Joe Clark: Media access

You are here: joeclark.orgCaptioning and media access
CRTC filings and interventions → Intervention in Broadcasting Notice of Public Hearing CRTC 2003-3

Posted 2003.05.01

Intervention in Broadcasting Notice of Public Hearing CRTC 2003-3

This intervention concerns the license renewals of the following broadcasting undertakings in Broadcasting Notice of Public Hearing CRTC 2003-3:

Canal Vie
CHUM Limited (and applicable partners)
Canadian Learning Television
Pulse 24
Space: The Imagination Station
CTV Television Inc.
The Comedy Network
Outdoor Life Network
Report on Business Television
Talk TV
Global (and partner)
Alliance Atlantis
HGTV Canada
History Television
Musique Plus Inc.
Rogers Sportsnet Inc.
Teletoon Canada Inc.
The Score Television Network Ltd.
The Score
TVA Group Inc.
Le Canal Nouvelles
YTV Canada Inc.
Treehouse TV

It deals solely with accessibility issues and the need for industry and the Commission to support an independent research project that will resolve outstanding issues in captioning, audio description, and related fields and will provide needed standardization in those areas.

Permanent location

This intervention is permanently located at the address:

Support for renewals

This intervention supports the renewals of the above licenses – with modifications and if licensees support the Open & Closed Project. We’ll support their licenses if they’ll support our project.


I, Joe Clark, am the sole author of this intervention. I have been involved in the field of accessibility for more than 20 years, starting with a fateful encounter with The Captioned ABC News in the 1970s. In my career as a journalist and author, I have written over a dozen articles on captioning, audio description, and Web accessibility, the latter being the topic of my book Building Accessible Websites (New Riders, 2003). In a September 2001 profile, the Atlantic Monthly called me “the king of closed captions.”

I have done and continue to do paid consulting work with public- and private-sector clients on accessibility. I have written audio-description scripts for first-run cinema. I maintain a large Web presence on media access at

The Open & Closed Project

My extensive, decades-long expertise in accessibility was a factor in the development of the Open & Closed Project, a first-ever research and standardization project in multimedia accessibility. I am director of the Project, which is independent of broadcasters, service providers, equipment makers, industry groups, and regulatory bodies.

The project is up and running, with a funded activity investigating best practices in online captioning. It’s a small beginning for an ambitious project. In this proceeding, we have a once-every-seven-years opportunity to bring the Open & Closed Project to maturity.

Undiscussed issues in accessibility

When the Commission and broadcasters discuss captioning and audio description, they never manage to talk about the whole issue – and rarely even discuss what’s actually important. This section will serve to break that silence and introduce new areas of discourse in accessibility of television programming.

Proliferation of techniques

Through the entire histories of captioning and audio description, each provider of the service has done things differently. The more modest providers might argue that their work is at least as good as everyone else’s. It’s typical for service providers to contend, usually adamantly, that their way is the best and that their captioning or description is better than everyone else’s. They can’t all be right.

The enormous variation in captioning and description techniques means that viewers with disabilities must continually relearn how to watch television with accessibility features. Every program, and every commercial within every program, is captioned differently. Even at this early stage of the development of description in Canada, programs are described differently.

The Commission and broadcasters may not view the proliferation of captioning and description techniques as a problem, but rest assured it actually is. A viewer who requires captioning or description must continually relearn how to watch a program with those features. You the viewer must sit there and consciously decode what the service provider is trying to communicate. You cannot simply understand and enjoy the program. Captioning and description become foreground issues, a puzzle you the viewer must solve.

When every service provider does its own thing, we guarantee that the goal of treating accessibility as a necessary and inseparable part of television programming will never be reached.

If every set of captions looks and acts different, and every track of audio descriptions sounds and acts different, we emphasize an unprofessionalism and a disrespect for the audience using those features. We declare, however implicitly, that deaf and hard-of-hearing viewers are good enough only for whatever captions we feel like giving them; blind and visually-impaired viewers are good enough only for whatever descriptions we feel like giving them.

Further, we fall victim to marketing: We buy the contention of service providers that their work is best (or good enough). We accept that accessibility is an ad-hoc feature pasted onto a program well after everything that’s really important is taken care of. We advance the attitude, by deed if not by word, that captioning and description are too trivial, mechanical, unimportant, uninteresting, distracting, extraneous, or bothersome to be worthy of systematic study that emphasizes rationality, fact, and research.

The Broadcasting Act, the Canadian Human Rights Act, the Charter, and the commitments of broadcasters themselves oblige broadcasters and the Commission to finally, at long last, reconsider their disregard for accessibility. Broadcasters and the Commission have a first-ever opportunity not only to increase the quantity of accessible programming but to guarantee that the accessibility complies with independently-developed standards.

Accessibility practices based on opinion

When service providers advance the claim that their way of doing things is better than everyone else’s (or at least as good), they essentially advance an opinion. When broadcasters and CRTC employees, who themselves do not live with accessibility features day in and day out, make decisions on captioning and description techniques, they too are advancing an opinion. With only the most unusual exceptions (e.g., the preference for real-time captioning for live events over teleprompter/electronic-newsroom live-display captioning), what broadcasters give us is whatever they consider good enough.

But on what evidence?

