Hi guys, I've been learning a new keyboard layout that I generated using the simulated annealing technique demonstrated in carpalx. What do you guys think? I've done a little non-scientific write-up here:
Colemak forum
Forum for the Colemak keyboard layout.
- You are not logged in.
- Topics:
- Active
- Unanswered
Introducing the RSTHD layout
- Started by xsznix
- 3 Replies:
Offline
- 0
I only had a quick look, but on the face of it, it seems to have some good features. I guess it's really an improved Dvorak-like layout, seeming to fix some of that layout's flaws (e.g. L placement) and reducing same-finger bigrams. Obviously because pretty much all the keys change, it makes it much harder to learn, compared to Colemak for example. For me this is the major downside of all layouts like this - a (debatable) tiny gain in efficiency is achieved at a cost of a much more difficult to learn layout. For those who *want* to change everything it may have some appeal though.
Some other observations:
1. Personally I am not a fan of layouts that have E on the thumb. I would rather reserve thumbs for modifier keys. For example, I would instead use that key to define an Extend Layer.
2. Changing shortcuts for copy/paste/undo and other common commands would be a significant barrier for many.
3. You have a fairly common key, D, in the centre column. The keys where you have P and W are much better IMO than the keys where you have D and M. I would swap both those pairs, especially D-P.
Last edited by stevep99 (17-May-2016 10:52:07)
Using Colemak-DH with Seniply.
Offline
- 0
stevep99 said:
I only had a quick look, but on the face of it, it seems to have some good features. I guess it's really an improved Dvorak-like layout, seeming to fix some of that layout's flaws (e.g. L placement) and reducing same-finger bigrams. Obviously because pretty much all the keys change, it makes it much harder to learn, compared to Colemak for example. For me this is the major downside of all layouts like this - a (debatable) tiny gain in efficiency is achieved at a cost of a much more difficult to learn layout. For those who *want* to change everything it may have some appeal though.
Some other observations:
1. Personally I am not a fan of layouts that have E on the thumb. I would rather reserve thumbs for modifier keys. For example, I would instead use that key to define an Extend Layer.
2. Changing shortcuts for copy/paste/undo and other common commands would be a significant barrier for many.
3. You have a fairly common key, D, in the centre column. The keys where you have P and W are much better IMO than the keys where you have D and M. I would swap both those pairs, especially D-P.
Interesting thoughts. I wouldn't say RSTHD is based on Dvorak, because the design principles and heuristics for optimization are very different.
I am indeed one of those people who want to change everything, admittedly. Why else would I buy an ErgoDox (non-staggered keyboards have a decent learning curve) and also switch around 8 of the numbers in the number row? I think that if you're going to make a better layout, it's better not to be held back by the poor design decisions of the past. And there are a lot of those in Qwerty as well as in the standard, non-ergonomic, staggered keyboard form factor itself. I seem to have learned the layout without a problem, so it can't be *that* hard to learn. Plus, making everything different makes it easier to separate your muscle memory for multiple layouts, which I think may have helped me be able to retain proficiency in Dvorak and Qwerty while learning RSTHD.
The efficiency boost is admittedly very debatable. Since there is very little rigorous research done on the subject of keyboard layouts, I'm not actually sure that there's an objective difference in how layouts perform compared to one another. But that doesn't bother me. For me, it's more a matter of comfort. There are certain strokes that I find annoying, slow, and hard to type quickly, and being able to optimize them away gives me a pretty big sense of satisfaction. Plus, a 54% decrease in same finger usage compared to Colemak and and a 72% decrease in jumps across the home row on consecutive key presses, among other things, doesn't seem that tiny to me.
As for the other observations:
1. I didn't think of it at the time that I created this layout, but putting more modifier keys on the thumb is also a very compelling solution. If I didn't put E on the thumb, that key would probably be a shift key, which would take a big burden off the pinkies, which are overworked in all layouts anyway.
As it is, the four small keys on each thumb are modifier keys that essentially give access to four Extend layers. Home/End/Pg Up/Pg Dn are on layer 5. The arrow keys are easily accessible from the home position, so they don't need to be on a separate layer. I also have a layer for special symbols, so I don't ever have to reach up to the top row for them.
2. Cut, copy, paste, and undo are on Layer 6 in the positions that you would expect them to be (bottom left), but I find myself only using them when I'm also using the mouse at the same time.
3. That's an interesting point. Personally, I still prefer D and M in the center column (for my index fingers, the center column is slightly easier to reach than the bottom row, and I also don't like same finger bigrams that hop into the center column like WN and PH… although MN is also relatively common at this point), but I can see how people with different preferences would want to swap D-P and M-W. Fortunately, swapping those keys doesn't change the overall analysis very much.
Offline
- 0
I agree with SteveP. Another issue of a thumb E would be compliance with other physical keyboards: If you encounter another computer it's fairly simple to use a software remapping on it but if you're used to having the most common letter on a thumb key it'll likely throw you off completely.
*** Learn Colemak in 2–5 steps with Tarmak! ***
*** Check out my Big Bag of Keyboard Tricks for Win/Linux/TMK... ***
Offline
- 0