I needn’t belabour the tenaciously-held belief by informed people that a text-only Web page isn’t an “accessible” one. A page, separated from its parent like a calf at a slaughterhouse, that contains no graphics and is nothing but unstyled text barely counts as a Web page in the first place. (I suppose that rules out three-quarters of what I’ve published on the Web. But I’m not making separate text-only pages.)

But now we have a doozy of a research paper by Giorgio Brajnik et al., “Do text transcoders improve usability for disabled users?” (PDF).

The authors did a small usability test and a big one. In the small test, 11 people were given an online shopping task. They used the real site and a text-only version of the site that had been run through a “transcoder.” It turned out that the subjects’ highly variable basic competence at using the Web and at using their adaptive technology made the results unreliable. (If you cannot find a Rolling Stones album because it’s listed under “the Rolling Stones,” should you even be allowed on the Web?)

In the larger test, 17 blind, seven low-vision, and five dexterity-impaired people had to complete tasks on a really bad Italian government site and on its transcoded text-only version. The results are, I should say, very densely presented (in part using a rather poorly designed table), but from what I can gather:

  1. The transcoded site improved the scores for the evaluation questions “I easily found the required information” (more commonly answered) and “I was tempted to go elsewhere to find the answer” (less commonly answered).
  2. The question “I always knew where I was in the site” was the same for both cases.
  3. Text-only transcoding improved the answers to the question “I’m satisfied with the solution I found” only for simple tasks, not complex ones.
  4. More people finished complex tasks using text-only pages than normal pages. There were other successful outcomes for tasks using text-only.
  5. 70% of subjects preferred text-only pages overall.

Interesting, nu?

Of course, this is not a fair comparison. Of course text-only pages will be preferable to an inaccessible page; the text-only version is less inaccessible. The trick would be to compare really good, standards-compliant sites (some with valid code or the next best thing, others with invalid but still semantic code) against text-only pages. I doubt that subjects would prefer text-only in that case. And a test using learning-disabled subjects would be a near-complete failure for text-only pages, unless they happened to use crazy fonts like brown Comic Sans on pale blue, which is what we keep being told this group wants and needs.

The foregoing posting appeared on Joe Clark’s personal Weblog on 2005.11.08 18:05. This presentation was designed for printing and omits components that make sense only onscreen. (If you are seeing this on a screen, then the page stylesheet was not loaded or not loaded properly.) The permanent link is:

(Values you enter are stored and may be published)



None. I quit.

Copyright © 2004–2024