Save your hands while going full speed: how to have your cake and eat it

(Note: this blog post was originally written on Jul 20, 2022 but in a private forum, so it has now been made public)

The backstory

When I was still in university, I had a few friends that were avid software developers. They programmed round the clock for fun; for them, this was their joy. However, about halfway through our time at university, two of them developed RSI; this had a profound effect on me at the time. I remember thinking, “OMG, they have not even begun their careers yet and yet the thing that they love to do just became that much harder for them”.

I became very afraid that if I did not care about this topic, it was only a matter of time before I, too, was affected. As they say, “the best time to act was yesterday but the second-best time is today”. I chose to act that very day.

The most obvious solution to the RSI problem is to just “type less” but that seemed equally as bad to me: I want to be able to do more while risking less (don’t we all).

So began a lifelong journey to both save my hands and become faster at inputting data into computers.

Faster typing is better

I have often heard the expression “I can already type faster than I can think” or “Thinking is the hardest part of my job” and this is used as an explanation for why you should not bother trying to become faster at typing. While I do agree that “thinking” is the most important part of the job, I completely disagree that should mean no investment in becoming a faster typist. In fact, I think you can use the same logic to reach the opposite conclusion.

Let’s run through my logic together:

• I find that I speak faster than I type; you are likely the same (since most people speak at 200WPM). So I know it is possible to get information out of my brain and into the real world faster than I currently do with typing.
• Because of this difference, all of the things I’m trying to write are pushed to a mental buffer that is more onerous than speech.
• My fingers then attempt to flush that buffer as fast as they can.
• This buffer flush is not free for me, unfortunately. The larger the buffer gets, the more mental energy it takes to maintain while also preventing context switching (I have to stay on task otherwise, that is the same as wiping or degrading the mental buffer).
• Eventually, I end up in an equilibrium where my thoughts have been slowed down enough that the speed that I’m putting words into the mental buffer is equal to the speed of my typing.
• This essentially brings my thinking speed down to my typing speed (or worse, because I’m concentrating on thoughts that I’ve already thought so that I can flush them out of this mental buffer).

Overall, my thinking speed probably looks something like this as a result:

(Red line is my max theoretical thinking speed over time and the green line is my theoretical typing speed over time and the blue line is my “real” thinking speed)

My bottom line opinion is this: the faster you type, the quicker you are at flushing that mental buffer and the more time you can spend on the important parts of your role. Whether that be as an engineer or a manager or really any professional that uses a computer for their day job. This hit me again lately as I realized that most of what I do as a manager looks like writing English words to people as fast as I possibly can. The faster that I can type, the more time I have to think, the more problems that I can tackle, and the better I can spend my limited time.

What we really want, in an ideal world, is for the green line to match the red line. That would mean that I am spending all of my time thinking, and the words just land on the computer immediately, requiring none of my mental capacity. That is the vision that I am aiming for.

If you are anything like me, then I hope that I have convinced you that it’s in your best interest to become as fast a typer as you possibly can. At the very least, I’m hoping that you’re questioning the “why bother becoming faster at typing when I spend more of my time thinking” fallacy.

The first step: measure twice, cut once

The first step, with anything, is to measure where you are right now so that you can figure out how to meaningfully improve in the future. In this case, this means measuring your typing speed. The online typing speed tool, that I enjoyed using when I first started on this journey years ago, is 10-fast-fingers.

Do me a favour, open that up, give it a few tries, and post your best typing speed below, or maybe share it with friends. This is my first try (which is usually slower than your other tries):

(My current roughly average speed. This is the result of following the advice I’m about to give)

When I first began this journey back in University, I used to type at 40–50WPM at my top speed. These days my top speed is ~110WPM but I sit very comfortably at the ~90WPM mark and find that most of my typing is at that speed in reality.

According to the WPM article on Wikipedia “An average professional typist types usually in speeds of 43 to 80 wpm”. I have also seen it said that the average software developer types at 70WPM. Can’t remember where I read that though.

If your typing speed is below that: never fear! You are just at the beginning of your journey. If you are above 110WPM then stay for the end of the blog post; there is still a progression for you.

Now that you have worked out how fast you write, your next step is to decide what the best course of action is for you to get faster AND save your hands.

