[EL_Loco]

The split keyboards and ortholinear keys aren't that much better. They're mostly ideas about what would be better ergonomically, but they haven't been as thoroughly thought out as they seem. For the split keyboards, look at the whole of human history and tell me, out of the hundreds of thousands of years of tool use all the way back to the stone age, how many of those tasks were done with your arms set straight out in front of you, hands 12 to 14 inches apart. I bet most of human fine movement tasks were done with the hands close together (sewing for example). The position with your hands extended and 12 inches apart actually puts your elbows in a quite uncomfortable position, and I wouldn't be surprised if that in itself would lead to some injuries in that area (elbow). As for the ortholinear key designs, that's not very ergonomic either. Make a fist with your hands. Now open your fingers. Do they open in a parallel line pattern like the orthogonal keyboards? No, they don't. They spread out, in a star-like pattern. If the keys were laid out similar to that pattern then, yes, I might concede a benefit there. But these two regular complaints about traditional keyboards while hailing split and orthogonal keybords as well thought out ergonomic solutions seem more like a fad to me. By the way, I don't think the traditional typewriter-like keyboards are great, far from it. I do think, though, that our hands are pretty resilient, but not infinitely resilient. Too much repetitive work, insufficient rest, personal biomechanics, can all eventually lead to injuries, on all these types of keyboards.

PS-> Who came up with calling it 'ortholinear'?

[tjoff]

I don't think it can be argued that they aren't more ergonomic than a regular keyboard. If you manage to create an uncomfortable position with a split keyboard you have the luxury to adapt... And I haven't seen people claiming that it is the holy grail, just that it is better.

Many otholinear keyboard are in a grid, yes. I guess mostly because it is easy, aesthetics and it fits everyone as good/bad. Many keyboards have their rows at different height to adapt to the different length of each finger. But how much? Depends on hand size and preference it is hard to imagine one being mass produced.

[nextos]

I don't have huge hand issues due to ergonomics, but I have experienced some.

I have tried most of what is out there: Kinesis Advantage, Maltron, Kinesis Freestyle, Matias Ergo Pro, Microsoft Sculpt, several Vortex...

Paradoxically, my best experience is with an Apple Magic Keyboard 2 (and I use Linux). I think that's because it's tiny, so I can move my hands and use mostly index and thumb fingers. Plus, actuation force is so light. Yet depth is acceptable. This is not a butterfly switch. A great side effect is tiny latency [1].

I also enjoy using the Magic Trackpad, which now works quite well in Linux. Scrolling using a finger gesture is very ergonomic compared to mouse wheels or even trackballs. At least for me. It promotes position variability.

Of course, these Magic Keyboard scissor switches are much more fragile than any mechanical one. In 4 or 5 years of intense use keys start falling apart. That's my experience with an equivalent keyboard in a MacBook Air. So, shortly, I will try to assemble a split keyboard with more shallow mechanical switches, inspired by the Magic Keyboard experience.

[1] https://danluu.com/keyboard-latency/

[the_pwner224]

Please post on HN when you get that project done, I'd love to see what you come up with!

I'm in a somewhat similar situation, where I have a split keyboard with blues. It's pretty good, but the combined height of PCB + plate on top + standoffs and plate below PCB (the 'case' is just two plates above/below the PCB) + switches + keycaps is quite tall. I'm also going to try to make a slim keyboard in the near future.

[jacobolus]

Ultimately, the ergonomic goal of a tool like a keyboard which is used for extended periods should be to encourage use with all joints in a relatively neutral position, with very low static load on muscles.

The thing that splitting the keyboard (while keeping it flat) helps with is reducing “ulnar deviation” of the wrist (outward rotation in the plane of the hands). It is also possible to avoid ulnar deviation on a standard keyboard by changing typing style a bit, which is why some people have less problem than others with a single-piece keyboard.

What helps most is not splitting the keyboard per se, but tilting the two sides upward toward the center (the ideal is probably something like a 45° tilt, but even a bit of “tenting” helps significantly).

Having the hands palms-down and close together (as forced by a standard keyboard) ends up causing an unpleasant trade-off between (a) forearms rotated uncomfortably inward about their axis (“wrist pronation”), or (b) elbows swung forward or out to the side to reduce the amount of wrist pronation, in the process causing shoulder and back strain, or causing people to rest their elbows, forearms, wrists, or palms on some surface to compensate for the shoulder strain, and often in the process flexing or extending their wrists.

Wrist flexion or extension is really the worst for any kind of repetitive motion. It dramatically reduces strength and responsiveness of the joints, and can cause severe repetitive strain injuries. You will notice that typical human tools are designed very carefully so they can be used with a mostly straight wrist (or tool-user technique has developed to work around tools which don’t make this obvious).

Tilting the keyboard the right amount front-to-back to match the height of the keyboard relative to the torso can help quite a bit; most keyboards I observe in the wild are tilted incorrectly. You want the plane of the keyboard to be roughly parallel to the plane of the forearm. So on a tall desk, the keyboard should be tilted up at the back. On a low keyboard tray they keyboard should be flat or even tilted slightly down at the back.

