Touch typing: To the Moonlander and back.

This picture is best enjoyed while listening to the theme from 2001: A space odyssey

Here's a super-fun blog post about me learning to touch type.

Why would I do this to myself?

Why would I, a 38yo (at the time) techie, suddenly decide to learn to touch type?

(What do I mean by touch type? I don't just mean "type without looking at the keyboard", which I've always been able to do... I mean "type using the correct fingers on the correct keys...and still without looking at the keyboard"...)

1) I wanted a Moonlander keyboard (above). They look so cool, like you're driving a space rocket. And they have pretty lights!

2) I thought I could type faster. I became aware that the folks in my team who could touch type were ~20% faster than me. Or more. What if every slack message, every email, every blog post about touch typing, was typed 20% faster!?

3) I felt like a fraud. Shouldn't I, a 38yo (at the time) techie, be able to touch type?

4) Ergonomics. I'm not getting any younger, and I know a couple of people who've had problems with RSI, CTS or back problems caused by poor posture...so I was keen to be proactive and not wait for myself to get hurt before making a change.

5) Can old dogs learn new tricks? As someone that had typed a lot every day for 30+ years, the muscle memory was fairly well locked in. I was genuinely interested...would it even be possible for me to change something so fundamental, old dog that I am?

(Full disclaimed: this was also in lockdown - so what else was there to do, really?)

Tell me more about this Moonlander thing...

You can use the left half by itself if you want to be a leet gamer.

The Moonlander is a keyboard made by ZSA. It is:

Split - it's in two halves, so you can keep your hands further apart

Tented - so the central edges are elevated from the desk, allowing a more ergonomic position for your wrists and forearms

Ortholinear - the keys are in straight lines vertically, instead of being staggered like bricks. In theory, this means you just need to straighten and contract your fingers to hit most keys, without travelling sideways

It's also backlit and highly configurable - more on that later.

I ordered the keyboard in black (it's also available in white for ultra-spaceship-but-hard-to-keep-clean vibes) with Cherry brown switches...nice and tactile, without being obnoxiously loud.

It's fair to say the keyboard isn't cheap - I think it was over £300 to buy the thing and pay the import duty. But, you know, lockdown...I needed to spend money on something.

Customisation

The keyboard is highly configurable. For each key you can decide:

What the key does - anything from simply typing a character through to running full macros
What the key does on different logical layers (ie when a modified key is pressed at the same time)
What the key does when pressed differently (long press, double press etc)
What colour the key should glow in each instance

The flip-side of this flexibility is that you actually kinda have to customise it. There will be a layout that makes sense to you and your hands based on your use cases, what operating system you use, etc. You just have to iterate towards that as you get used to the keyboard.

All of this configuration is achieved via fairly intuitive software:

Once you've configured the firmwear, you get to upload it by pressing a pinhole switch on the keyboard. There's an amazing sense of jeopardy every time!

Setup

This isn't exactly what I started with, but here's my current setup... things that matter for typing: I've got a desk that's the right height...and a good chair that has heigh-adjustable arms:

Please get in touch if you're a fellow fan of the Office Ladies podcast.

Now concentrate…This is the science bit…

Being a "data-driven" professional and keen to measure "ROI", I wasn't going to just switch keyboard and hope for improvement. I needed to show my receipts. So, for a couple of weeks before I switched I collected some control data on typing speed...and then I set about testing myself over the first week.

I set myself one basic rule - I had to type properly on the new keyboard. That is to say, I had to start with my index fingers on the home keys (F,J) and use the right finger for each key strike. There's a lot of stuff online about which finger goes where, won't repeat the theory here.

So, let's see how week one went:

I think in future economics books, this will commonly be referred to as a "Truss curve"

6wpm.

Yep, my first test scored me 6wpm.

Thankfully, this was largely caused by my inability to find the key that acted as the space bar. Once I found that, my pace increased quite rapidly.

It was hard, though. Having to really concentrate on something that you normally do completely automatically is a bit mind-bending. And my fingers hurt! I didn't know that "tired hands" was a thing that could happen when typing... but in reality my former typing technique only employed four fingers regularly...so the other six found this whole experiment a bit of a wake up call.

I also faffed a lot. I tried to increase the tenting angle gradually, to get used to not being flat any more...and I worried a lot about the placement of the keyboard. How far apart are the two halves? What angle are they on? How far from my chair are they?

I'm unsure if these things ever really mattered...but they certainly don't matter now. I can put either half basically anywhere and still tap away quite happily. And I've come to view the slight variation this gives my body as being a good thing.

I think the most important thing in this phase was that I kept at it. I used the Moonlander for everything from day 1 - no matter how pressing the deadline for a document, no matter how quickly I needed to ping someone an email... And I did a lot of practice, too...spending maybe an hour a day doing typing tests, writing fiction etc.

Practice?

