Staggered/Columnar/Ortholinear analysis?

[ergo_typing]

Does anyone know of any research/analysis into the relative merits of Staggered/Columnar/Ortholinear layouts in relation to reducing keyboard related pain?

Splitting a keyboard layout seems to be an obvious win to reduce ulnar deviation induced pain but I was curious if anyone had run any studies on Staggered/Columnar/Ortholinear pain reduction.

[Findecanor]

The TypeMatrix study, and the TRON keyboard paper point to ortho-linear and columnar fan-layout respectively being beneficial for 1) minimising finger extensions, and 2) minimising wrist movements from the home row.
However, they don't really say why either in itself would be a good thing,

I know that some people with movement issues in their hands greatly prefer small columnar keyboards so as to minimise finger movements though.
When it comes to avoiding moving the wrist or the entire hand, there exist also an opposing view out there: that it is good to move your hands around to avoid static tension, which could lead to muscle pain.

Note that the original TypeMatrix and the Tron keyboards both minimise ulnar deviation by separating the left-hand and right-hand keys.
(An ortholinear or columnar keyboard without enough hand-separation/opening angle is worse than a traditional keyboard IMHO. Most people type on a row-staggered keyboard with the keyboard left of centre, the left hand at around 0° angle, and their right at ~20°)

[ergo_typing]

Thanks for the links!

The "why" you mention really is the thing isn't it! Also, the way Raymond P. Bello set up the TypeMatrix study is very odd. They had 43 participants and they gave all of them a complete ergonomic work station adjustment, had them all do an exercise program and gave them a new keyboard. I surely would hope that the majority of the participants noticed an alteration in their symptoms. I don't think it really in any way proves that the TypeMatrix was responsible for that change though.

The TRON paper is an interesting design study but doesn't attempt to prove that the new design reduces symptoms. The design goals of "gain speed" and "lessen the fatigue" are admirable, but not really proved, more asserted. But as this appears more of a design paper rather than an ergonomic related pain study, fair enough!

The TRON thumb cluster really made me think of the new Dygma one:





You interestingly mention "the left hand at around 0° angle, and their right at ~20°" do you have a citation for that, or is it through empirical observation?

You also mention "minimise finger movements": I wonder if this is in relation to arthritic pain in finger joints, rather than RSI in general?

And "good to move your hands around to avoid static tension": I fall into this category for sure, but I come from a classical music background, so I've always felt that typing was some cut down way of playing the piano

A couple of papers that I've found talk in more objective measurements about splitting the layout but not staggered vs ortholinear that might be of interest:

Kroemer 1972 https://doi.org/10.1177%2F001872087201400110
Thompson 1990 https://doi.org/10.1177%2F154193129003400428

[Findecanor]

The TRON thumb cluster really made me think of the new Dygma one:

There is a long tradition of columnar keyboards with many thumb-keys, and in an arc, especially in Japan. Those have certainly influenced the DIY ergo keyboard scene and thus Dygma.
http://xahlee.info/kbd/Japan_M-Type_TRON_keyboards.html

You interestingly mention "the left hand at around 0° angle, and their right at ~20°" do you have a citation for that, or is it through empirical observation?

That's only my impression, yes. I think there is also a large proportion of people who hold their left hand at near 45°  (columns T-F-C, R,D,X, etc).

You also mention "minimise finger movements": I wonder if this is in relation to arthritic pain in finger joints, rather than RSI in general?

I'm not sure what the reason, but I've seen it happen to relatively young people.  AFAIK, arthritis is uncommon for people younger than 65 y/o.

A couple of papers that I've found talk in more objective measurements about splitting the layout but not staggered vs ortholinear that might be of interest:

I've seen the Kramer paper referenced, but not found a free copy.

[nevin]

yes, split and hand separation will be the biggest change and improvement.
whether you go with split staggered, split ortho, or split columnar stagger will all just be..... well... "splitting" hairs. (ha, ha)

i don't have studies to link to but i can speak from my experiences.
- split is definitely the way to go. i've been in front of a computer for the last 26 years for work, got a split ortho in 2018 (i won't go back to one piece boards). the hand separation alone is worth the change.
- i have very little experience with columnar stagger boards. though, i do have an ergodash. i'm not a big fan of the ergodash or others mainly because of the thumb keys that usually come with columnar stagger. while i whole heartedly believe that modifiers (control, alt, shift, etc.) should be towards the center of the board and not mandated to your pinkies. i found the thumb clusters made the stretches for commands/shortcuts harder/further away then how they were on my current board (keeb.io viterbi)
- i do more shortcuts/commands during the day than actual typing (designer) so my use case may be different than most.
- 14 column can be setup extremely close to standard qwerty (as a transition to alternate layouts, ortho, etc., not speaking to qwerty as far as efficiency of a layout, that's a whole other ball of wax.) with 14 columns you only have to move one or two keys.
- your posture and desk/setup ergonomics have more to do with "keyboard related pain" than staggered, ortho, columnar.

hope this helps, answers some of your questions.

feel free to ask me if you have questions i might be able to answer.