Over the last 20 years, I have learned the hard way that a field considered picayune in the grand scheme of the broadcasting industry – captioning – evokes passionate disagreements, and in many cases outright flames. My own contributions to that discord are largely a matter of public record. It’s time for all of us to grow up. Providing captioning and description based on opinion rather than fact, research, or evidence is an idea whose time has long since gone – if it ever was a good idea to begin with.

From now on, whenever a provider claims their captions or descriptions are better than the competition’s or are good enough, our response should be: “Prove it.” When broadcasters use specific techniques or hire certain service providers to deliver accessible programming, our response to their actions should be: “Back it up.” When the Commission approves the accessibility promises of broadcasters, our response should be: “Justify it.”

We now have the capacity to do just that.

Research and evidence

Rather unbeknownst to service providers, broadcasters, and the Commission, there’s already a reasonably large corpus of research on issues concerning captioning and audio description. (I have the largest known clipping file in Canada – three linear feet of reports, papers, and studies.) We have quite a bit of information available to us right now that can settle disputes about the right and wrong ways to do things – and of those two, currently it is a more pressing matter to rule out the wrong ways.

The research is not, however, complete, and in many cases leaves room for too many ambiguities, or loopholes, that service providers and broadcasters uninterested in high-quality captioning and description can take advantage of.

Existing standardization efforts

There simply aren’t any credible standardization documents anywhere in the fields of captioning and audio description. That is not an exaggerated claim. There are no standards manuals that meet the criteria necessary for practical use:

  1. The manual is based on available research and evidence and de-emphasizes feeling and opinion.
  2. Where such research and evidence were lacking or ambiguous, the authors of the manual engaged or sponsored the needed research.
  3. The recommended practices were tested with a sufficient range of audiences.
  4. The manual provides detailed recommendations for a wide enough range of programming types, situations, and exceptions that it is practical for real-world use.
  5. The recommendations are customized for language, technology, and related factors that unavoidably limit what is and is not possible.

That list of requirements will strike many readers as news. If it seems a high standard to meet, it is, but responsible readers will concur that each of those requirements is justified and necessary.

Current standardization documents fall into definable categories:

The three categories are actually variations of each other, and none is sufficient to defend any given method of providing captioning or description.

Defenses and rationalizations

As it stands now, service providers and broadcasters tend to defend or rationalize their practices by asserting one or more of the following:

That’s how we do things here.
Everyone else says the same thing about their own work. Can you prove it’s better than any other way, let alone that it’s the best way?
It’s good enough.
Do you apply the same standards to picture and sound quality and other production values?
I can read it. I can understand it.
  1. Are your personal reactions a relevant criterion?
  2. Are you a person with a disability, or do you have actual expertise and experience in evaluating captioning or description?
  3. Have you attempted to actually sit down and watch several such programs using the accessibility features, and compared the program to similar programs with different features?
We haven’t had a complaint.
  1. Absence of complaints is no evidence of success. The absence of a negative does not prove an affirmative.
  2. How is it even possible for people with disabilities to file complaints any more detailed than “You said it would be captioned but it wasn’t”? Indeed, a Canadian Hard of Hearing Association survey (HTML) states:

    The weakest awareness levels emerge with knowing who to contact when experiencing problems with captioned TV. This could potentially provide one of several explanations behind the low incidence of complaints being placed by those experiencing problems with captioned TV.

    Also in this study:

    • 69% of CHHA members report having “experienced problems with captioned TV” (nonmembers: 17%).
    • 40% of CHHA members (nonmembers: 25%) “are unable to identify where they would direct a complaint regarding captioned TV.”
    • “A significant proportion of [nonmembers] (25%) and CHHA members (40%) are unable to identify where they would place a complaint regarding captioned TV.”
  3. Should people with disabilities be expected to police captioning and description when nondisabled people have no similar duty?
  4. Why should this system be predicated on the filing of complaints, which are post facto and adversarial?
  5. This is a complaint.

The Commission’s hands are hardly clean in this regard. For two decades, the CRTC has demonstrated surprising ignorance and misunderstanding of captioning and description, to the point where:

  1. intervenors have had to explain the differences among captioning types
  2. interventions on captioning are ignored
  3. broadcaster commitments are rubber-stamped
  4. no evidence has ever been provided that CRTC staff actually watch TV, let alone watch TV with captions and/or descriptions
  5. factual inaccuracies are accepted at face value, apparently because they are filed on broadcaster letterhead
  6. staff fixate on details (e.g., the lack of absolute verbatim accuracy in real-time captioning) that, while legitimate unto themselves, are no substitute for detailed understanding of accessibility as a whole
  7. the Commission cannot even get the terminology right when dealing with audio description, claiming that audio description and descriptive video are different
  8. obstacles and policy issues that a responsible observer could see on the horizon (e.g., delivery of audio description on simulcast U.S. programs) are ignored until a crisis forces a complaint to be filed

The Open & Closed Project’s solution

The foregoing provides a general airing of largely undiscussed issues in captioning and audio description for television.

The question then becomes: How are we going to fix it?

The Open & Closed Project has a plan. It won’t produce immediate results, but if broadcasters and the Commission do not support the Project at this critical early stage, the Project stands a good chance of foundering, at tremendous cost to television viewers with disabilities.


The Project adopts a broad definition of accessibility: Accommodating characteristics a person cannot change or cannot change easily.

