Ok, Dvorak fans, give me some data that contradicts this to convince me to continue with my Dvorak studies.
Knowing that I really needed to focus on my studies, I instead spent an hour hacking out a Perl script to compute distance and alternation metrics for QWERTY and Dvorak keyboard layouts over text documents.
The experimental setup. I pretend that the only thing the user types with his fingers are alphabet characters. Keys on my keyboard are 0.75" square, with the second row offset 3/16" horizontally from the first, and the third row offset 9/16" from the first.
Distance is measured as the sum of the distances traveled by each finger. (It's as if distance were essentially computed independently for each finger and then summed across all eight fingers.) The alternation ratio is the fraction of key pairs where the keys are typed one by each hand.
The results. I ran the experiment on three texts; here are the results.
| text | document size (# alpha characters) |
qwerty | dvorak | dvorak improvement | |||
|---|---|---|---|---|---|---|---|
| distance per alpha | alternation ratio | distance per alpha | alternation ratio | distance per alpha | alternation ratio | ||
| Qwerty article | 5056 | 0.563 | 49.66% | 0.444 | 65.47% | 21.3% | 15.8% |
| Hamlet Act I Scene 1 | 6322 | 0.537 | 51.36% | 0.416 | 62.72% | 22.4% | 11.36% |
| IBM-Redhat | 4268 | 0.570 | 48.27% | 0.451 | 66.87% | 20.9% | 18.6% |
Conclusions. Your fingers move 20% less (laterally) when typing in Dvorak. That's 20% less motion of your tendons through your carpal tunnel, and of course you might expect to type 25% more keys in an hour if you move your fingers at the same rate. You also alternate hands about 15% more often than you do when typing in Qwerty; this contradicts the Economist article's uncited claim that Qwerty has the upper hand (pun intended) on alternation. Perhaps the most poetically just result is that these statistics hold true of the Economist article itself. :v)
Should you bother learning the Dvorak layout? Perhaps Dvorak typists try to deceive themselves to justify the cost of learning the keyboard layout, but every one I talk to simply prefers the feel of typing in Dvorak. Once you are competent at Dvorak, typing in Qwerty just feels like a ridiculous amount of extra work.