Decisions, Decisions, Decisions

When making this decision, the key points are:

1. Better ergonomics
   You want to be able to type as much as possible while straining yourself as little as possible. There are many points that factor into this, but the key points are a good keyboard, a good chair, and a good desk setup.
2. A better keyboard layout
   You want to use a keyboard layout that facilitates reducing strain on yourself as much as possible. This means that your fingers should be traversing as little as possible, and you should be able to be as precise as possible.
3. Touch typing
   You want to be able to type without looking at the keyboard (ever). This means not just knowing where the alphanumeric keys are but also never looking down to reset yourself and knowing where the number keys and other special keys are without looking too.

For the best ergonomics, if you are a professional that spends most of their day in front of a computer, I hope that you’re already sitting on an Aeron chair (or something like it) and have an excellent screen setup at home / work. If not, make that your first priority.

Get yourself a good keyboard

However, it is entirely possible that you are just using a Mac keyboard. I would rate this as “less than ideal”. In order to save your back (and prevent hunching), you want to avoid needing to squish your hands next to each other while typing: this means that you should get yourself a split keyboard.

For Split Keyboards, you have heaps of options. My favourite two are perhaps:

Kinesis Advantage

(the keyboard I currently use for character-based typing)

This keyboard is a dream on your fingers, wrists, neck and back. I cannot rate it highly enough. For most people, this will be the perfect solution and you can stop right here. It works even if you don’t want to change your keyboard layout away from Qwerty (which you should). I have been using this keyboard for years, and only during EOFY season where I’m spending 12 hours every day for multiple days in a row writing reviews did I even feel so much as a twitch with any finger.

Ergodox Moonlander

More flexibility in the split, and more programmability than the Kinesis Advantage but no “cup” to hold your fingers in. It’s got pros and cons, so it may be more your cup of tea. This is simply much more customizable than the Kinesis Advantage, which is only a pro if that is what you plan to do, otherwise, I would just stick with the Kinesis.

In short, start with a good keyboard as this is literally something you can just pay money for and the problem is solved; the rest of this blog post is instead going to cost you time or effort. Or both.

Get a better keyboard layout and touch typing

The two most popular keyboard layouts, apart from Qwerty, are Dvorak and Colemak. When I first started on this journey my goal was better ergonomics AND to be able to touch type. Out of the two keyboard layouts I gravitated to Dvorak for a reason that is best depicted in this article:

Dvorak is not proven to be faster — the highest recorded speed on QWERTY is 227 WPM, while the highest recorded speed on Dvorak is 194 WPM. However, there are many more people who have practiced QWERTY for their whole lives than Dvorak. Perhaps if more people used Dvorak there would be a fastest Dvorak typist. Even though QWERTY has a much higher top speed, most top typists agree that Dvorak has the potential to reach that same speed so do not let this scare you away from switching layouts.

Even if Dvorak is not proven to be faster than QWERTY, it is definitely more comfortable. When typing with Dvorak, your fingers move only 62% as much as they do with QWERTY. If you are someone who experiences pain while typing or types a lot, switching to Dvorak could help prevent injury. You can see in the heat maps below that Dvorak is much more centered around the home row.

(Source: The Dvorak Keyboard Layout — Das Keyboard Mechanical Keyboard Blog)

In summary, by switching to Dvorak:

1. I would be forced to learn how to touch type because even looking at the keys will no longer help me. This alone will let me become a much faster typist.
2. The way that Dvorak is designed makes it much more ergonomic that Qwerty because your fingers remain on the home row more frequently and don’t move that often. See the heatmap comparisons from that blog post for example.

Qwerty Heatmap

Dvorak Heatmap

Colemak Heatmap

Hopefully, you can see the significant difference in ergonomics from that alone.

I’ve also included a Colemak heatmap just to show that it, too, is much better than Qwerty from an ergonomics perspective. I ultimately chose Dvorak because it had common vowels on the left and common consonants on the right; this resulted in an almost musical left-right rhythm typing experience. For example, take the word “musical” itself. In terms of hands, that is:

• M — right hand
• U — left hand
• S — right hand
• I — left hand
• C — right hand
• A — left hand
• L — right hand

This means that Dvorak also uses both of your hands pretty evenly; also reducing RSI strain. It has the added benefit of a typing mistake also “feeling wrong” because you have lost your left-right hand rhythm; as though you’re playing left-right on the bongo drums and you hit with your right hand twice.