Under this definition, a deaf or hard-of-hearing person cannot stop being deaf when confronted by a TV soundtrack; a blind or visually-impaired person cannot cease to be blind when confronted by television images; and a person who does not understand the source language cannot learn it in time to understand a program he or she may wish to watch.

Viewers who do not speak the source language, while usually not disabled, have requirements that nonetheless constitute accessibility. Moreover, this form of language accessibility overlaps with disabled accessibility: Subtitled and/or dubbed programs are captioned and/or described. In fact, all four variations – subtitled program with captions, or the same with descriptions; dubbed program with captions, or the same with descriptions – have real-world examples already.

For all these reasons, then, we define the four fields of audiovisual accessibility to include captioning, audio description, subtitling, and dubbing. This intervention focus on the first two.

Standardized practices

The chief goal of the Open & Closed Project is the creation of the first-ever unified set of recommended practices for the four fields of audiovisual accessibility.

At the end of our project (likely four or five years after launch), we will have developed and written one single manual in each of the captioning, description, subtitling, and dubbing fields. Here, “manual” will probably mean:

We want everyone to understand the ramifications of this plan. At the end of the Project, there will be one manual on how to do captioning, one on audio description, one on subtitling, one on dubbing. We can assert that these manuals will be unique in their fields because they will be developed out of existing research and scholarship and with brand-new research the Open & Closed Project conducts or sponsors. Throughout the Project, evidence and fact will preferred over feeling and opinion. The manuals will be exhaustive – enormously more thorough than existing documents. You’ll actually be able to use these manuals in your work. You’ll be able to learn from them, use them as a reference, and find researched solutions to myriad problems.

We are children of the Internet and understand that closed development processes simply do not cut it in the 21st century. The Open & Closed Project will go to unusual lengths to conduct its research and consultations in public. Through online consultations, face-to-face meetings, and individual contacts, the Project will encourage and permit the participation of all interested parties. And we really do mean all: We want all “stakeholders,” as they are often called, involved in development.

However, because this is a new and thorough research project, there is no service provider currently operating whose procedures will meet the standards articulated in the manuals; there will be differences from current practice.

Training and certification

Once the manuals are published, it will then become possible to train practitioners. Training is a huge lacuna in current accessibility provision. Every service provider trains its own staff. Captioners, describers, subtitlers, and dubbing artists are essentially self-schooled or home-schooled, where the home in question is whatever provider they work for.

While it is possible to earn a diploma in court reporting (even with a specialization in real-time captioning), and while a couple of subtitling courses are offered in Europe, and while it is quite possible to earn a degree in translation, there simply is no post-secondary institution or private-sector business where you can enrol in a course to learn how to caption, subtitle, or create dubbing or description tracks.

In part, that’s due to a lack of curriculum: What would those courses teach? Inevitably, at present, the answer would be Yet Another Different Way of Doing Things. The Open & Closed Project’s recommended practices will have the added benefit of providing the first courseware for a standardized training program. We intend to aggressively develop training courses for learners of all ages who wish to work in the four accessibility fields. We will of course retrain existing practitioners if they so desire. Training could be delivered online and in partnership with a range of educational institutions worldwide.

Once the trainees graduate, we can envisage a certification process. The model of the Registry of Graphic Designers of Ontario (portfolio evaluation plus written test plus personal interview) seems reasonable. Learners who clear this stage will then be able to deem themselves trained and certified practitioners of the Open & Closed Project’s methods.

At that point, it will become possible for broadcasters, producers, regulators, and other responsible parties to insist that all accessibility provisions meet the Open & Closed Project’s specifications; those parties can insist on Open & Closed Project–certified practitioners.

“Legacy” accessibility

By then, the accessibility industry will boast a coterie of practitioners certified in techniques that were developed through an exhaustive open process of evidence-gathering and research. We can expect an immediate improvement in quality of captioning, audio description, subtitling, and dubbing, and an immediate payoff for viewers of those services, who will no longer have to relearn how to watch programming with access features.

Needless to say, “legacy” accessibility provisions will still be around. It’s still possible to watch 1980s-era captioning on television, for example. The entire landscape will not have been standardized. But as technology changes (in particular, during the transition from analogue to digital television), old programming will likely be recaptioned, redescribed, resubtitled, and/or redubbed. (By that time, those adaptation processes won’t have to start from scratch due to the Open & Closed Project’s efforts to create a universal file-exchange format, to be discussed shortly.) As programs are updated to new digital standards, so can their accessibility provisions be updated to Open & Closed Project standards.

Infrastructure development

Standardized practices are not the only development goal the Open & Closed Project has. We’ll also develop necessary infrastructure.

Of immediate interest are our plans for the creation of a first-ever exchange format for accessibility files. After nearly 30 years of closed captioning, a decade of description, and a century of subtitling and nearly as long a history of dubbing, it is still impossible to transfer caption, description, subtitle, and dubbing files from one system to another.

The Open & Closed Project’s Accessibility Exchange Format™ (.xex™) aims to solve the complete problem. The Format will be an XML (extensible markup language) document-type definition that will permit files to be exchanged from system to system with all manner of information kept intact – not just the raw text of captions and subtitles, for example, but timing, placement, case, spelling, and version information, to give but a few examples. We emphasize that .xex will solve the actual problem of file transfer and will not be a piecemeal effort.

