Evolving a better keyboard layout
klausler.com
klausler.com
I would recommend in switching that you do NOT move your keycaps around, or draw new letters on your keyboard, or any other mental crutch, but instead simply memorize the new layout mentally. It won't take long, and that small step will hasten your transition.
There were a few interesting side effects however. For the first couple of days, I realized that I could no longer spell some words -- my life is so terminal oriented that much of my rote spelling knowledge was based on muscle memory.
In the first couple of days while feeling really inept at the keyboard, that feeling also carried over into regular voice conversations -- I had to remind myself that I wasn't speaking impaired, only typing impaired.
I easily exceeded my old qwerty performance within the first several days.
During the transition, it was an added challenge to do everything, and that reminded me the previous generation of punch card programmers describing their work. Suddenly it became more efficient to think about the problem harder before trying solutions, rather than just firing rapidly from initial thoughts until I had bludgeoned a solution.
It's amusing to secretly change some unfortunate person's keyboard to dvorak, watch the bewilderment, then type a few sentences, and walk away. "Works fine for me. I don't know what your problem is..."
My default home position has my right hand in the conventional position, but my left hovering over the bottom left-hand corner of the screen. I spend far more time using those meta keys than typing. Ctrl+e and ctrl+y (vim scrolling keys) and alt+tab (window switching); ctrl+pageup/down (browser switching) and ctrl+a (gnu/screen escape key) than I do bulk entry of text. It also puts me closer to the escape key. In Windows I heavily use control characters based on win+r, win+e, and tab and shift-tab. I suspect the reason I have the right hand in the 'correct' position is because the dimple on the keys there allows me to consistently return to a known position in between occasional mouse usage. A small change to the keyboard layout (e.g. trying to use a UK English keyboard when I grew up with US English) is annoying not only because of the different positions of some control characters I use a lot, but also because it changes the feel of the keyboard and gives a different feel when I'm trying to return to a known position. I can't slide along the left shift there to navigate my left hand to the right position.
I suspect I'd have an easier time learning a different combination of alphabetical characters than I'd have adjusting to changes in the shape of the keys. I've not seen a typing survey about 'optimal layout' that considers such things, and given how individual it is I'm not surprised. The only reliable way to get figures out would be to take different populations of people at birth and grow them into different layours. With the entrenchment of qwerty and the tie between applications and particular layours (e.g. navigation keys in vim, popular positioning cut/copy/paste hotkeys in qwerty and the important role those keys play in users moving away from the mouse and to keyboard control) I doubt such a study would be practical.
I wonder if he used source code as the primary input into his algorithm, how the results would change. I can't type on the home row in qwerty, because my right hand is always shifted right about 2 keys, because I use those other keys so often.
Most relevantly, it puts the brackets, parens, and other programming symbols as primary on the number keys, and the digits themselves as the shift-modified versions.
Seems like it'd make lisp/scheme/etc a good amt easier to type :-)
I'm up for doing a conversion experiment on myself, once I figure out what my new work keyboard's going to be... And if they're going to stop me from swapping it for a kinesis.
What do people here think? Is it worth the time investment to switch?
I've used Dvorak for over two years now, and in my experience:
1. If you spend a lot of time typing on computers you have control over, particularly if your career involves typing, it will probably pay off within a few months. (It works very well with Emacs, IMHO.) If you spend much of your time fixing other peoples' computers or something, it probably won't help you much.
2. If you're genuinely curious about whether it would be worthwhile, it probably will be, even if only for the experience of shaking up really deeply wired habits.
3. Don't do it if you have deadlines coming up! Seriously, don't. Plan on a week or so of typing 5-15 wpm, and any "cheating" during the first week or two will really confuse you. (You'll probably also feel really stupid at first, like you've forgotten how to talk. It will pass.)
4. You will be able to type on a Qwerty keyboard, but switching back and forth within the first month or two will really confuse you. It's not a big deal once you're used to it, though. Having contextual associations ("my keyboard is Dvorak, the one in the school lab (or whatever) isn't") also helps.
this is quite similar.
It's somewhat related to keyboard layouts and efficiency...
Different strokes, I guess!
(Plus, my WPM is already plenty good. I can hit over 100WPM.)