Me typing on a Dvorak keyboard; left, right, left, right, left, right…

With these three changes (keyboard, new layout, and touch typing) made years ago, the strain on my hands is negligible, even with a full day of software development or writing annual reviews. I simply don’t worry about getting RSI or other work-related injuries.

Not only that, but it only took me 4 weeks after going “cold turkey” in a switch from Qwerty to Dvorak to get back to 30WPM and approximately another month to get back to 50WPM. This means that if you switched keyboard layouts from Qwerty to Dvorak over a holiday (like the December-Jan period), then you could be back at work with a decent typing speed and a much, much more ergonomic setup.

My method of learning the new layout was:

1. Go cold turkey. Make the switch to Dvorak and refuse to use Qwerty ever again. (Don’t change my keyboard keycaps to Dvorak, though to ensure that looking down is a habit that won’t provide any value thus ensuring that I learn how to touch type).
2. Memorise the keyboard layout until I could see the entire keyboard in my head. (Does not take that long, it’s only 26 letter keys after all)
3. Practice writing as much as I could. (I especially liked typing “The quick brown fox jumps over the lazy dog.” because as an English Pangram it made sure that I hit every key on my keyboard at least once and thus proved that I knew the layout)

Was it annoying to be much slower for a few months? Absolutely. Was it worth it and one of the best decisions that I ever made? Absolutely. I would highly encourage everybody to make the switch.

If you wanted to stop here, I completely understand. This is where I stopped for many years and it has been completely sufficient for my professional career so far. This is actually where I would recommend that most people stop as it is the best place to be in terms of cost-benefit: you have a low chance of RSI while also attaining a speed that is much higher than the average professional.

If you have made it here: congratulations!

But can we do better?

I want to be the very best, like no one ever was

Many would argue that being able to type at ~100WPM consistently with a low chance of ergonomic issues is brilliant. Why go further than that?

The problem is: that the speed of speech is ~200WPM and we’re only typing at ~100WPM, there is still too much buffering going on, slowing us down and preventing us from getting our thoughts into the computer at a more natural speaking speed.

There are two broad schools of thought on how to get to this much faster speed:

1. Speech-to-text tools
2. Stenography

Speech-to-text tools are certainly one way of solving this problem. There are certainly many people in our industry attempting to make this happen and they have been attempting to make it happen for years. Despite this, the idea has never really caught on and is quite difficult to learn to use well. The key problem is consistency, it’s hard to know how to press the right “buttons” of a speech-to-text tool and get the output you were expecting. As a result, getting a high level of accuracy is very difficult and may take years to master. That said, as AI improves over the next decade I could imagine this changing, maybe. It’s a bit of a gamble. Even then, you are giving up an entire communication channel by making speech be your data input mechanism: no talking in a meeting while taking notes typing for example.

Conversely, have you ever wondered how court reporters or live-captioning works? If they can’t type faster than people can talk, aren’t they always just falling behind and trying to race to keep up? How do they do it?

The tools of the trade for a court reporter. (Source of image)

People called Stenographers (not to be confused with Steganography) can easily type at above 200WPM consistently:

Court reporters train on stenotype machines, and the minimum speed required for certification by the National Court Reporters Association (NCRA) is 225 wpm.

(Source: How Fast Is the World’s Fastest Court Reporter? | Connor | Indianapolis | )

If you are like me, then your reaction is “holy heck that is fast”! But it gets faster:

NCRA member Mark Kislingbury of Houston, Texas is the Guinness World Records title holder for the Fastest Court Reporter in the World. Kislingbury achieved 360 wpm with 97.23% accuracy. That’s insanely fast. We doubt that he’s ever had to work that quickly in the courtroom.

