I had completely debilitating RSI in both arms for years, eventually getting to the point where I was using Dragon Naturally speech recognition for any computer interaction, and a hamfisted stylus for using my phone since actually using fingers with any dexterity hurt too much.
On my better days typing for short periods was ok, but any interaction with a mouse was immediately painful. Once I discovered I could use a trackball mouse on the floor with my feet I was able to get back to a point where I could work professionally again. But if I ever went too far I would be out for at least the next few days without being able to do much.
A year or two after I realized that the pain was generally worse when my arms were cold. After some experimentation I found that almost all of the active-use pain went away when I kept my arms warm.
Since then I've been able to get back into the full swing of things. I'm consigned to always be wearing sweaters, jackets, and hoodies while working, but that has been a small price to pay for being productive again.
Any time I forget and use a computer bare-armed it comes right back again, but so far keeping them warm has been the silver bullet for me. If you're at the end of your rope give it a try!
You know what happens when athletes perform without warming up? They get injuries. It's the same thing. Don't type with cold arms. Beware armrests which suck the heat out of your arms. I have a Herman Miller Embody chair and I had to get some custom armrest covers made for it because I noticed they were sucking the heat out of my arms in the winter.
Ask a doctor. There are far more variations of RSI than most people know. Taking someone else's advice may work, but you may be trying something that fixed a problem you do not have.
My problem ended up being a genetic variation in the structure of the tendon sheaths in my wrist. Many people have it, but if you fall in the middle of the Venn diagram of people with this problem and people who type a lot, it is a problem. Surgery was the only answer to fix it, in my case.
Again, my case will surely be different than yours. For many other people, ergonomics are all that is needed. But ask someone who specializes in these things. They can save you a lot of time and trouble.
De Quervain's tenosynovitis was my RSI. But not all cases of that are the same anomaly as I had - this RSI is a case where the tendon sheath through which 2 tendons slide becomes inflamed. My specific root cause of it is that my sheath was bisected, making two sheaths, so there was less room for the tendons, and once inflamed, it just wouldn't heal. So the fix was to cut it open to make room. Sounds mildly barbaric, but it worked. Almost literally overnight.
No doctor will tell you to go for surgery for this, or any other RSI, immediately. Surgery is always the last answer. I ended up at a surgeon who specializes in wrists, but started with my family doc, then a physical therapist. If your family doc is not helping... which it sounds like they are not, seek out a wrist specialist.
Try what they suggest. Let them know if it doesn't work. But go sooner rather than later - these things can be fixed easier if you get help before it gets bad.
I swear by these mice: https://www.logitech.com/en-us/product/wireless-trackball-m5...
You don't move your hand.. just rest your hand on it and lightly move your thumb by less than an inch in any direction (half inch typically.. if you're moving it much more than that, increase your pointer speed). There are ball mice that use your middle fingers, but I find those put stress on the tendons on the top of your hand.
Add a keyboard that requires very little force to press the keys. I use this one (without the number pad).. it's decent: https://www.microsoft.com/accessories/en-us/products/keyboar...
I don't know that the split in the keyboard matters all that much.. I think the pressure required to press the keys is the most important. Buy whatever kb you like with soft keys.
I also have an articulating keyboard tray:
That lets you move the tray in every direction (back&forth, up&down, side to side, and tilt angle) so you can position the keyboard where it's suppose to be. Add a computer chair.. and you can position everything... even if your posture, etc isn't the best it could be every day.
All the fancy keyboard make no difference if you type too much. Pre RSI I'd type out some code, consider it carefully, rewrite it, reconsider and rewrite. Only then would it be run. It multiplied up the amount of typing I'd do. Now I'll think it through before putting hand on keyboard. No experience with dvorak but logically less typing must beat a 'better' layout.
Emails are shorter now. Just practice getting to the point. Writing docs, I used the dragon voice recognition. It works acceptably when trained, though that is not a quick and painless process.
I have a kinesis keyboard and its splitting/tilting don't help. But, because it lacks a number pad to the right it means I can place my trackball closer, so less risk of shoulder stress from reaching further out sideways. That's very worth it.
I found a mouse to be bad, something to do with pressing the button on top. I now use a trackball. I suspect that's specific to me though.
Above all, when it starts hurting, stop, and for a decent period of time - "But I can't afford to stop!" will be your response. Well, you will stop, nolens volens, it'll just be whether you choose to or your body packing up.
Hand RSI is really bad, I have it now and it'll never go away. It gets better but it's taken over a decade and I'll have it forever. So please note my the para about it hurts so stop.
