I was inspired by Englebert's "Mother of all Demos" shown at Super Happy Dev House 42 and decided to actually try out the device that was in his left hand.
It was amazing on the IMAX. I was a little sad that it only displayed on about 1/5th of the screen infront. I assume it only fit onto one projector instead of the software cutting it up to fill the whole dome.
It is a little slow at first, but once he gets to explaining the tech, it was wonderful.
Cute. But I'd need a lot more than this for programming, which uses many special symbols. Even regular English requires being able to type quotes, question marks, exclamation points, numbers, etc..
Also, as far as the implementation of this particular typing tutor goes, it needs to let the user retry to type a given chords until he gets it right, instead of immediately moving on to the next one.
And the chords should be taught in small batches, rather than trying to teach them all at once.
While we're on the subject of chording, I should mention that there's a vim plugin that will let you map chorded keystrokes:
I kept getting caught up cognitively on the fact that the thumb corresponded to the 5th box in all of the chord diagrams. I feel like some visual differentiation would have made that easier to grok.
I also had the same issue, oddly enough it got worse as i got more accustomed to the program. I think mainly because I was visually counting the boxes from the right to left.
maybe instead of boxes use a numbering or finger image? But yeah definitely some differentiation between the 'f' key and the 'space' key
Isn't keyboard-related RSI caused by moving wrists? This system doesn't require you to move your hand at all.
BTW. I've just read on Wikipedia that there's no causation proven between RSI/CTS and keyboard typing. Interesting, as from what I've seen, it's a widespread belief.
It's a problem with keyboard ghosting. That's the problem with chording keys like this, once you're hitting >= 3 keys, you'll find some just won't work.
is there a standard mapping for chorded keyboards? i feel like X has a much easier combo than E. would be interesting to use data on people's error rate for different key combos and then remap frequently used letters to easier combos.
I remember reading about Steve Roberts building chording keyboards into his "handlebars" on his Winnebiko years ago. Since he had well more than 2^5 bits using both hands[1], he was able to assign macros that helped significantly with his effective typing speed.
Englebart says that his layout is not intended for speed typing, with the typical trained typer being able to type 30-50wpm. He observed that typing speed on conventional keyboards does not have a strong correlation with typing speed on his chording layout. Sight-reading piano players have a bit of an advantage here, of course.
I've always wanted one of the Twiddlers, and am kind of excited they're back on the market, but I'd vastly prefer a wireless version. I've also been looking forward to the release of the FrogPad. It would be nice to have a better-than-touchscreen keyboard for emails on the iPad or iPhone.
[1] I can't remember the exact layout now. If it used all ten fingers, or if it had multiple positions for, say, thumbs, à la the BAT keyboard, it's gone from my memory. Regardless, it used both hands.
Interesting idea, but it's rather odd that the keystroke for P is pressing <SPACE>, but the keystroke for <SPACE> is a four character combination when <SPACE> should be its own chord. While the binary incrementing keystroke combination is somewhat logical, I would imagine that a more frequency-based assignment would ultimately be more successful as muscle memory is far more important than remembering that the chord for V is binary+1 of the chord for U.
29 comments
[ 4.5 ms ] story [ 109 ms ] threadIt is a little slow at first, but once he gets to explaining the tech, it was wonderful.
Also, as far as the implementation of this particular typing tutor goes, it needs to let the user retry to type a given chords until he gets it right, instead of immediately moving on to the next one.
And the chords should be taught in small batches, rather than trying to teach them all at once.
While we're on the subject of chording, I should mention that there's a vim plugin that will let you map chorded keystrokes:
http://www.vim.org/scripts/script.php?script_id=2425
I kept getting caught up cognitively on the fact that the thumb corresponded to the 5th box in all of the chord diagrams. I feel like some visual differentiation would have made that easier to grok.
I also had the same issue, oddly enough it got worse as i got more accustomed to the program. I think mainly because I was visually counting the boxes from the right to left.
maybe instead of boxes use a numbering or finger image? But yeah definitely some differentiation between the 'f' key and the 'space' key
Maybe I should build a Dvorak-ish version of this...
BTW. I've just read on Wikipedia that there's no causation proven between RSI/CTS and keyboard typing. Interesting, as from what I've seen, it's a widespread belief.
I guess you can tell that I'm not an emacs user :)
Paul, can you change time intervals to to catch less accurate chors?
When in doubt, make a configuration option right? ;)
https://secure.wikimedia.org/wikipedia/en/wiki/Stenotype
And there are a fair number of competing standards and keyboard layouts (especially if you count layouts/standards for various languages).
Stenotypes machines and musical instruments are really the only chorded keyboards in wide enough use to develop any kind of standard.
Englebart says that his layout is not intended for speed typing, with the typical trained typer being able to type 30-50wpm. He observed that typing speed on conventional keyboards does not have a strong correlation with typing speed on his chording layout. Sight-reading piano players have a bit of an advantage here, of course.
I've always wanted one of the Twiddlers, and am kind of excited they're back on the market, but I'd vastly prefer a wireless version. I've also been looking forward to the release of the FrogPad. It would be nice to have a better-than-touchscreen keyboard for emails on the iPad or iPhone.
[1] I can't remember the exact layout now. If it used all ten fingers, or if it had multiple positions for, say, thumbs, à la the BAT keyboard, it's gone from my memory. Regardless, it used both hands.
http://www.cc.gatech.edu/classes/cs6751_97_fall/projects/fol...
Right hand for navigating (mousing), left hand for typing.