My proposal of difficulty map for grid keyboard

[Piotr Dobrogost]

I like the idea of difficulty/strain map used to create Workman layout – http://www.workmanlayout.com/blog/
I think it could be improved. Below is my take on difficulty level for keys on grid keyboard like TypeMatrix.



Original OJ's difficulty map:


Notes:
My map is more fine grained than the one of OJ as I believe differences in difficulty/strain among fingers are more subtle than the scale 1, 1.5, 2, 3, 4 can reflect. My map uses 9 step scale instead of the original 5 step scale.
I left the original values for home keys for index and middle fingers as typing these keys is the most comfortable.
Ring finger being weaker than index and middle fingers results in its difficulty raised to 1.5 from the original 1 for home row key. As pinky is even weaker its difficulty was raised to 2 from the original 1.5 for home row key.
Difficulties for index fingers reaching horizontally was raised to 4.5 from the original 3 as horizontal movement is very unnatural kind of movement for fingers (without moving whole hand) and this is not mitigated by index fingers being the strongest ones. The rise of 1.5 and not less comes from the fact that I find horizontal movement more difficult than reaching diagonally which I left at the original value 4.
I find diagonally folding index fingers more difficult than both diagonally reaching and moving horizontally that's why I raised difficulties to 5 from the original 4.
I don't agree with the original difficulty of reaching with index finger being greater than difficulty of reaching with ring finger so I raised difficulty of reaching with ring finger to 3.5 from the original 2.
I find folding middle finger really difficult due to its length, more than reaching with index finger so I raised this difficulty to 3.5 from the original 3 making it 0.5 greater than difficulty of reaching with index finger.
I find reaching with pinky more difficult than folding it due to its shortness so I raised difficulty of reaching to 4.5 from the original 4.

I'm curious if you agree with my adjustments and if not then what would you change?
The map was created with http://www.keyboard-layout-editor.com/
Below is source in json which you can import and modify:
[{x:1,c:"#ff3d3d",a:5,f:7,fa:[2]},"4\n\n\n\n\n\n4.5",{c:"#ffa0a0"},"2\n\n\n\n\n\n3.5",{c:"#00c100"},"2\n\n\n\n\n\n2.5",{c:"#00ff03"},"3\n\n\n\n\n\n3",{c:"#ff7a7a"},"4\n\n\n\n\n\n4","4\n\n\n\n\n\n4",{c:"#00ff03"},"3\n\n\n\n\n\n3",{c:"#00c100"},"2\n\n\n\n\n\n2.5",{c:"#ffa0a0",fa:[1]},"2\n\n\n\n\n\n3.5",{c:"#ff3d3d",fa:[2]},"4\n\n\n\n\n\n4.5"],
[{x:1,c:"#12b40e"},"1.5\n\n\n\n\n\n2",{c:"#10a20d"},"1\n\n\n\n\n\n1.5",{c:"#128e0e"},"1\n\n\n\n\n\n1","1\n\n\n\n\n\n1",{c:"#ff3d3d"},"3\n\n\n\n\n\n4.5","3\n\n\n\n\n\n4.5",{c:"#128e0e"},"1\n\n\n\n\n\n1","1\n\n\n\n\n\n1",{c:"#10a20d"},"1\n\n\n\n\n\n1.5",{c:"#12b40e"},"1.5\n\n\n\n\n\n2"],
[{x:1,c:"#ff7a7a"},"4\n\n\n\n\n\n4","4\n\n\n\n\n\n4",{c:"#ffa0a0"},"3\n\n\n\n\n\n3.5",{c:"#00c100"},"2\n\n\n\n\n\n2.5",{c:"#ff0000"},"4\n\n\n\n\n\n5","4\n\n\n\n\n\n5",{c:"#00c100"},"2\n\n\n\n\n\n2.5",{c:"#ffa0a0"},"3\n\n\n\n\n\n3.5",{c:"#ff7a7a"},"4\n\n\n\n\n\n4","4\n\n\n\n\n\n4"]


ps.
I originally posted this at http://www.workmanlayout.com/forum/viewtopic.php?id=72 but it seems that forum has very few users so I decided to re-post it here.