Also be really cautious about any device that claims to help RSI such as those gyroscopic hand exercisers. They can help when it's improved (IME anyway), but when it's bad it will aggravate it.
All the above my own experience, with a grain of salt as everyone's different - except for the bit about "when it starts hurting, stop" cos I think that's probably universal.
It is entirely possible to type (at high speed in continuous bursts of 20–30m, with short breaks between) for 5+ hours per day every day for decades without injuries. The key is to reduce static load on the muscles and put joints in a neutral position, and make sure to occasionally move around.
It used to be that people who did such work first went to secretary school where they were trained in the proper form by experts. Now, everyone teaches themselves with no external feedback, and uses whatever equipment they happen to find lying around, most of which was incompetently designed, with a heavy priority on cost cutting. Almost no office workers are trained in office ergonomics.
The office furniture and computer hardware industries (along with government research funders, etc.) have been underinvesting in ergonomics research for 40+ years. The keyboards now are worse than what was typically used in the 1970s–80s, and desks and chairs are generally about the same. Official ergonomics standards developed by national and international organizations simply codify existing bad practices instead of trying to work out better designs from first principles.
What ergonomics research is done typically investigates how (small samples of) people will react to very slightly different design choices in the short term if given no training and little adjustment time. Usually the measurements taken are methodologically suspect. It’s a mess.
Split keyboards allow you to keep your keep your wrists at a more neutral position vs having your forearms rotated inward but then twisting the wrists outward in order to type on a flat keyboard.. It absolutely helps with my RSI and if I have to use a flat keyboard for an extended period of time I start to get wrist pain.
A split keyboard and ergonomic trackball such as a Microsoft Trackball Explorer or Elecom Deft Pro, combined with a weight training workout, essentially solved RSI for me.
This is my main RSI issue as well. Suddenly stopped being able to click a mouse one day, and still can't 10+ years later without pain. Even doing the motion in the air causes finger pain. Would love to find a way to fix it. Instead I use a TrackPoint with thumb buttons.
Before that, I could not touch-type on QWERTY, but learned the Neo layout a couple of years back. I gave up on it because I was using too many other machines then. It felt more natural, but I think what really makes the difference for me is the keyboard. On the Kinesis, it is very obvious which finger to use for each key. I was a lot more confident in the physical mapping within days than after decades on standard staggered layouts. I wish there were ortholinear options for laptops.
The first 2 months was stressful, after that it was fine. I think it probably took a year to get back to the same level as I was at with Qwerty though in terms of speed (I was pretty fast), but I only did typing tutors for that first 2.5 weeks, so I probably could have improved things by using them more.
I actually switched back from an ergonomic keyboard to my MacBook Pro keyboard as I no longer needed the ergonomic layout and I could type faster on the MacBook. Had brief RSI symptoms for an hour once since switching to Dvorak, vs a few hours every few weeks on Qwerty. I imagine I'll end up back on an ergonomic keyboard eventually, but between that and Dvorak I think I'm set for my career.
Edit: you can learn the basics with typing games in a few days but then you commit and switch to it full time and your WPM drops to like 10. It takes a while to build up the muscle memory again and it's a tough experience. Like learning to walk again.
Mind you, I never claimed to be a particularly fast typist on any layout, although I have been touch typing for several decades, and on that note I am just as productive on Dvorak as on qwerty -- but at far lesser expense. That is to say, don't switch for the (mythical?) increased typing speed, switch for the ~30%(!) reduction in finger movement and pleasant "rolling" motions you gain.
True, it's not easy to share computers, be it you using somebody else's machine or somebody else using your machine. The latter can easily be addressed by quick layout changes (again, a pain to get right on Windows but a breeze on Linux), while I find that the former doesn't really bother me so much -- curiously, when I can't touch type (such as on my phone), I am hopelessly slow on Dvorak but quite efficient hunting-and-pecking on qwerty (possibly because I retain training from before touch typing?). The most glaring downside is the situations where keys and shortcuts are clearly laid out for qwerty and won't allow you to remap them (case in point: WASD in games); in reality most shortcuts are easy enough to use regardless of layout (side note: I've always been using Ctrl/Shift+Insert/Delete for cut/copy/paste, so the locations of X, C, and V don't affect me -- but modern keyboards with misplaced or missing Insert keys do).