[aasasd]

Personally for myself, I found the best option to be MS Natural 4000's backward tilt (down at the back)—I'd like it to tilt even more at low desk heights but oh well. That way, the wrists stay straight while the fingers drop downwards a bit, in a feeble emulation of Kinesis Advantage's pits. I even use a wrist brace sometimes, which pads the wrist on the palm's side giving even more of a raise—feels really great for the fingers, but alas can't be reproduced in a stock wrist support because of the thumb getting in the way.

[jacobolus]

On a tall desk with a low chair, a “backward” tilt to the keyboard almost inevitably causes some uncomfortable hand/arm positioning.

If you have a high desk and low chair and you can’t change the furniture, you want the keyboard to be tilted (sometimes very aggressively) “forward” so that the plane of the keyboard is parallel to the plane of the forearms.

The keyboard should be kept close to the torso so that the upper arms can hang loosely down with relaxed shoulders. A bent elbow isn’t generally too big a problem.

You’ll notice that typewriters from the middle of the 20th century have a very aggressive tilt to them, to compensate for the standard high desks and low chairs (which were optimized for a middle-sized male to write with a pen, but made quite poor furniture for typically female typists.)

> wrists stay straight while the fingers drop downwards a bit

I have observed people doing this; usually their wrists are not actually straight, but are extended (i.e. bent downward), which is also a bad idea.

Resting palms, wrists, or forearms on a surface is generally undesirable during active typing in my opinion. It’s fine when the hands are not being used but are just sitting there.

Also, the fingers should not “drop downward”. The primary joint which flexes to type a key on a keyboard is the near knuckle where the finger attaches to the palm. You want these first knuckles to be straight when the finger is on the top of the unpressed key, and then flex downward to press the key; this is the strong and efficient part of that joint’s range of motion.

When reaching the near rows of keys, the further knuckles will be bent. When reaching the far rows of keys (which should ideally have tall keycaps) the further knuckles will straighten to reach the keytops.

My personal opinion is that the Kinesis Advantage is not the most biomechanically sound design; it was more or less a cheaper-to-produce ripoff of the Maltron, made by people who had done less thinking about typing than Malt. With careful attention it is IMO possible to significantly improve on both of those designs.

YMMV.

[dasKrokodil]

> My personal opinion is that the Kinesis Advantage is not the most biomechanically sound design

I agree. I use an Advantage, and while it's certainly better than traditional keyboards, I think it has a couple of flaws in terms of ergonomics:

• the keywells are not rotated inward and the distance between them is not adjustable. Therefore, unless your shoulder width is exactly right for this keyboard, you have to rotate your hands in- or outwards

• the thumb clusters are too far away. For me, if I relax my thumbs, they drop right in that little space between the space/backspace keys and the keywells.

• the thumb clusters are also too high, making it somewhat painful to rest the thumbs on the keys for longer periods of time

• the tenting angle is insufficient

[UkiahSmith]

I have a split keyboard, and I think the real benefit is not that I can rotate my wrists, but that it allows me to pull my shoulder blades closer together. I sit more upright and have less back pain.

[ianhowson]

This. I also enjoy being able to move the two halves around through the day, shifting the muscular load around. It makes it easy to accommodate different seating, standing and lying positions.

(I never figured out, despite 'ergonomics training', how I was supposed to hold my extended hands and arms upright for hours per day without shoulder and neck pain.)

[bsdubernerd]

Are you kidding, right? Some people have a natural predisposition to issues such as carpal tunnel and don't need to use a keyboard at all to exacerbate the problem.

Traditional keyboards are simply not great for a lot of reasons. I experimented with tons of different keyboards and keyboard layouts mostly out of curiosity. You don't need to go far to feel the difference: microsoft curved keyboards are already much better than most regular keyboards.

Once you realize you can reduce finger travel by orienting the keyboard to the natural direction of the finger flexion you won't go back easily, and you'd likely want to go further with something you can orient in any way you like.

And yes, 'ortholinear' plays a big part in how you position the keys. I've used a lot earlier models of fully split keyboards such as the "comfort keyboard", but it never felt as natural as a simple linear grid, as I tended to put the keyboard at weird angles just to achieve what a simple vertical array of keys does. Heck, I'll go as far as saying that if you want to experiment without going crazy, the first step would be to try an ortholinear keyboard. It makes more difference.

I've personally settled on a Kinesis Advantage Pro. It does many things right out of the box. I can only complain about the function keys, but I barely use them anyway. It's not fully split, but the position is good enough, and it avoids the issue of fixing the keyboard pads to the table in order to avoid movement which is an issue I had with all split keyboards (and no, a rubberized mat is not enough).

I think the price is right for the build quality, and I didn't want to go the DIY route mostly because I have other projects going on (a DIY keyboard is a lot of work - I have plenty of respect for those who make even a flat one).

Edit: typo.

[wwright]

The Advantage has been updated into the Advantage II (I have two and adore them ). One of the big changes is the function keys, if that is ever of interest to you.