[LuX]

I have to say I don't agree with most strain maps out there. Maybe my fingers are just weird, or it depends on technique... I would rate my strain like this:

Obviously this is all extremely subjective. People have different sized fingers, for example men have a shorter index than ring finger, women have the opposite.
Instead of starting with a premise that some fingers are weaker, I decided that index home key is a '1', and began subjectively comparing how much harder a different key is to press, and so I ended up with this map, it may not be 100% accurate.
I you float your hands or not while typing has a major effect on how I would rate the keys, and placement of hand initially as well, so making a really accurate strain map is nearly impossible.
I should point out I made that map based on the ErgoDox, but for all intents and purposes that keyboard is nearly ortholinear (or grid-like).

When designing a layout, I think it's more important to consider n-grams, rather than the individual placement of keys, not to say it's not important, but still. It's the main reason I prefer Colemak over Workman.

[Piotr Dobrogost]

I have to say I don't agree with most strain maps out there. Maybe my fingers are just weird, or it depends on technique...

Well, there's rather small chance you would agree with entire map.

Instead of starting with a premise that some fingers are weaker, I decided that index home key is a '1', and began subjectively comparing how much harder a different key is to press

I did the same.

I you float your hands or not while typing has a major effect on how I would rate the keys, and placement of hand initially as well, so making a really accurate strain map is nearly impossible.

Isn't floating hands while typing discouraged?

I should point out I made that map based on the ErgoDox, but for all intents and purposes that keyboard is nearly ortholinear (or grid-like).

Nearly but not the same. With ErgoDox the placement of columns reflects differences in lengths of fingers. I think this removes difference in strain between index and middle finger in regard to reaching and folding.

When designing a layout, I think it's more important to consider n-grams, rather than the individual placement of keys, not to say it's not important, but still. It's the main reason I prefer Colemak over Workman.

I wouldn't want to use layout with comfortable rolls where pinky would be more loaded than index or middle finger Every aspect should be taken into account.

All being said there are quite a few similarities between our maps;
- reaching with middle finger is easier than with index one (however I would suspect these to be equal on ErgoDox)
- folding index finger is easier than folding middle one (here I would suspect these to be equal on ErgoDox, too)
- reaching with index, middle and ring fingers is easier than moving index horizontally
- reaching with pinky and reaching/folding in the center columns have the most strain

...and quite a few differences;
Do you really find typing in home row and reaching with ring finger as easy as typing and reaching with index finger? Did you take into account how weak ring finger is in comparison with index one? Also index finger is totally independent of other fingers whereas ring finger is connected with the middle finger and pinky. I recommend the following test. Place your hands flat on desk in front of you and tap 30s with both index fingers as quick as you can and then do the same with your ring fingers for another 30s and see what was easier and in which case you did more taps.
Do you really find moving your index fingers horizontally easier than reaching with them diagonally? In the latter case the distance is longer but I think it's more than compensated with the fact it's much more natural move.
Do you really find folding your ring finger easier than reaching with it? I generally find reaching being more natural move than folding and thus easier. Exceptions are pinky, because it's extremely short and index finger, because it's shorter than ring finger what makes folding a little bit easier than reaching.

[chuckdee]

Probably a redundant question to some, but I haven't really been able to find anything on it- what do the numbers represent?

[Piotr Dobrogost]

Probably a redundant question to some, but I haven't really been able to find anything on it- what do the numbers represent?

Strain when typing given key; lower is better. In addition in small font I placed original values used at http://www.workmanlayout.com/

[LuX]

Floating is definitely better, but it also requires skill. The problem with novices and people using a new layout is that they anchor their fingers too much to the home row. When you allow your hand to float completely, you don't have to twist your wrist in an arc, and stretching is practically non-existent, there are mostly 'dives'. That is why I generally value rolls much more than individual placement.

The stagger on the 'dox is very subtle. We're talking about millimeters here. It looks much more significant on paper.

We can argue about finger strength all day long, we are just different, my map is what it is. I feel like you keep all your other fingers glued to the home row and try to reach the keys using 100% the strength of your finger. Allow your fingers to be in the air and move along with your hand as you type, it becomes much easier to reach and press keys, as half of the strength comes from the wrist and arm motion.