To answer you question: I very specifically switched during the week before New Year's eve, when I could be reasonably sure that things at the office were slow enough to allow me to struggle through an unaccustomed layout. Honestly, the first days were atrocious, but within a week (so ~35 hrs) I had internalized the layout to a reasonable degree, and within a very few weeks (3? I don't really remember) I was up to my usual typing speed.
Bonus: here, have a go and try for yourself: http://gigliwood.com/abcd/lessons/ (no affiliation)
* Specifically, the Norwegian variant of Dvorak, because I need some localized keys. This has been a challenge, but a surmountable one, back on Windows (along with changing the keyboard layout for the Windows login screen, which truly defined the pinnacle of obscurity) -- on Linux it is dead simple to set up and just works.
I switched to Vim, and a Dvorak layout. I really like the feel of Dvorak - every time I have to type Qwerty for an extended amount of time, my hands get uncomfortable.
I also like modal editing. You cut down on the amount of key chords you need. I've since switched back to Emacs with Evil.
Another trick is to implement some break routine. A typing break of 30 seconds every five minutes may already be enough. There are many tools ta enforce these. I used the venerable Workrave, but it's ancient and has bad UX.
Also, read Divided Mind by Dr Sarno.
The reason for it is obvious when you look at it. When you write in a notebook as a student, your arms are almost crossed and your hand is bent at an angle relative to the table. You can maintain this position for hours a day, years on end when you're writing notes. When you're using a mouse, your arm is bent outward and your hand is parallel to the table surface, which twists your wrist. That ends up pinching a nerve in your wrist, which causes all these other problems.
I switched to a pen mouse (back in 1997) which immediately fixed the problem. However, the pen mouse broke after a few months, so I went back to my regular mouse, but I first put it in front of my chest, in front of the keyboard, and I cup the mouse so that my hand is at an angle to the table. My arm from the shoulder to the elbow is resting against my chest, just like when I write on paper, and I don't let it get off my body. When I type I move my mouse out of the way, and when I use the mouse I move the keyboard out of the way.
It's been 22 years now without any repeat of the problems, so I know at least for me this fixed the problem.
To keep your wrists straight, pull the keyboard fairly close to the edge of the desk, then tilt the keyboard so that the top surface is parallel to your forearms, when you have your arms in a relaxed position with your upper arms close to your torso and your shoulders back.
For a high desk/low chair, you need to tilt the keyboard up at the back end. The taller the desk relative to your torso, the more aggressive the tilt needs to be.
Bought an ergodox, within a week the symptoms started going away, which prompted me to buy a second one for home. It's about one month now, and my elbow only has the occasional twinge.
Yesterday I discovered you could map mousekeys on an ergodox. Haven't needed the mouse today, which was the cause of pain.
I still haven't found the optimum ergodox layout though. My current one allows for plenty of mispresses of Ctrl and space. Also some kind of software feedback where if I am in emacs the keyboard switches to an emacs layer would be fantastic.
One thing I noticed that’s not in this article (and from my perception is commonly missing from resources about RSI) is cortisone shots. In my case I tried for a long time to get by with better ergonomics, stretching, NSAIDs, and ice but getting cortisone shots in both my hands made the pain stop for long enough that I could actually rehab my hands properly and function at work.
After the shots I started lifting weights and focused on my ergonomics and it never came back. I’d highly recommend asking about them if you have an inflammation issue. In my case it bought me the time and space to really fix the underlying problems.
To me, the key insight was not to focus too much on the immediate source of the pain, and instead think about overall posture. Fix the posture, and the pain will go away. Adopt unnatural remedies for e.g. the wrists, and the pain will, at best, wander somewhere else.
The section "Desk & Keyboard Positioning" in the article, with the emphasis on right angles, seems the most important to me. In my experience, monitors are often set up too low, so that's the first thing I check when RSI issues crop up.
Personally, I also swear by getting regular massages, so back and neck issues never build up too much.
People just need to remember that they can swivel their eyes without rotating their whole head/neck downward. You want to sit in a non-slouching position, with neck and head straight. The spine is quite good at supporting the weight of the head when the torso and head are straight. But holding up a tilted head for a long time puts a lot of static strain on neck muscles.
Personally I would recommend putting the top of the display at least slightly below eye level, pushing the display as far away as is comfortable (the further away, the less work the eye muscles have to do), and tilting the display backwards at the top so the center of the display points at the face.
Putting the display very high up in the field of view (like the article author’s pictured setup) encourages slouching and tilting the head back.
