication Commission’s captioning technical specifications in Part 15 of the FCC 47 CFR regulations
The CAB “standard” was cooked up behind closed doors in an intentionally secret process. It involved broadcasters’ and lobbyists’ writing down what they thought they knew about captioning. (The “standard” cited a great many research papers, most of which I provided to the committee almost post-facto, but still managed to get things wrong.)
If the AODA committee had enough knowledge to cite this “standard,” presumably it had enough knowledge to Google it (even using IE6 on Windows) and learn how many things are wrong with it.
The FCC technical specifications really deal with transmission and are superfluous.
Video description: Requirements for video description shall meet or exceed the ITC Guidance on Standards for Audio Description, May 2000 (Ofcom).
It’s audio description, not video description. I guess somebody told the committee there really weren’t any standards for audio description, except this old one from a foreign country. (Tell me: Must description tracks be recorded in a British accent to comply? Which British accent?)
And again, if the committee has enough skill to locate this “standard,” surely they also located a critique of it.
English Braille: Requirements for English Braille shall be in accordance with the “English Braille, American Edition, 1994” approved by the Braille Authority of North America.
French Braille: Requirements for French Braille shall be in accordance with: Le Code de transcription de l’imprime en braille (1989) or L’index de l’abrège ortographique français entendu (1955)
Does that mean Braille should, must, should not, or must not be contracted?
Large print: Documents shall conform to the CNIB clear print guidelines .
This is the most infuriating requirement of all. These are, of course, the guidelines the CNIB itself notoriously misnamed “Ckear Print.” Based on half-assed research that concluded that no conclusions were possible about what to do to make large print legible, CNIB forged ahead and ignored its own results. Never having met a Windows computer font it didn’t like, CNIB attempted to impose Windows 98–era typography on the entire country.
According to this marvel of nonresearch, to be truly accessible, a document should use blue type on yellow (or just black on white), in 12 to 18 point on 4 to 6 points of lead, respectively, preferably in Arial or Verdana but only in a monospaced font, never using italics for emphasis.
As such, the Clear Print guidelines essentially ban graphic design. (And 12-point type isn’t “large print” in any sense.)
If the AODA committee has its way, this nonsense masquerading as a “guideline” would have the force of law, and you could be faced with a human-rights complaint if you fail to do what it says.
Captioning and description standards need to be developed independently and tested to prove they work. Having the industry jot down its own best guesses is self-serving at best. It’s been attempted over and over again and has failed.
I never used to be interested in government funding for the Open & Closed Project, but if this is what the Ministry wants to do, it’s time to submit an application. The government can’t force every organization in Ontario to comply with rules that somebody else pulled out of thin air.
The standard will not prescribe specific technology so as to promote innovation.
That’s badly written. More substantively, staying technology-neutral can produce incomprehensible standards (Cf. WCAG 2). Many sections of the current “draft” refuse to say what they mean. What are they actually talking about?
Some technologies, like actual fonts, are specifically required. (That isn’t a stretch; see my Clear Print discussion. And technology-neutral is explicitly defined as “not predicated on any individual technology or style of interface” [emphasis added]. Fonts are a style of interface.)
Being technology-neutral isn’t a bad idea; it’s just very difficult to do in practice.
So let’s go through some examples.
Access to information provided in visual form: Persons or organizations providing information in visual formats shall have the capacity to provide/shall provide upon request:
- At least one mode that does not require user vision; and
- At least one mode to users with visual acuity between 20/70 and 20/200; and
- At least one mode that does not require user colour perception; and
- At least one mode that does not require user capacity to decode printed text.
So, as I read this, every single piece of printed matter must be available in audio recording or Braille and also in large print and also completely translated into symbols. Those aren’t the only possibilities to comply with the section, but they are plausible ones.
Then they go on to list “acceptable solutions for communicating information in printed text”: Large print, “Braille-ready electronic format,” ASL/LSQ interpretation, or E-text, or “a functional alternative solution that meets the technical requirements.” There is nothing preventing a person from insisting on other formats, including any of those I have listed.
Now, what if your document is electronic? The committee acts as though “electronic text” could not be accessible in and of itself. You have to provide not only the E-text but an ASL/LSQ translation or an audio recording. This is foolish, as E-text can be and will be voiced by screen readers. A person without a screen reader might need other accommodation, but that isn’t how the spec is written.
Acceptable solutions for communicating information using graphics and videos: Persons or organizations presenting information in graphic or video form will be compliant... by providing the following:
- Text description of all images; or
- Video description in text and audio form; or
- A functional alternative solution that meets the technical requirements.
First of all, it isn’t a text “description” (“This is a picture of a logo of a company and it looks like this”). It’s a text equivalent, a replacement for the image. And you could meet this requirement by snail-mailing a printout of text equivalents. As written, you can decouple the text equivalent from the image if you want, rendering the text equivalent useless.
For “video description,” they mean “audio description.” There is no known use case whatsoever for a text file of audio description. It isn’t cued up with the original video and is completely useless. If the intention is to serve deaf-blind people, they’ll need more than an E-text file to understand a video they cannot see and hear well. The committee has no user testing to back up such a requirement. (No such user testing exists.)
They revisit this requirement in a later section, 7.3. It seems they really mean it.