I'm happy with my layout and the way I type, obviously you aren't, or not enough. Are you designing a new layout or just trying to create a new weight map for other designers?
In the end only you can say what feels best for you. Whether it applies to your situation or not, I can provide alternate ideas to think about, confirmation of a problem doesn't add value to the question.

So, my recommendation is that you give floating a try. It's a skill that will take some time to learn and even longer to master, but it's definitely worth it.

I wouldn't want to use layout with comfortable rolls where pinky would be more loaded than index or middle finger Every aspect should be taken into account.

Workman sounds good, but isn't really. It hardly takes every aspect into consideration. It tries to fix one issue but introduces many more. My opinion anyway. Colemak isn't perfect either, but it sure is much more balanced than most layouts out there.

[chuckdee]

Probably a redundant question to some, but I haven't really been able to find anything on it- what do the numbers represent?

Strain when typing given key; lower is better. In addition in small font I placed original values used at http://www.workmanlayout.com/

I get that.  But what I was asking was is that an objective measure given somewhere,  or just a subjective measure that you created?

Sent from my SM-T710 using Tapatalk

[Piotr Dobrogost]

I get that.  But what I was asking was is that an objective measure given somewhere,  or just a subjective measure that you created?

The latter one.

[chuckdee]

I get that.  But what I was asking was is that an objective measure given somewhere,  or just a subjective measure that you created?

The latter one.

Is it just relative,  or do the units have a meaning?

Sent from my SM-T710 using Tapatalk

[Piotr Dobrogost]

Is it just relative,  or do the units have a meaning?

It's relative.

[chuckdee]

Is it just relative,  or do the units have a meaning?

It's relative.

Thanks! I was just trying to put the map into perspective

Sent from my SM-T710 using Tapatalk

[jacobolus]

Can you take a top-down picture of your hands in their most neutral resting position over the physical keyboard you’re trying to optimize for? Instead of having each finger precisely on a single keytop, relax all your fingers as much as possible so they’re in a kind of medium arc, and put your hand in the center of where you expect its range to be, with your arms/wrists in the same position you use for typing.

The reason I ask is that your numbers here don’t bear any particular resemblance to the difficulty I notice for reaching various keys, so I’m curious how your diagram matches up with a picture of your own hands. (E.g. maybe your hands are a very different shape from mine.)

On a standard keyboard, my fingers’ easy reach range ends up something like:

(This picture goes a long way toward explaining why I think the completely square grid design is stupid: overall, keys end up about the same if not slightly harder to reach than on the standard keyboard, but you lose the advantage of sticking to a standard design. Anyone who cares enough to build keyboards in a custom shape can do a lot better by adopting a column-staggered layout with more separation and a bit of rotation between halves, ideally some significant tenting, and a better design for the thumb keys.)

Also note, the ease of reaching different rows depends substantially on the keycap shape and key profile. With a chiclet laptop keyboard for instance, the ZXCV row is very easy to reach while the QWER row is more of a stretch, whereas on a manual typewriter the ZXCV row is a real pain to reach but the QWER row is pretty easy. So are you designing for the TypeMatrix specifically, or for some other grid keyboard?

[yellowfour]

I thought of something to say about Workman's effort model now that others are also copying and modifying to their own whims. It is counterproductive to assign such nuanced grades for the first nine keys or so for each hand. I mean no need to be so specific, from 1.0 1.5 2.0 2.5 etc for the best nine keys on each hand. You just need to make about three grades 0, 1, and 2, and assign three keys to each grade. This is to allow some flexibility for keys to swap to maximize certain metrics.

One, the efforts between these keys aren't that far apart. Some people erroneously assume (as I once did) that individual finger flexibility has a big effect on typing speed and comfort. That is, people tend to short-sell how nimble the big fingers really are (and over-sell the pinky).

Two, these graph-makers tend to miss the forest for the trees. A layout is not just individual keys, so they neglect other important factors in typing holistically, such as rolls. For example the top ring finger is usually assigned a worse grade. But that ignores the fact that it is also one of the better keys to start inward rolls.

Three, they assume that all reaching should be done by the fingers alone. That's very unergonomic. The arms should move freely to bring the fingers closer to the keys without over-stretching or over-curling the fingers.

This is the effort model that I use:
.

And very successfully created advanced layouts from which compete with the current top well-known layouts (namely MTGAP) (My layouts are named BEAKL):