* * *
> If you hold your elbows at a right-angle, your keyboard needs to be just below your hands in order to keep your wrists straight. You want to aim to hover, not rest at an awkward, upwards angle.
Holding the elbows at a right angle is not something to prioritize. It’s fine, if the furniture allows, but if not, allow the elbows to be at whatever angle they need, and highly prioritize the wrists. IMO the most important RSI-preventing step to take is positioning/orienting the keyboard so that the wrists are not extended or flexed while typing.
> [watch out for a] wrist-breaking upwards slope. I don’t know why exactly this is (perhaps it’s inherited from mechanical typewriters needing their keys in the correct place), but most keyboards slope upwards, causing your wrists to bend as you type.
An upward-at-the-back tilt to the keyboard is not just okay, but is strictly necessary when using a tall desk and low chair. If you use a down-at-the-back tilted keyboard with a high desk, the wrists will end up flexed, which is a really bad idea. The appropriate front/back tilt of the keyboard depends entirely on the height of the keyboard relative to the torso (which determines the orientation of the forearms).
If the desk can be lowered or (even better) if a tall saddle-type stool can be used or if the typist can stand up, then letting the elbows be relaxed (more than a right angle is probably ideal) is also fine. When the desk is low relative to the torso, then the tilt of the keyboard must be adjusted to match the orientation of the forearms.
Dramatically “tenting” a split keyboard (aggressively tilting the two halves upward toward the center) will make a big reduction in static hand/arm/shoulder strain required. Separating and rotating the two halves are relatively minor by comparison.
Finally: the author of this article puts the keyboard very far from the edge of the desk, meaning that his arms are going to be reaching far forward. This puts a lot of strain on the shoulders. Pull the keyboard closer, so that the upper arms hang straight down, with elbows close to the body.
1. Clean your keyboard and mouse.
2. Feed your tendons.
3. Resistance training, such as weightlifting.
I suspect one of the hidden benefits of recommendations to buy a different (ergonomic) keyboard is that simply having a new keyboard means it's probably cleaner than the one you've been using. Makes me wish someone would do a study involving three options:
1. New, clean keyboard same as the old keyboard in terms of layout.
2. New, clean ergonomic keyboard.
3. Just religiously and thoroughly cleaning your keyboard.
It would be interesting if we could quantify how much the recommendation for a new keyboard was beneficial simply because it's clean. (Though I guess you would also need to rate the filth level of old keyboards to get more meaningful data.)
A few weeks of consistent resistance training with bodyweight exercises has worked like magic for me after years of chronic wrist pain. I’ve been gradually working my way into this routine: https://www.reddit.com/r/bodyweightfitness/wiki/kb/recommend...
Pull ups/chin ups, ring rows, and push ups (or their easier/harder variations) all seem to help a lot, though I suspect the pull ups and chin ups are having the most impact.
I personally find it a lot more fun and easier to motivate myself to do these kinds of exercises compared to weightlifting, but I’m sure that works too.
What's frustrating is that _not a single doctor_ suggested physical exercise. Some prescribed expensive PT. Some prescribed some stupid hand stretching exercises which only seemed to make things worse. One prescribed massage (nope, doesn't help with RSI).
Ever since then I'm pretty convinced that most musculoskeletal and tendon chronic ailments are due to either the lack of muscle development (especially in the back and legs/knees) or due to unbalanced muscle development.
But, yeah, generally speaking, I'm pretty frustrated with how common it is for doctors to not tell patients about simple fixes that don't involve drugs or surgeries. Sometimes you get generic handouts with information about, for example, how cleaning can serve as a first line of defense with allergies and respiratory problems, but we don't really promote such approaches in earnest. Interventions that don't involve drugs or surgery are often treated like they don't really count in some sense.
This dirty keyboard theory seems entirely implausible to me. Do you have any evidence supporting it? Or is it purely speculative/anecdotal?
A search of the academic literature for «"subclinical infection" "repetitive strain injury"» turns up zero results.
On the one hand:
It doesn't and I didn't say it did. I only suggested it might be helpful for some people.
On the other hand:
Your hands come into regular contact with the keyboard, sometimes for many hours a day. Some keyboards are shockingly disgusting.
I inherited a shockingly disgusting keyboard at my corporate job. My tendon issues got worse. I began cleaning it regularly. It helped.
You can call it anecdotal if you wish.
It would also would be a good idea to regularly clean keyboards on public computers, e.g. at libraries.
But not because of concerns about RSI...
Edit: “I cleaned my keyboard and my joint pain seemed subjectively better” is a good example of anecdotal evidence; no need to italicize anecdotal.