The .xex format:

  1. will be the property of the Open & Closed Project but will be freely licensable to all
  2. will be painstakingly designed to work with existing XML(-like) or related interchange file formats, of which we know of no fewer than 13
  3. will be compatible with existing attempts to solve a limited subsection of the problem (like the World Wide Web Consortium’s Timed Text Working Group and the Canadian Satellite Users Association’s caption-exchange project), and in fact, our development process will accelerate theirs
  4. will be built from the ground up in an open consultation process, as with other Open & Closed Project activities


  1. Save money. You can recover caption or description files from previous tapes and recordings.
  2. Save time. If properly handled, the files will contain full text (in unambiguous character encoding), position codes, and timecodes, to use the captioning example.
  3. Put an end to format wars. As with service providers, everyone says their software is the best. It won’t really matter anymore: Everyone will be able to exchange files.
  4. Protect your investment. You’ll eventually have to switch or upgrade your systems. Your old files will still be usable.

Other infrastructure projects are planned and will be announced as circumstances warrant.

Applicability of the Project to license renewals

The Open & Closed Project has a small window of opportunity – unreasonably small, given the hugely accelerated intervention deadline for dozens of complex broadcasting licenses – to demonstrate that broadcaster claims are either unproven or are at variance with the existing evidence. We also need to demonstrate that current practices are unsupported by evidence and research. Yet further assertions and activities could be verified or disproved with further research.

An educational opportunity

We urge broadcasters and the Commission to take advantage of the scholarship and information that this intervention, and the Project itself, will provide. Our goal is full accessibility that meets high standards. We have a large reserve of evidence, and plans to fill the holes in that evidence, that will bring us to that goal.

When this intervention points out inaccuracies and flaws in broadcasters’ contentions, we do so under the assumption that broadcasters don’t know any better. We do not impute ill motive; we assume a lack of education, which we are here to remedy.

A New Deal for accessibility?

Given all the foregoing, we have another way of looking at things:

Maybe we shouldn’t expect broadcasters to care much about, or learn everything there is to know about, captioning and description. Broadcasters are interested in profits (or at least cost reduction) first, programming second, and perhaps a few other issues here and there.

Maybe accessibility should be left to the experts. Quite possibly, broadcasters have better things to do than become experts in these fields. Perhaps they should simply accept independent expert advice that is based on research and evidence.

This intervention may thus be the beginnings of a New Deal for broadcasters, the Commission, and viewers who rely on accessibility features.

Adequacy of the CAB captioning manual

The Canadian Association of Broadcasters publication Closed Captioning Standards and Protocol for Canadian English-Language Broadcasters is a valid first effort in standards- and evidence-based captioning manuals, but it is not adequate. I edited a latter-day draft of the manual (apparently the very last draft) and can speak authoritatively.

A full enunciation will be provided at the hearing, but the manual’s weaknesses include:

Issues and discussion


A note, first of all, on the format of CRTC interventions, which are unnecessarily adversarial. Broadcasters make certain points, which are apt to be believed by the Commission. It is up to intervenors, who never have the money or staff that broadcasters and the Commission themselves do, to counter broadcaster claims.

Under this structure, even pointing out a difference of opinion, let alone a broadcaster claim that is at variance with existing evidence, takes on an air of j’accuse. In such a proceeding, it is difficult to disagree mildly.

I can only counsel readers not to take things personally and not to overreact. The assumption here is that broadcasters don’t know any better. The Open & Closed Project emphasizes evidence and research in its development processes, and to the extent possible, those are emphasized in this discussion.

Exclusions from description

Some broadcasters in this proceeding claim that their program types, or their complete networks, are either not suitable for description or are not a priority for it.


in both the sports and news genres, because much of the programming is play-by-play, discussion and analysis, description is, in fact, present.

The Score (emphasis added):

The Score has been in touch with NBRS and has described [sic] how the Score serves the visually-impaired. The nature of the vast majority of the programming on the Score does not lend itself to a standard approach.... Play-by-play announcers and colour commentators provide what the visually-impaired viewer requires to follow the action.

Both claims are unsupportable. It is unwise to take the opinions of one service provider as solid evidence.

Sports and news programming are amenable to description. In sports, the differences between play-by-play and audio description have been documented for nearly two years. I’ve already had a discussion with WTSN, the only digital specialty service currently on the air that has a description requirement, and have tried to discuss the issue with TSN. Live audio description, for example, has already been done (it’s the norm in theatrical description); what we need are standards and practices for carrying it out.

It is not the case that sportscasters are telling blind viewers as much as they need to know. Sportscasters assume their audience can see what’s happening. (Keep in mind that other broadcasters in other proceedings have argued that sport doesn’t need to be captioned because it is so visual.)

CTV claims: “In the case of sport services, our on-air commentators have been trained in this regard and much of the programming calls for descriptions to be constant.” The claim is unverified by my experience. On TSN, I have never witnessed anything remotely resembling enunciation of the full onscreen text (not just bits and pieces of it), let alone live description of action that cannot be understood from the main soundtrack alone.

In news programs, CTV must surely be aware that the failure to voice onscreen text is an ongoing irritant to blind and visually-impaired viewers. (Canonical examples are phone numbers, names of speakers and guests, and weather warnings.) CTV’s claim that newscasters, “in practicing their craft,” provide description is also unsupported.