What if you give people an “electronic audio” document? Well, you have to transcribe it. But you have to do that anyway: “Acceptable solutions for communicating information using audio” are three types of captioning, a deaf-blind intervenor, ASL/LSQ interpretation, notetaking, or an assistive listening device.
From what I can tell, every audio CD and radio station in the province would have to be provided in some or all of these alternate formats, even if technically impossible or ridiculous.
Incidentally, I read this entire section as requiring all art galleries to provide nonvisual access to their collections. While this has been done by some galleries and is achievable in many cases, the committee needs to be honest about what it is asking for.
A whole raft of requirements about “user motor movements” seems to be a euphemistic way of requiring kiosks to be accessible. Again, if that’s what is meant, say so.
Information must be “timely,” that is, “in a timeframe equivalent to that experienced by others.” (There’s no verb in the sentence.)
This is impossible in practice. Only in the highest-profile, highest-budget cases do accessible versions of books, for example, appear exactly when the regular-print editions do. It is imaginable that standard and accessible versions could be produced simultaneously, but not without undue hardship in some cases. Of course, for certain electronic documents, the standard version is the accessible version.
As written, every publisher in Ontario that refuses to issue large-print books (solely the copyright holder’s prerogative), or that does so after the regular-print book is issued, could be hauled before a human-rights tribunal. And that will happen, because members of the Alliance for the Equality of Blind Canadians have been talking about doing exactly that for years. Now they’ll have a legislative method to launch such nuisance complaints.
Many definitions are ill-written or simply incorrect. (Like the entire document, copy-editing is atrocious and I have cleaned up their numerous Windows 98–style mistakes.)
- accessible formats/alternate formats/multiple formats
- An accessible or alternate format is when a print document is made available in a format other than print so it can be accessed by a person with a disability. Common accessible formats include but are not limited to: accessible Web sites; large print; audio cassette or digital audio talking book; Braille; computer diskette or CD-ROM/DVD; described video; SignWriting, and on-screen text/E-text.
Nobody uses SignWriting in the real world. The overuse of slashes here as elsewhere – another sure sign of a government document – has unintended consequences, as a slash can mean “and,” “or,” or “and/or” (it’s recursive). Hence a person with a disability could insist on CD-ROM and DVD.
Also, large print is print.
- audio description
- The process of verbalizing text on a television or movie screen.
No, it is not. That’s what the CRTC thinks, but they’re wrong too. Audio description is the true and sole generic term for the creation of an additional narration track for blind viewers. “Verbalizing text” is merely one thing that is done in audio description.
- described video
- Provides audio descriptions of what appears in a video. Just as closed captions provide display text corresponding to the audio in a video, described video enables someone to hear what is being portrayed visually.
All that “described video” does is read out onscreen text? If I’m blind but I can hear, I already can “hear what is being portrayed visually.” This is such a half-assed definition I have trouble believing it actually made its way to a released document. I’m not even going to help them out here; they can find nice tidy airtight definitions of audio description, the sole correct term, by carrying out easy searches.
The entire definition of “barriers (physical)” is off-topic and reads like an Easter Seals fundraiser rather than a real specification.
- Captioning is the process of converting the narration, dialogue, music and sound effects of a video production into text that is displayed on screens.
No, it isn’t. Captioning is a transcription of dialogue (narration is dialogue) and rendering of sound effects. The “displayed on screens” part isn’t technology-neutral and is superfluous. On a bad day, one could say “visible text.”
- culturally deaf
- Deaf culture consists of the norms, beliefs, values, customs, moral attitudes and manners shared by members of the deaf community. Culturally deaf people in America use American Sign Language.
Perhaps, but we don’t live there, and the government has no business writing a standard just for this group.
Plain language is one of the requirements of the federal government’s communication policy.
Perhaps, but we don’t work for them. (Also, the U.K. Disability Discrimination Commission, mentioned elsewhere, no longer exists.)
- Section 508
- Section 508 of the Rehabilitation Act of 1973, as amended, is a U.S. law requiring electronic technology used by the government to be accessible. Specific requirements are maintained by the Access Board in the Electronic and Information Technology Accessibility Standards (36 CFR Part 1194). This standard is structured as a set of provisions, each identified by a paragraph. There are 16 paragraphs applicable to Web technology, most of which correspond to WCAG checkpoints.
Perhaps, but we don’t live in the United States and the document doesn’t discuss WCAG anywhere else – except of course in another definition, which falsely claims that “the current version” of WCAG is 1.2.
The entire section headed “Types of standards” barely reads as English, let alone as workable definitions.
- Web accessibility
- Web accessibility refers to the practice of making Web pages that are available through an Internet browser interface, accessible the widest possible audience, including those with disabilities and other groups with accessibility constraints.
- Web accessibility refers to access to the Web itself, ensuring the equality of access to information that is available through this communication medium.
The first definition doesn’t explain very much and assumes everybody uses a browser (surely only IE6). Web accessibility is solely concerned with disabled users.
The first definition is actually a badly-written restatement of the idea of Web universality, except that nothing after the comma makes any sense.
They haven’t finished two other topics yet (among more than eight):
With the CNIB on board, there will inevitably be an insistence on using only Tiresias for every sign. The committee, I speculate, is trying to figure out how to finesse the timing of the release of its own spec to coincide with WCAG 2 so it can require the use of WCAG 2 everywhere in Ontario.
Posted: 2007.11.26 ¶ Updated: 2008.03.20 11:32