[kajahtaa]

You cannot prove a keyboard reduces pain. You already messed up the body. Go to PT.

I've used ortho and columnar stagger almost 5 years. Never had pain. Nothing is proven.

Had pain from reverse curls tho. Still do reverse curls.

[ergo_typing]

@Findecanor that ashlee.info Japanese link is amazing - weird keyboards are such a delight. 45° wrist angle makes my wrists twinge just thinking about it - that must be doing such nasty things.

Re arthritis: I think this is one of the big issues when discussing keyboard ergonomics - so many different aspects of user pain get rolled into one thing. Let alone one would assume the large amount of such pain is self reported, rather than diagnosed.

@nevin "split and hand separation will be the biggest change and improvement" totally agree. I think the default position of keyboards forcing ulnar deviation is, for me, the biggest issue. What led you to go with an ortholinear keyboard?

Re "thumb keys that usually come with columnar stagger" One of the first ergonomic keyboards I tried was an Ergodox and the thumb clusters where what made it less useful to me. The thumb clusters just started giving me a new novel pain the thumb near the wrist! I wonder, with the ergonomics of the thumb, whether a generally useful thumb cluster solution is ever possible. The traditional keyboard long space bar allows for such small adjustments of thumb position on a per user basis, something which a thumb cluster will inherently never allow because of key size constraints.

Re "posture and desk/setup" absolutely - if you are ergonomically in difficulty before you even start typing, there's no hope. I wonder what impact actual typing technique makes? Not just fluid touch typing but how heavily the user bottoms out too. All interesting ergonomic questions.

@kajahtaa "You cannot prove a keyboard reduces pain" An interesting thought - but as a possible counter, when one works with a physio, a common thing they will do initially is an assessment of the injury, but also how it was caused and what actions exacerbate the user pain. For example, if a user has a their main computer monitor too low and not straight in front of them, they might present with neck pain. A physio would of course assess the damage in the neck muscles but would probably advice repositioning of the monitor. Likewise with wrist pain, if the user gets regular pain from typing on traditional keyboards relating to ulnar deviation, the physio might suggest trying a split keyboard.

Proving a causal relationship between a specific keyboard and a specific user is always going to be difficult. It's not like when debugging a complicated piece of software where one can specifically isolate a small segment of the code. There will always be multiple factors when working with people. However, one could design a study whereby a sample of users presenting with similar or related pain are randomly split, one group switch to working with a split keyboard and the other group continuing as before. Then look at the results over time. An interesting thought experiment!

[Findecanor]

Re "thumb keys that usually come with columnar stagger" One of the first ergonomic keyboards I tried was an Ergodox and the thumb clusters where what made it less useful to me. The thumb clusters just started giving me a new novel pain the thumb near the wrist! I wonder, with the ergonomics of the thumb, whether a generally useful thumb cluster solution is ever possible. The traditional keyboard long space bar allows for such small adjustments of thumb position on a per user basis, something which a thumb cluster will inherently never allow because of key size constraints.

Yet another totally personal observation/theory is that when typing on a traditional row-staggered keyboard there is a slight wrist angle of the right hand against the keyboard's plane, but when typing on a columnar or ortho-linear keyboard, the hand is kept parallel to the keyboard.
This affects of course also the angle of the hand to the thumb keys.

Some columnar keyboards, such as Maltron, Kinesis, Dactyl/Tractyl (DIY) and Moonlander (if tented) have instead the thumb cluster on a plane rotated slightly so that the thumb-keys are pressed at a more natural angle.

The ErgoDOX copied the Kinesis' thumb cluster ... but has it in the same plane as other keys.

[kajahtaa]

OP if you research this and find solutions for people you'd be making a massive contribution to board and medical community.

Sincerely hope this goes well for you.

[ergo_typing]

@kajahtaa my speciality is in computer science not medical research but it would be amazing if a medical researcher designed a study round this area.

[ergo_typing]

@Findecanor

traditional row-staggered keyboard there is a slight wrist angle of the right hand against the keyboard's plane

very interesting observation. I wonder if this could be alleviated by floating the wrists a little more.

plane rotated slightly so that the thumb-keys are pressed at a more natural angle.

that rotation makes the hand position much more neutral doesn't it - more of a handshake position.

The ErgoDOX copied the Kinesis' thumb cluster

I always thought that was an odd choice - I'm assuming they made it to simply the manufacturing process, rather than an ergonomic decision.

[ak66666]

I kind of performed such a research, the result is not a written article but a design.
Most of the ideas were listed in this thread, most important:
- neutral arm and hand position
- reduced hand/finger movement

The starting point was Let's Split and Dactyl-Manuform.
From them I got the idea of columnar and vertical stagger.
Other points:
- Thumbs operate on maximum two keys each.
- Keyboard to be tented by default.
- Layer operations to be mostly controlled by thumbs.
- Cursor, mouse to be on the keyboard.
- Key wells/Curvature can be successfully imitated/approximated with PCBs.
- Angle between the halves (in case of a unibody version) is around 44 degrees.
- No key/modifier use should result in splaying the palm, or stretching/rotating in to an awkward position.
- No key/modifier use should result in discomfort within the palm.
- Key press should go straight down (no sideways pressure on the joints.) ==> Switches to be placed of a flat surface (not curved like in DM or tented like in LS.)
- Standard QWERTY layout with close-to-standard modifier (Ctrl, Win, Alt, Menu) placement to be used.
- Index finger prefers stretching than bending in.
- Poke/punch not squeeze - the key needs with the hit/punch motion, not like squeezing the ball.