CTV later provides the contradictory statement that, with TalkTV, “the opportunity also exists to train the on-air talent to consider the needs of the visually-impaired... and we will endeavour to increase the awareness of the needs of the visually-impaired.”

CTV’s intent to convene “focus groups to define specifically which program formats require DVS [sic]” should be carried out by an independent body like the Open & Closed Project. As written, it sounds like a plan for a broadcaster to ask a roomful of blind people which programs the broadcaster does not need to make accessible.

In any event, we already have sufficient evidence that all programming genres require description and that blind and visually-impaired viewers want them all described. See, for example:

  1. Packer and Kirchner (1997) found the following preferences for described programming:

    Genre Survey 1 Survey 2
    Dramas or Mysteries 85% 83%
    Nature or science 67% 72%
    News and information 61% 68%
    Comedies 59% 77%
    Music programs or videos 40% 44%
    Sports 37% 26%
    Game shows 26% 41%
    Daytime soap operas 21% 28%
    Talk shows 20% 29%
    Shopping programs 16% 12%
    Children’s programs 15% 38%
    Other 14% 11%

    Note the strong desire for description of “news and information” and “sports.”

  2. In the U.K., the Independent Television Commission has refused to exempt programming categories from description requirements:

    The ITC issued its Code in February 1997. For... audio description, 2% by the start of the second year rising by 2% every two years. The Commission decided not to exclude any categories of programming from the minimum requirements.

    In addition:

    Excluded programs: The legislation allows the ITC to specify certain classes of programs which are to be excluded from these requirements altogether, or to which reduced requirements should apply. The ITC has decided, however, not to make any blanket exclusions or reductions of this kind for any of the three types of provision.

    The ITC goes on to note that:

    The ITC nevertheless recognizes that it may not be appropriate for the sign-language or audio-description requirements to apply to some digital program services which are of a themed nature (i.e., consisting of only a very limited range of program types). The ITC will therefore consider, on a case-by-case basis, requests from licensees providing themed services for exemption for those services from these requirements. The ITC would only accept such a request where it was persuaded that the application of the requirements would be inappropriate, and following consultation with representatives of those deaf or blind or partially-sighted people for whom the provision is intended.

    At present, the ITC does not intend to use its discretion to apply reduced targets to “excluded” programs, but will review the situation in the light of experience with any themed services which are exempted in the way described above.

    TalkTV suggests that its entire service should be exempt from description requirements, except inasmuch as it needs to convene a focus group to authorize such exemption, and except inasmuch as its programming is already self-described, and except inasmuch as its on-air talent needs to be trained to provide description.

    TalkTV does not note that even if certain programs might not benefit hugely from description, other programs on that same network might. A broadcaster- or genre-wide exemption is unwarranted for any broadcaster or genre in the current proceeding.

CTV also states:

[I]n the case of our news-based services, including Newsnet and ROBTV, our policy is to attempt to ensure that wherever there are visuals, such as charts or text, these are also voiced over.

My viewing experience fails to bear out CTV’s claims that voiceovers actually happen.

Note that CTV’s policy is merely to attempt to ensure. Moreover, there is no known research on how to turn “charts” into speech. It may be impossible. We draw charts in the first place because the relationships they embody resist encapsulation in words. Some parallels with Web accessibility and adaptation of books into Braille and talking book can be drawn, but the upshot here is:

While audio-description exemptions for genres or networks should not be given, research on these issues is in order.

Satellite distribution of audio description

Numerous broadcasters – including CTV (TalkTV, ROBTV, Newsnet) and Alliance Atlantis (HGTV, History) – claim that it is impossible or unworkable to deliver digital specialty channels with audio description.

The claim is unsupported. In 2002, I wrote a consultant’s report for Bell ExpressVu listing ways in which a direct-to-home satellite service can accommodate visually-impaired viewers. One method involves the use of virtual channels with open description. (Virtual channels were ExpressVu’s idea.)

The system works as follows.

Bell ExpressVu has implemented the system for numerous Canadian channels with description. It’s happening now. This example proves description is possible on satellite. CTV’s application for TalkTV states:

In the case of DTH, our experience... has been that Star Choice is not capable of providing DVS [sic] to its subscribers and has not passed through our conventional DVS signals, while ExpressVu, on the other hand, has indeed undertaken to pass through DVS signals on a request basis.

(The latter four words are incorrect or at least ambiguous.)

Alliance Atlantis states (emphasis added):

Through the CSUA and other broadcasting entities it has been determined that the BDUs for the most part are technically not capable of delivering the descriptive video signal to all of our channel’s subscribers at this time. It is imperative that we... not create an imbalance between our subscribers by sending our described signal to only a few BDU’s....

We are actively working within the CSUA on its task force to help find a technically accepted solution so that the whole chain is complete from program service to subscriber.... Therefore, we feel it is inappropriate for us to commit to any level of weekly broadcasting of descriptive video knowing that this effort and expense will not benefit all of our viewers. When this matter is fully resolved, then we would be very willing to commit to a condition of license.

In fact, description as currently delivered in Canada already does not reach “all” subscribers or viewers. There is no country on earth that airs description “all” viewers can watch.