Edit 2: I removed my comments below, because this whole conversation is off topic and unproductive.
Please explain to me how I should interpret that if it isn't a dismissal of the idea.
Interesting possibility: Perhaps that provided just enough exercise for your hands, as suggested by others.
Though I intend for this to be my last reply to the endless stream of dismissive bullshit regarding the suggestion to clean your keyboard.
There are multiple other comments in this discussion also saying, in essence, "anecdotally, X helped me." None of them are getting the ridiculous levels of pushback this is getting.
"Clean your keyboard" (if you are in so much pain that you are desperate for solutions and will try random shit you tripped across on the internet) is such an innocuous suggestion that I find it impossible to see the replies here as anything other than a giant pile of sexist bullshit on multiple levels.
It's not like I'm suggesting people douse themselves in gasoline and light themselves on fire or spend zillions of dollars. Geez.
So if your wrists hurt, don't just look at how you hold your hands. Change your sitting posture, try a standing desk, etc. - there's a chance that just sitting differently will help your RSI.
(This has happened with me. I used all kinds of fancy ergo keyboards, plus expensive regular chairs. Now I use a Varier Balans plus a simple 68% mechanical keyboard, and all my wrist and back problems went away.)
There's a bunch of factors that go into reducing stress on the hands, mainly however reducing stretching and improving posture. The main issue with mice is that your hands need to move back and forth and back and forth. You should try and minimize this motion as much as possible using keyboard shortcuts/tiling managers and whatnot. But if you must alternating between a thumb and finger trackball is helpful. Using just one runs the risk of putting too much strain on a single finger. If you can enable drag scrolling with your trackball, I've found that to be less aggravating then using a wheel.
Keyboards like the Dactyl Manuform and MS Natural are great allowing the hands to separate and reach a more neutral positioning but many don't leverage the programability of keyboards to improve RSI. Usually this comes with a necessary reduction in size. Boards in the 40% category do a decent job of reducing the amount of work but you also need to factor in the weight of the switches you're using. I'm a bit of an extremist and swear by 12g Kailh Chocs but something like Gateron Clears (35g) would be more ideal for regular typists.
The main thing is that with a programmable board you can jam more features into less keys. Momentary layers, Vim-style leader keys, multi-modifers (Ctrl-Shift-Alt on a single key), and all sorts of shenanigans (like on-board text snippets). It's not as straight forward as regular boards but like the author noted, it opens so many damn doors.
In my opinion Gergo (while smaller then many keyboards) is too large and runs the risk of pinky-overloading (Modifiers, Bksp, tab, etc). Georgi with a optimized firmware and layout is closer to the ergonomic solution (Providing you're willing to put the time to adjust to chorded QWERTY). It's a shame the DataHand died as it did, the layout was damn near perfect ergonomically.
For example, from the article:
> The main mistake was chording. Chording is where you use one hand to type two distant keys. Press Shift-T on your keyboard with only your left hand and you’ll see what I’m talking about.
I can see how it's a stretch if you follow home-row resting position from standard keyboarding advice, but I taught myself touch-typing while playing StarCraft several years before taking a keyboarding class. My resting position isn't home-row, and shift-t isn't a stretch - my hands rest on the keyboard angled instead of straight on, so bending my fingers a bit is enough.
It even hints at the "not blindly follow advice" at the end, while not saying it explicitly:
> Throughout my experimentation, I tried several other strategies that seemed to be ubiquitously recommended across the internet yet were useless pieces advice for me personally.
> I also read a lot of advice that told me to stop using my mouse. This didn’t really make much sense to me as I could clearly feel my wrists hurting more when typing than when using my mouse.
The mouse one is interesting. I've seen two main ways to use a mouse, often based on how big your hands are: Rest your wrist on the mousepad and use your fingers to move it around, or rest your hand on the mouse and lift your arm, using your elbow and shoulders to move the mouse around. The first method I can absolutely see contributing to RSI, the second not so much because you're not twisting your wrist. I'd guess the author reflexively does the second, people giving this advice were taught to do the first.
(In my particular case the conclusion seemed to be that nerves from the arm were being compressed as a result of some combination of poor posture and underdeveloped core/back muscles.)
this is where i found out about it: https://www.nytimes.com/2017/10/30/well/trying-the-feldenkra...
- I stopped using an external mouse (replaced partially by mouth)
- I switched to vim
- In chrome I use the vimperator extension
- I learned as many shortcuts as I could.