Google, who have historically not given a shit about Web standards, now employ two of the three severest standardistas roaming the earth: Ian Hickson (“disclosure”) and T.V. Raman, the latter of whom confirmed his new employ in an E-mail.

The third, Tantek (q.v.), works for a failed concern whose competition is already better.

Go for a clean sweep, Google?

I live in the hood and I don’t know how to pronounce it. (Care? Car? Cur? Probably care.)

(Q.V.)

Most captioning viewers are hearing people. If you are a native speaker of the language being captioned, you sit there and follow the speech and captions simultaneously. You do that for all kinds of programming, including how-to shows, dramas, music videos, porn, or whatever else. The captions appear and disappear (or scroll or paint on) under someone else’s control.

But it turns out that your understanding of instructional programming may actually be impaired by simultaneously reading text and hearing the same words.

A paper by Slava Kalyuga et al, “When redundant on-screen text in multimedia technical instruction can interfere with learning,” conducted several experiments with teenage learners involving simultaneous or sequential presentation of visible and spoken text. Generally, the subjects had to sit there and think harder when the two media were presented simultaneously. Presenting words and speech one after another resulted in better test scores.

The trainees’ control over the pace of instruction could be an important factor influencing our results. For example, [another study] showed that replacing on-screen text with audio narration (the modality effect) was effective only when the pace of instruction was set by the time of the narration and students had no control over the pacing. […]

The instructional format based on an auditory-only presentation of text was significantly more efficient than the concurrent audio and visual presentation format…. These results indicate that a redundancy effect was obtained under conditions that required learners in the concurrent text group to read and listen to the same text simultaneously for a limited time and that required learners in the auditory-only text group to listen to identical auditory explanations for an equal amount of time without reading them. Having the same information presented in two modes simultaneously is less effective than when it is presented in one mode alone under conditions in which the pacing of instruction is controlled by the system.

So:

• Does this same effect apply to watching instructional programs on TV?
• Does it apply to watching TV at all? Do the moving images interfere with the effect?
• How does the effect vary according to program type?
• Does the same thing apply to hard-of-hearing people? (I doubt it applies to totally-deaf people.) How about ESL learners?

A research report from June 2005, entitled “Information now! Enhancing digital access to learning materials for Canadians with perceptual disabilities” and available in several unfriendly formats, covered a trial of a DAISY electronic-talking-book player with 56 “adult learners,” most of them blind.

(DAISY is a structured talking-book format based on XML, and I’d give you a better reference for it if I knew of one that made any sense whatsoever. The online documentation is terrible. DAISY allows you to mark up your original text in structural ways that will be familiar to any standardista. You can optionally link to voice recordings.)

The most surprising part of the study? Many users could not understand the concept of navigating by headings.

Although tasks, such as placing bookmarks or the “Go to page” commands, were rated as fairly simple and helpful, results did show that the more complex navigation options, such as navigating through headings, were more difficult to execute. In particular, participants had trouble distinguishing whether they were navigating by phrase or by heading. […] “Navigation by heading level” proved to be the most time-consuming task for all three groups of participants.

Visually-impaired people took more time to navigate by heading on a hardware DAISY player, while blind people needed more time on a software player (which had a habit of talking at the same time their screen reader did).

Terminology used for describing heading levels

Participants indicated that the tasks which seemed to take the longest to complete, for all disability types and levels of technological skill set, were those which fall under the Navigation of Book category…. [S]ome participants had difficulty grasping the concept of how a book is structured into levels comprised of chapter headings, subsections, paragraphs, and phrases…. Part of the confusion on behalf of users with the navigation structure may rest with the terms or words used to describe the levels. One participant commented on this during her test: “Language is too confusing – e.g., ‘element,’ ‘levels,’ etc.”

I don’t understand this at all. How is the concept of “headings” difficult? Braille and large-print readers will be familiar with it, as will advanced screen-reader users who have used standards-compliant Web sites. In fact, we are continually reminded that Windows screen readers allow you to navigate by heading, that people use those features frequently, and that, as a result, WCAG 2 should require heading elements that make sense when removed and remixed.

Are headings a problem or not?

At Web Essentials ’05: Don’t use address to mark up a postal address, because that’s a mistake people make in reading the spec. Also don’t use definition lists to mark up dialogue, because that’s a mistake within the spec.

The perfect restaurant for the Dungeon Master with a hankering for teff?