The claim that satellite-delivered channels can provide description only when “all” subscribers can enjoy it raises an impossibly high and unattainable standard that defies current reality and ensures that broadcasters will never have to provide description. Also, we can’t wait seven long years – until the next license renewals – to impose description requirements. We don’t have that kind of time.

We already know it is possible to deliver described programming via satellite. It’s happening already.

The ExpressVu technique costs bandwidth, but then again, so does everything in satellite broadcasting. After a certain number of virtual channels has been set up, it may make sense to retire the virtual channels and instead use the alternate language tracks that certain DTH systems make possible. The issue there is the inaccessibility of the onscreen menu systems.

The Open & Closed Project proposes to carry out independent research on:

We suggest a funded one-year development process, resulting in a report that is publicly released in accessible formats. Contents of the report could be detailed later, but at a minimum the report must explain the current technology and limitations and set out a range of attainable options to solve the inaccessibility problem. An implementation plan will be included.

Canada already has a salutary history of overcoming technological barriers to delivering described TV programming. There’s no reason why the Open & Closed Project’s independent research cannot solve the full problem.

Note that the solution may have corollary benefits worldwide and, indeed, may be exportable. (For one thing, solving the problem of inaccessible visual menu systems will require the cooperation of U.S. and foreign DTV developers. If we get it working for Canada – in two languages, no less – we get it working for everyone.)

This proposal is consonant with broadcaster submissions. CTV (TalkTV):

CTV submits that some time is required, as well as a phase-in period, before specialty services, in cooperation with distributors, can offer DVS [sic] signals. We look forward to working with the industry in this regard.

So let’s get working.

We would emphasize that description on satellite-delivered channels could be postponed (e.g., for 18 months) until a tolerable technical solution is found (we already have one that’s up and running), but under no circumstances must the CRTC fail to pass a description requirement in this round of renewals.

I am a pioneer in the field of captioning of music videos. My guest editorial in Billboard back in 1989 introduced, promoted, and essentially authorized the captioning of music videos for the entire recording industry. (I built on the experience of two other leaders in the field, record producer Ed Stasium and Donna Horn of the Caption Center.)

Music-video networks in Canada have evaded their responsibilities to provide captioning for 14 years. Their arguments boil down to the complaint that “Labels don’t provide us with captioned videos.” While this is certainly true – after all this time, it’s still difficult to get captioned U.S. videos in captioned form here in Canada – as a justification for inaccessibility it does not stand up to scrutiny. Under the Broadcasting Act (§3[1][h]), broadcasters have responsibility for programming, not producers or record labels: “all persons who are licensed to carry on broadcasting undertakings have a responsibility for the programs they broadcast.”

Essentially, broadcasters say “Our hands are tied. What do you expect us to do? Caption them ourselves?” Yes, in fact. Over the years, the Commission, evidently not taking music videos seriously, has accepted broadcasters’ claims wholesale, despite repeated interventions laying out the facts.

The piecemeal approach to music-video captioning in use for 14 years really means:

Had the CRTC asserted its role as a regulator 14 years ago, today captioning viewers would enjoy continuous captioning of music videos on networks. The entire problem of uncaptioned videos would have been solved. For 14 years, the Commission has made it clear that it believes access to music-video programming is something deaf and hard-of-hearing Canadians, and other captioning viewers, are not legally entitled to have.

Now broadcasters propose even further rationalizations for refusing to provide accessible music-video networks. MuchMoreMusic recapitulates the CRTC’s early mistake and writes “the Commission has stated that it would not be appropriate to apply its general approach for English-language services to MuchMoreMusic.”

On what evidence? The facts are available:

It is noted that certain live programs and, after lobbying, VideoFact-funded videos are captioned on MuchMoreMusic. (I have documented examples of recent VideoFact-funded videos without captions.) MusiMax claims that MaxFact-funded videos are now captioned.

MusiMax makes the claim (paraphrased into English) that there are complex copyright questions involved in captioning uncaptioned videos. The evidence contradicts this claim.

MusiMax claims that the issue of adding captions to uncaptioned videos “raises complex questions of copyright that remain, at present, unanswered” (my translation). What MusiMax claims is unresolved they are already doing. What MusiMax claims it cannot do its corporate partners – MuchMusic, CITY-TV – are already doing.

In a proceeding such as this, it is not enough to claim that a vague application of copyright precludes legal requirements for accessibility. The Charter’s provisions trump those of the Copyright Act. Broadcasters’ responsibility to avoid contravening §15 of the Charter, with its prohibitions against discrimination, supersedes any claim that the Copyright Act forbids the mere transcription and standards-based onscreen textual presentation of lyrics, music, and non-speech information.

MusiMax further muddies the waters by citing «questions de traduction de droits d’auteur,» perhaps meaning “transformation,” “enlargement,” or “interpretation,” but suggesting that the issue involves translating English-language lyrics into French. MusiMax also claims that producers hold copyright, a fact that is not true. (Be careful with the term “producer” when discussing recorded music.)

In addition, MusiMax assumes copyright holders (a) care about post-facto captioning and (b) aren’t already doing it themselves. If a Canadian broadcaster recaptions a video that was already captioned in the U.S. but never made it to Canada with captioning, what does the rightsholder care? What does the rightsholder care at all?