Again, this was not consolidated in writing, but was produced as a series of designs, which were observed in the natural habitat, the negative traits eliminated and the positive ones carried over into the new iterations.
The current version turned out to be stable enough and does not seem to require any significant changes.

[nevin]

@ak66666 - what does your current iteration look like? curious to see what you ended up with.

[FrederickWood]

I kind of performed such a research, the result is not a written article but a design.
Most of the ideas were listed in this thread, most important:
- neutral arm and hand position
- reduced hand/finger movement

The starting point was Let's Split and Dactyl-Manuform.
From them I got the idea of columnar and vertical stagger.
Other points:
- Thumbs operate on maximum two keys each.
- Keyboard to be tented by default.
- Layer operations to be mostly controlled by thumbs.
- Cursor, mouse to be on the keyboard.
- Key wells/Curvature can be successfully imitated/approximated with PCBs.
- Angle between the halves (in case of a unibody version) is around 44 degrees.
- No key/modifier use should result in splaying the palm, or stretching/rotating in to an awkward position.
- No key/modifier use should result in discomfort within the palm.
- Key press should go straight down (no sideways pressure on the joints.) ==> Switches to be placed of a flat surface (not curved like in DM or tented like in LS.)
- Standard QWERTY layout with close-to-standard modifier (Ctrl, Win, Alt, Menu) placement to be used.
- Index finger prefers stretching than bending in.
- Poke/punch not squeeze - the key needs with the hit/punch motion, not like squeezing the ball.

Again, this was not consolidated in writing, but was produced as a series of designs, which were observed in the natural habitat, the negative traits eliminated and the positive ones carried over into the new iterations.
The current version turned out to be stable enough and does not seem to require any significant changes.

Can you provide a detailed explanation of the design choices and features incorporated into the keyboard, including the inspiration and starting point of the design, the specific elements that were focused on such as neutral arm and hand positions, reduced hand and finger movement, columnar and vertical stagger, thumb operation on a maximum of two keys each, and any other notable considerations such as cursor and mouse placement, default tenting and key wells and curvature that can be imitated with PCBs, and the final design stability and any further improvement planned? When it comes to essay writing, the keyboard is an essential tool. That's why I am so interested in this subject because I am a writer for this site https://assignmentbro.com/us/do-my-english-homework and I want to improve my skills. A well-designed keyboard can make the process of writing an essay much more comfortable and efficient. A keyboard that is designed with ergonomics in mind can reduce the risk of repetitive strain injuries, which is especially important for students and writers who spend long hours typing.

[anawilliam850]

To my knowledge, there isn't much research comparing the pain reduction of Staggered, Columnar, or Ortholinear keyboard layouts. However, some studies have suggested that Ortholinear layouts may reduce wrist deviation, which can lead to reduced wrist pain compared to Staggered layouts. Columnar layouts have been found to reduce finger movement, which can help reduce repetitive strain injuries. Ultimately, the best keyboard layout for reducing pain would depend on the individual's typing habits, wrist posture, and comfort level. It is recommended to try out different layouts and see which one feels most comfortable for you.

[ak66666]

@ak66666 - what does your current iteration look like? curious to see what you ended up with.

Other variants are on thumbsup.shop

[ak66666]

Can you provide a detailed explanation of the design choices and features incorporated into the keyboard, including the inspiration and starting point of the design, the specific elements that were focused on such as neutral arm and hand positions, reduced hand and finger movement, columnar and vertical stagger, thumb operation on a maximum of two keys each, and any other notable considerations such as cursor and mouse placement, default tenting and key wells and curvature that can be imitated with PCBs, and the final design stability and any further improvement planned?

I put together some ideas on the keyboard layout here:
https://www.allthingsergo.com/case-study-one-approach-optimizing-ergonomic-keyboard-layouts/


For the physical design I haven't got such notes, not as a single paper, but rather as replies on Reddit. Probably will consolidate them in an article some time soon enough.
For the new designs I have very little new ideas. Mostly it is about implementation, different MCU, split versions.

As for the history of development you may see how it evolved here: https://www.thingiverse.com/ak666666/designs

[nevin]

@ak66666 - that's a brilliant idea for multilevel keyboard, and impressive that the risers/connectors are part of the matrix. very, very, smart / unique. two big thumbs up!

[ak66666]

@ak66666 - that's a brilliant idea for multilevel keyboard, and impressive that the risers/connectors are part of the matrix. very, very, smart / unique. two big thumbs up!

Thank you very much!