QWERTY, DVORAK, and COLEMAK, Oh My

One of the first, and most important decisions, when the typewriter was first commercialized in the late 19th century, was how to organize all of the letters on the keyboard. Thus, the QWERTY layout was born. Because of the success of the machine that used the QWERTY layout, it became a standard and was used by most other manufacturers of typewriters in that era. Since standards tend to be self-reinforcing, we've all been using QWERTY ever since!

But there has been a robust debate over the years about QWERTY and its alternatives. Is QWERTY the best layout for a keyboard? Is it worth trying others? Am I wasting years of my life by typing less efficiently than I could?

All worthy questions. This page will not answer all of those questions, but it might help you think more about your typing and keyboard than you ever have before. And maybe it will inspire you to try something different...

QWERTY Key Click Patterns

First, let's take a look at QWERTY. Here you can see the entire book Anna Karenina being typed, starting with the first sentence in slow-ish motion. As those letters are typed, it is impossible to see whether QWERTY is bad or good.

Once we then animate in all 2,000,000+ additional characters (not including numbers and punctuation not visible in this abbreviated visual keyboard) we can easily see that the QWERTY layout leaves a lot of gaps for letters that are used less frequently. We can also see that the left hand does heavier lifting than the right. And some fingers get quite a workout, while others, like the right pinky, do practically nothing. (Note that we're simulating a true full-size keyboard touch-typist - we're not considering "hunt and peck" typists or thumb-driven mobile keyboard typists! Also, thumb measurements are from hitting the space bar, though the space bar is not visible on this keyboard.)

Of course, testing a keyboard against one book alone isn't fair - Anna Karenina, for instance, might have more A's and N's than the average book! And if we were to test using a physics textbook or a million lines of code, the results might look very different. (At the end of this interactive, you can run various tests yourself against different texts, to see if you can detect any important differences.)

DVORAK Key Click Patterns

The DVORAK keyboard layout, patented in 1936, was designed as an improvement over QWERTY. It was specifically designed to reduce errors and typing fatigue while speeding up typing. One primary strategy was to reduce the amount of finger travel since 70% of typing occurs on the "home" row - the middle row of keys. It also, therefore, reduces jumping of rows, and was meant to decrease the number of words that are typed with just one hand on QWERTY, such as the word "are". But for a variety of reasons, including that it only offered a 5% improvement over QWERTY, DVORAK never really took off.

It is clear from this visual that the fingers on the home row are doing most of the work when typing on a DVORAK keyboard. However, since all of the vowels are on the home row, the infrequently typed letter "U" is wasted "home" row space, and perhaps could be replaced with the more commonly typed letter "R". You can also see that the bottom row of letters is much more lightly used on DVORAK than on QWERTY, for good or bad, I don't know.

COLEMAK Key Click Patterns

The COLEMAK keyboard layout is very new - it was just designed in 2006. COLEMAK was designed to solve the same problems with QWERTY that DVORAK was targeting while also addressing some of DVORAK's issues. By making COLEMAK closer, conceptually, to QWERTY, it is supposed to be easier to learn by those who learned to type on QWERTY (i.e., the vast majority of us!) It also tries to solve some of DVORAK's other issues, such as the fact that it swapped the left-hand dominance of QWERTY for right-hand dominance. This lopsidedness is still a problem, COLEMAK afficionados claim. And it moves the "R" to the home row, which is a glaring error with the DVORAK layout!

Evaluating the visual, it does appear that COLEMAK is even more even-handed (pun intended) than DVORAK and QWERTY. And it also does an even better job, it seems, at keeping the fingers on the "home" row.

ABCDE Key Click Patterns

What if we decided to be less scientific and more alphabetical with our keyboard? What if we just placed the keys, in order, starting from the upper left?

This is surprisingly even-handed, but certainly does not take advantage of the "home" row to the same degree. So we can assume that fingers will be jumping around more, which is probably not a good thing. Let's investigate this jumping around thing next.

QWERTY Key Jumps

In this view, we're simply measuring when the same hand jumps rows - from the top to the bottom row or vice versa. The thickness of the line shows the number of jumps between two letters using the same hand. For instance, you can see that in Anna Karenina, at least, there is a huge number of jumps between the "I" and "N". Perhaps that's due to all of the present participle verbs - all those words ending in "ING".

JUMP Key Jump Travels

Come to think of it, jumping rows isn't just about awkward movements, though that's a huge part of it. It's also about distance traveled. In this visual, we're doing the math. Assuming it's about 1 1/2 inches between rows of a keyboard, this means that if you type the entire book of Anna Karenina, you end up traveling almost 3 1/2 miles in those jumps - enough to walk from the Lincoln Memorial to the U.S. Capitol building and nearly half way back again! That's tiring just to think about - imagine walking that distance, 1 1/2 inches at a time (on your fingers)!

INTERACT

Now it's time for you to play around a bit. Select what you want to see: "Key Click Patterns" for the key and finger frequency; "Key Jumps" for the connecting lines; and "Key Jump Travels" for the Lincoln Memorial to U.S. Capitol travel distance. Then select the text you want to analyze. Finally, select the keyboard you'd like to evaluate against. Note that the line thickness for "Key Jumps" is unique to each text/keyboard - there is no shared scale, you can't directly compare line thickness across the texts/keyboards - though you can compare patterns. Click "Go!" to see it in action.

CONCLUSION

These visuals should help explain why certain keyboards are easier or harder to use, and may even help you pick the right keyboard for you to use going forward. However, unless you're typing Anna Karenina or one of the other texts here, these tests may not reflect your actual results! If you're a computer programmer, your considerations may be quite different - numbers and different letter and punctuation keys would be important to test, which are not included here. And remember that there has been a decent amount of research into keyboard layouts and the jury is decidedly still out on the relative effectiveness of the options!

If you have a text - such as a long page of code or different types of English-based text that we can insert into this test, I'd love to have it. Please get in touch and let me know!