The evidence argues persuasively against MusiMax’s copyright contentions. The conclusion is applicable to all music-video services: They should in no way be exempt from CRTC captioning requirements during the full broadcast day.

More discussion on this topic will be engaged at the hearing.

Music-video “presentation” segments

MuchMoreMusic commits to captioning “at least 90% of all non-music programming... including presentations by program hosts.”

Some questions come to mind:

  1. Do all parties realize MuchMoreMusic’s statement means that throws and VJ patter must be captioned, but videos must not? This situation is like the dilemma of newsroom captioning in reverse: Live segments are real-time captioned, but taped segments (the videos) may not be.
  2. Is the Commission willing to compound error after error by authorizing 90% captioning of everything but the most important part of a music-video service?

I urge the Commission to correct its previous mistakes and require captioning of music videos on music-video networks.

Audio description on music-video channels

“Given the nature of MuchMoreMusic as a music-video specialty service,” it writes, “CHUM believes that a guaranteed minimum number of hours of described programming would not be appropriate.”

Guaranteed minima are in use in the U.K., the U.S. (before the FCC lost its case), and Canada. No evidence has been tendered that MuchMoreMusic’s documentary and interview programming and series could not be described, among other items. No genre or network should be exempt from description requirements.

Live French captioning

On the topic of live French-language captioning, broadcasters continue to make submissions to the Commission that contradict existing evidence. MusiMax claims there is a lack of an efficient, affordable technology to provide quality closed captions for live television (my translation). The claim is unsupported.

There are already two adaptations of English-language stenography for French-language captioning.

It must be reiterated that both systems use English stenographic equipment. They can be viewed as different software platforms on similar hardware; Canadian broadcasters have no fear of different software in offline captioning, but in French real-time captioning, broadcasters consistently maintain the Radio-Canada system is all there is. The position is unsupportable.

On the issue of human resources, broadcasters correctly state that not enough trained real-time reporters are available. And for how many years have broadcasters been making this claim? Seven? How many real-time writers could have been trained in seven years?

Why is it that there appears to be more live Spanish-language captioning in the U.S., where Spanish isn’t even an official language, than there is live French-language captioning in officially-bilingual Canada?

Voice recognition

Broadcasters’ claim that French captioning will have to wait until new voice-recognition technology is chosen, let alone perfected, contradicts available fact.

Here we see the perils of industry-centred accessibility research. Armed not with the facts (let alone linguistic or technical knowledge), Canadian broadcasters attempt to reinvent the wheel using unproven vapourware technologies.

It is incumbent on the Commission to rein in such speculative development. Money and time should be focused on improving a system that actually works – stenography – instead of guaranteeing years of uncaptioned live programming during development of a competing system that no evidence shows will actually work.

Use of scrollup captions

In the current proceeding, OLN, Treehouse, Prime, and the Comedy Network all use scrollup captioning for fictional narrative programming. Such a practice is not supported by standards and is difficult, bordering on impossible, to follow. It sure is cheap, though.

Here we recall license interventions of years past, when the Commission statutorily charged with regulating accessibility in the broadcasting sector had to be painstakingly shown the difference between real-time and live-display captioning for newscasts.

The Comedy Network submission makes the claim that “To ensure the quality, reliability and accuracy of the closed-captioning, we use only the highest-quality captioning services.” In fact, no evidence has been provided that CTV uses other than the cheapest offline captioning services.

Fictional narrative programming on, for example, the Comedy Network is captioned using the scrollup method. What’s wrong with that?

The captioning of daytime soap operas in the U.S. added a new wrinkle to the use of scrollup captions. The earliest captioned soap operas looked like every other dramatic program, with pop-on captions. Later, turnaround times grew ever closer to airdates, and, I am reliably told, producers began to resent the added cost of pop-on captions. The inference I draw is: “Sure, we’ve got to caption it, but it’s only a soap opera, for heaven’s sakes!”

At present, then, the only prerecorded fictional narrative programming that is captioned with scrollup on the major U.S. networks is the soap opera. (I don’t know of a single soap opera captioned with pop-ons. Even Spanish-language telenovelas are captioned with scrollup.)

I suppose it could be argued that the use of scrollup for this kind of fictional narrative programming authorizes its use for all other kinds. That is not the case.

Scrollup not authorized by existing style manuals

No style manual in my possession authorizes the use of scrollup captions for fictional narrative programming.

Based purely on the existing evidence, the use of scrollup captions for fictional narrative programs should be prohibited in all cases save for emergencies.

Quick-turnaround programs

I have discussed the issue of turnaround time with captioning staff at various networks. An employee of a network not party to the current proceeding explained that scrollup captions were necessary because there isn’t enough time before airdate to create pop-ons.

I question this blanket statement. Even in the case of U.S. soap operas, in many examples producers use scrollup captions purely to save money and not because there isn’t enough leadtime. In any event, an undiscussed issue is the lack of 24-hour-a-day, competent, standards-compliant captioning facilities in Toronto and Montreal, home to the applicants in the current proceeding.

In Los Angeles, it is perfectly possible to deliver a tape in the late afternoon and receive an encoded sub-master the next day – with pop-on captions. Same-day turnarounds are not uncommon at all, particularly for shows like The Simpsons that are edited right down to the wire. U.S. captioners have various efficiency mechanisms that Canadian captioners are presumably unaware of.

