[jerf]

That's definitely the biggest one I was thinking of. QWERTY v. Dvorak has gone in many strange directions, but if you look for the actual science, it is at best quite split, and at worst, all the relevant studies were done by people with very strong preexisting opinions, leaving us with not much to go on. (I include Dvorak's own studies here, for the sake of argument.)

But the claims that Dvorak has no advantages over QWERTY really don't pass the smell test. It's almost certainly more a matter of whether it's worth it for someone to switch, on which I'm very ambivalent.

There's also the interesting question of whether it would be better to start on Dvorak, which I'll have to ponder here in the future now that I have a baby. Personal experience would suggest that someone raised on Dvorak is much more likely to learn actual touch typing.

(I was on QWERTY for over a decade and still doing the same wandering-hands thing everybody else does, because QWERTY doesn't reward touch typing. Touch typing on QWERTY is like the official way to swing a baseball bat; everybody has to learn it, but hardly anybody does it and even at the pro level everybody does their own thing. Dvorak and most of the other alternate layouts reward it very strongly. You don't even have to teach it, it just happens.)

[silentbicycle]

> It's almost certainly more a matter of whether it's worth it for someone to switch[.]

Agreed. For programmers, I would say the odds are good - you'll be doing a lot of typing, and RSI is a real concern. For people who do a lot of typing on other peoples' computers, planning around what's widely available (qwerty & vi on Unix, for example) is probably a better choice.

> QWERTY doesn't reward touch typing.

Good observation! Dvorak seems more clearly designed with touch-typing in mind, I think.

(As a data point: I type "wandering-hands" on Qwerty at 95-105 wpm, and about same on Dvorak. The typing isn't the bottleneck.)