
On Fri, 12 Dec 2003, Taavi Burns wrote:
Beware of placebo effect. *Expecting* that typing on a Dvorak layout will feel better has a strong tendency to *make* it feel better.
There are strong arguments that a Dvorak layout should feel better than qwerty regardless of speed. The most commonly used keys are in the home row, the second most commonly on the top row, and least common on the bottom row. This does indeed reduce the number of times your fingers have to move away from the neutral position. That's gotta be a win, eh?
It seems likely, but (a) sometimes the real world delivers surprises, and (b) even assuming it is a win, that doesn't mean it's a *significant* win. Placebo effect is powerful, and it's the reason why tests of things like new drugs absolutely must be done as "double-blind" tests: neither the patients nor the experimenters know which patients are getting the real pills and which are getting the dummies, until the experiment is over and the sealed envelopes are opened.
I do find it hard to believe that there is NO improvement whatsoever from using a keyboard layout designed for use with electronic interfaces with careful attention to useability versus one that was actually designed to reduce jamming on old mechanical typewriters.
Uh, say what? Dvorak, like Qwerty, was designed for manual typewriters. (And to the extent that it had any significant advantage, it may have been reduction of muscle fatigue in that physically-more-demanding case.) The major experiments hyping its superiority were done during WW2, and the first unbiased trials -- which largely ended interest in it -- were done in the 50s. The only currently-extant keyboard layouts which can be said to have been designed for electronic interfaces are the split keyboards. The mere fact that Dvorak does not assign home or near-home positions to RETURN and BACKSPACE demonstrates clearly that it was not designed for electronic keyboards. Henry Spencer henry-lqW1N6Cllo0sV2N9l4h3zg at public.gmane.org -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml
participants (1)
-
henry-lqW1N6Cllo0sV2N9l4h3zgļ¼ public.gmane.org