In the wake of the Vlug decision, CBC has been required to caption everything it airs, save for outside commercials. Though CBC is in noncompliance with its 100% requirement, it still manages to caption programming close to deadline, often using pop-up captions.

Cheaper, not better

The question of turnaround time is a canard in many cases. I know from conversations with various broadcasting undertakings that, of the list of reasons why scrollup captions are used, cheapness is number one. It’s all about saving money.

The CRTC has abetted this approach. For two decades, it has predicated captioning requirements on claimed availability of funding, and has been unworkably vague about captioning quality. It was a modest breakthrough when the Commission finally figured out the difference between live-display and real-time captioning. But in the last five years, the use of scrollup captioning for fictional narrative programming has exploded, and the primary reason is cheapness, not turnaround.

Viewer-hostile captioning

I can and will provide evidence at the hearing that fictional narrative programs captioned via scrollup are nearly impossible to follow. It will be instructive for broadcasters and the Commission to watch the very same program with scrollup and pop-on captions. I can make a believer out of anyone.

Why are scrollup captions hostile to the viewer when used for fictional narrative programming?

Overturning captioning history

One ought to proceed with caution in claiming that certain captioning conventions should continue to be used simply because that’s the way we’ve always done things. But when the proposed alternative is worse, we can indeed decide that the conventional practice became conventional for a reason. The new method provides the contrast we needed to recognize the old methods as superior.

In this case, the method used for decades to caption fictional productions – pop-on captioning – is indisputably superior. The use of scrollup captions for fictional narrative programming must be terminated in all but uncommon and limited cases.

Emergency captioning

An exception clearly exists: The program really and indisputably does arrive too late to produce pop-on captions.

In such a case only, it is certainly true that some captioning is better than none. A broadcaster who truly cared about caption quality would take the following steps:

  1. Caption the emergent program in scrollup. Real-time captioning could be used only on the very latest of late-delivered programs (and not because it is even cheaper than offline scrollup).
  2. Make sure the program is never repeated on another day with scrollup captioning. Either get the original pop-on-captioned tape (as in the case of U.S. productions) or recaption with pop-on.

Emergencies are inconvenient for everyone. If the Commission and broadcasters are truly committed to caption quality, they will use scrollup captioning for fictional programming only as a last resort and will remedy the problem for that program by recaptioning in pop-on for later airings.

Recaptioning pop-up-captioned programming

The Prime network in particular is guilty of using scrollup captioning for programs that were already captioned with pop-on captions. MASH is an example, given that the Caption Center spent months captioning all 255 episodes as far back as 1993. In the intervening years, I occasionally saw an uncaptioned MASH episode, but only on Prime have I ever seen episodes recaptioned in scrollup, a waste of resources and a serious decrement in captioning quality.

Claimed adequacy of in-house captioning offices

Broadcasters would have us believe that their in-house captioning offices do excellent work. The claim is unsupportable; evidence will be tendered at the hearing.

Captioning of preschool programming

Treehouse claims its existing exemption, under which it is not required to caption programming “targeted to preschoolers,” should remain in place “since preschoolers are generally not literate.”

But older kids and preschoolers’ parents are literate. The exemption should be lifted. Further details of the intricacies of children’s captioning will be presented at the hearing.

Rerunning of real-time captions

TalkTV is among the Canadian stations that captions a program on first airing using real-time stenography and simply reruns that same tape for later airings. The result is the halting, error-prone, piecewise presentation found in the original airing even though the time pressure that forced the use of real-time captioning has been removed.

Broadcasters interested in quality captioning do the following:

The advantages?

Yes, it costs more, but not much more. U.S. newscasts use this approach for Eastern and Pacific time-zone broadcasts, for example. It’s even possible to reuse real-time captions for later pop-up captioning.

TalkTV has also not demonstrated that it is limiting the use of real-time captioning to live or near-live programming. Offline scrollup or pop-on captioning could and should be used in those cases.

Remedies and recommendations

The remedies and recommendations advanced by this intervention include:

Audio description

  1. All broadcasters receive a requirement to air described programming. No broadcasters or program genres are exempted.
  2. For satellite-based services, broadcasters to support the Open & Closed Project’s research efforts to identify real-world solutions, after the manner of the Bell ExpressVu example.
  3. Broadcasters’ description requirements to become active after the report is issued. The Commission could hold another hearing to specify and actuate the description requirements, but it should not fail to issue description requirements in the current proceeding.
  4. Support the Open & Closed Project’s research and develop efforts in live description, particularly of news and information and sports programming.


  1. Cessation of the use of scrollup captioning for fictional narrative programming within 30 days of renewal. The method can still be used for first airings of such programming in emergency cases or cases of genuinely short leadtimes, but repeats must use pop-on captioning.
  2. Music-video networks must begin to caption the entirety of their broadcast output irrespective of whether music videos arrive at the station originally captioned or not.
  3. In French-language real-time captioning, emphasize the existing stenographic systems. Do not put off captioning in French until a claimed voice-recognition system is developed. Train new stenographers first.
  4. Reruns of programming originally captioned using real-time stenography must use cleaned-up scrollup or pop-on captions.

Support for the Open & Closed Project

  1. Broadcasters to support the Open & Closed Project’s efforts to develop independent standards and infrastructure based on research and evidence.