Who knew that practicing typing was a thing you could do? But practice I did, mainly using:

KeyHero - essentially the first google result that came up. This is what I did all my early testing in. It gives you a decent mix of vocab and punctuation. But some of the phrases are deliberately difficult to type and yield depressingly low scores!

monkeytype - even if you've no interest in typing tests, you should check out the UX on monkeytype, really excellent. Limited vocabulary and no punctuation, but using monkeytype feels much more like playing a game than a school exam.

The Typing of the dead. Are you a fan of typing? And B movies? And old-skool shooter games? This has you covered. Type the words. Kill the dead. Enjoy the horrifying plot development.

Never has typing felt so perilous.

There are tonnes of other options out there - these are just the three that I've relied on over time.

How long did it take to break even?

I didn't collect datapoints every day after the first week, just sporadically, resulting in a curve like this over the first few months:

The long hard road out of typing inadequacy.

So I could type as quickly* on the Moonlander as I had been able to on the old keyboard within about four months.

* in most circumstances. Things that were off the beaten path...function keys, random special characters etc - probably still took longer to find.

...two years later?

I stopped collecting data regularly in June 21... so where did I get to a couple of years later, by the end of 2023?

In a word, faster.

Around 25% faster, on average. What was a "good" test result on my old keyboard was a "poor" test result on the Moonlander in 2023. My initial range was around 65-80wpm and on the Moonlander it was 80-100wpm. Woot.

There were still some special characters where I was less accurate, but most of it was completely natural.

So...was I over the moon (lander)?

Yes. I loved it.

I could type quicker, more comfortably and I enjoyed both tinkering around with the configuration of the keyboard and continuing my typing practice from time to time.

But, all was not totally perfect.

1) My wife doesn't get on with the Moonlander. It's not very easy to use for someone who isn't...me. Even for basic web browsing. As there is no lettering on the bulk of the non-alphabet keys - and as I type entirely on muscle memory -I'm not even able to tell her which key does what without giving it a go.

2) I have kids. I'm not sure the Moonlander is what they should be learning to type on - would this actually help them to touch type when faced with a normal keyboard?

3) Thinking about it...Did it actually even teach me to touch type when faced with a normal keyboard? I wanted to learn to touch type - not learn to touch type on a Moonlander. Would old habits come straight back if I tried using a regular keyboard? There was only one way to find out...

Enter the Ducky

So, I bought myself a pretty new keyboard... a Ducky One 2 TKL (tuxedo colourway!) and got straight to collecting some data.

My ESC key says "ESC" - but you get the idea.

Dawn of the Ducky: Day 1

I didn't write a diary about this process. If I had, day 1 would have read

"Dear Diary, Oh sweetest diary...what have I done!?"

There were difficulties.

Who put the C, V and P keys in such weird places?

Which finger are you supposed to hit B with, anyway??

Why are enter and backspace so far away???

Seriously, going back to a "normal" keyboard wasn't fun. I struggled a little bit with key placements, as above. There was also a little bit of muscle memory that I had to resist... But the overarching thought?

I feel like a t-rex playing a tiny toy piano.

My hands are so close together. And they're flat. My shoulders are so closed. I feel so hunched! Why would anyone do this to themselves?

But, you know, let's give it a chance.

Day 1 results:

If you're just scrolling and looking a the pictures, "hi!" - this is a graph showing how my typing speed changed when I stopped using the Moonlander and went back to a normal keyboard. These results are from day 1.

Getting up to around 50wpm on day 1 wasn't too bad. 50wpm is fine. It doesn't feel like you're typing quickly, sure...but it's not too frustrating to go through life typing at 50wpm.

Week 1 results:

Stats from monkeytype - wish I'd done this all along, it would have made this whole blog entry so much prettier!

I really quite enjoyed week 1... there was pretty constant progress and a couple of definite "breakthrough" moments... when my hands got used to finding C, then V, then P.

And now?

I've now had the Ducky for several months, and I alternate a bit between the Ducky and the Moonlander. How do I feel about things?

I can touch type.

I can type well enough on the Ducky to happily say that touch typing, in general, is a skill that I've acquired.

I'm still fastest on the Moonlander.

While I'm now ~15% faster typing on the "regular keyboard" than I was at the start of this process, I still have more practice touch typing on the Moonlander, so am still fastest there.

I still much prefer the Moonlander.

The sense of being a t-rex playing a tiny toy piano has never really left me when using a "normal" keyboard. Starting to use the Moonlander was definitely more difficult, but it always felt like I was learning to do something better. Going from the Moonlander to the Ducky just feels like a massive step backwards. So much less comfortable, less natural...

And while I've come to develop quite a nice smooth motion typing on the ortholinear layout, staggered keys feels really jerky and needlessly deliberate in comparison.

I feel like I finally understand what Acid Burn was talking about.

Classic Crash and Burn burn

So there we go. A whole blog post about learning to touch type. I thought it would be a riot and I'm sure you'll all agree that is was definitely that. To the brave few that made it this far - thanks so much for reading. Take care!