So, given that, how do these court reporters manage to get that fast? What even is Stenography? I think that now is the time to let this relatively recent “Half as interesting” video explain:

To which your reaction might be “well, that is all well and good, but what about for ‘real’ software development work”. To which I would respond that there are already software developers out there (a whole community actually) that use Stenography every day for software development work. From what I can tell they seem to get regularity to about 160–170WPM. Here is a video of one of them explaining their system (put aside a spare 30 minutes):

And here is a more recent one:

If you are in another profession that does not program but instead spends most of your time writing English words to people (my fellow Managers, I’m looking at you) then the results you should be able to get are even better.

The next question, then, is ergonomics. Is this better or worse than Qwerty or Dvorak? The answer seems to be: much better. The reason is that a “word” is considered to be five characters in most typing tools, so typing at 100WPM means actually typing 400CPM (ignoring the spaces). 400 presses of anything every minute is nothing to scoff at from an RSI perspective. Conversely, stenography is a “chorded keyboard” which means that you press multiple keys at a time and whole words are spit out. For example, pressing BG gets you the word “being”. You just got 5 characters for the price of two keys pressed or 1 “stroke”. From my research, it seems that most stenographers that write at ~200WPM are actually only stroking 50 times per minute (50SPM). Compare that to a regular character-based keyboard where you would be making 800SPM to accomplish that speed and you can quickly see how, despite the higher speeds that stenography allows, it is also much much MUCH easier on the hands for the equivalent speed. Stenography wins “hands down” 😜 on the ratio between strokes and output generated ratio and thus is much more ergonomic and easier on the hands.

This is why, since Apr 27, 2022 I have been attempting to learn Stenography and as of today ( Jul 20, 2022 ) I can type at a blistering 20WPM to type a full story (I can type faster on common words and more standard typing tests):

This brings us to the two major caveats of Stenography:

• You have to memorise a lot more than 26 keys; maybe even going so far as to memorise an entire “theory” for typing like Pheonix Steno or Magnum Steno. This in turn means that the learning journey is quite slow: it seems to make ~1 year of practice for most people to hit 80–100WPM (though some have done it much faster).
• Since you type phonetically in Steno, your ability to spell words is no longer as relevant, instead, your ability to pronounce them becomes more relevant. This means that you are undergoing a paradigm shift while learning a new layout which was not the case when switching from Qwerty to Dvorak. (As an aside, removing this dissonance is why the CharaCorder exists

Overall, I see this as just a much higher cost for an equivalent higher gain. If it takes me two years to get to 200WPM but I’m typing at that speed for the rest of my career, it was totally worth it for me. I’m willing to continue this experiment and give it a try and hope to get to a speed where I can use Stenography for my day-to-day work rather than as a hobby I practise in my spare time.

If you’re interested, my starting journey was pretty simple:

1. Started on a regular keyboard and bought a hobby stenography device: The Uni v4
2. Get started with Plover and the excellent learning resources: Beginner’s Guide: Get Started with Plover
3. Practice a little bit every day and try and make it fun.

For the record, here is my current keyboard:

(One amusing thing about the Steno community: you end up getting lots of pictures of people’s legs because that’s one of the more ergonomic places to put the keyboard)

Conclusion

If I had to surmise this entire blog post into the actions that I would love to see people do, in order:

1. Measure your current typing speed and assess how much fatigue you experience after a day of work and how fast you are able to communicate to, or through, a computer.
2. If you don’t have an ergonomic keyboard, get yourself one now. Your arms and back will thank me later.
3. If you are still using Qwerty, I strongly encourage you to reconsider. Your best bang for buck is likely to be to move to Dvorak (or Colemak) so start there. If you really insist, you can jump straight to Steno.
4. For those that are already pretty fast (or are still in school/university): Steno might be for you, I would encourage you to look into it if you are interested. I wish I had started when I was in that phase of my life. Fair warning, it will take a large amount of consistent effort to get into, but conversely, the payoff is insane typing speeds for the rest of your career. Your call.

In the end, I hope that I have, at the very least, made you consider this problem space very seriously and that some fraction of the people reading this post will be inspired to try and improve the ratio of time that they spend thinking vs typing.

Good luck in your journey and, if you have any questions, don’t hesitate to ask.

P.S. My ideal input device would be one that I could use while standing up on public transport or lying down in bed while also writing at >200WPM and would work on mobile and desktop devices alike. That’s my dream; I’m happy for it to take time and effort to learn such a device as it would be game-changing.