KEYSTORM Play Now

KEYBOARD LAYOUTS

QWERTY, Dvorak and Colemak compared

QWERTY was arranged to stop typewriter bars jamming, and nothing has jammed since about 1975. That makes the alternatives sound like an easy win. In practice they reduce how far your fingers travel, they cost you a month of being slow, and they probably will not make you faster.

Where QWERTY came from

Christopher Latham Sholes arranged it in the early 1870s for the first commercially successful typewriter. The popular story is that he deliberately slowed typists down. That is a distortion of something true. On those machines each key swung a metal typebar up to strike the ribbon, and two bars whose letters were struck in quick succession could collide and jam. Sholes separated the letter pairs that appear most often in English so their typebars would come from different directions.

So the goal was not slowness, it was avoiding jams, and the mechanism it was solving for has not existed in any keyboard made in the last fifty years. QWERTY survives for the same reason railway gauges do. Once a standard is in every school, every office and every phone, being better is not enough to displace it.

What is measurably wrong with it

The interesting critique of QWERTY is not historical, it is mechanical. Three things are genuinely awkward about it:

The usual summary figure is home row usage: the share of keystrokes you can make without leaving the resting position. The commonly quoted numbers are about 32 percent for QWERTY, around 70 percent for Dvorak and around 74 percent for Colemak. Treat these as rough, because the answer depends entirely on which body of text you measure against, and figures published by a layout's own advocates tend to use text that flatters it.

Dvorak

Patented in 1936 by August Dvorak and William Dealey after several years of studying letter frequency and finger motion. The design rule is easy to state: all five vowels and the most common consonants sit on the home row, vowels under the left hand and consonants under the right, so ordinary English text alternates hands naturally.

It genuinely does what it claims. Finger travel drops sharply and long words often stay entirely on the home row. The catch is that it moves almost every key, so nothing you know transfers. The shortcuts are the sharpest version of that problem: Ctrl+C and Ctrl+V end up in awkward places, which matters more in daily work than it sounds like it should.

Colemak

Released in 2006 by Shai Coleman, and designed with the benefit of seventy years of hindsight about why people fail to switch to Dvorak. It moves only 17 keys from their QWERTY positions and deliberately leaves several things alone:

The result is the highest home row usage of the three and a much cheaper migration than Dvorak. It is the layout most people who switch today end up on.

Workman, and the ortholinear crowd

Workman, published by OJ Bucao in 2010, argues that home row usage is the wrong thing to optimise. Its case is that reaching straight up with the index finger is easy while reaching sideways or curling the little finger is not, so the correct target is comfortable travel rather than minimal travel. Colemak-DH makes a similar correction for the awkward inward reaches.

These distinctions matter most to people using columnar or split keyboards, where the keys are in straight columns rather than the staggered rows inherited from typewriter linkages. On a standard staggered board the differences between Colemak, Workman and Colemak-DH are small enough that choosing between them is not where your time is best spent.

LayoutYearKeys moved from QWERTYMain argument
QWERTY1873NoneIt is what everything already uses
Dvorak1936Nearly allHand alternation, vowels under one hand
Colemak200617Most of the gain, far less to relearn
Workman2010Around 20Comfortable reaches beat short ones

Will a new layout make you faster?

An alternative layout will probably not make you faster, and anyone who tells you otherwise is selling something.

Above roughly 60 words per minute your speed depends on how quickly you recognise a word and dispatch it as one motion, and on how often you stop to correct something. Both of those improve with practice whatever layout you use. The world record holders on QWERTY are faster than almost every Dvorak typist alive, which would be impossible if layout were the limiting factor.

What an alternative layout does deliver is less finger travel for the same text, which for some people means noticeably less fatigue over a long working day. That is a real benefit. It is just a different benefit from the one usually advertised.

The cost is four to eight weeks. Expect to drop to about a quarter of your current speed on day one and to need a month or more of daily use before you are back to where you started. During that time you are slow at work, and you will find that switching between your own laptop and any other computer becomes unpleasant. Nobody should start this in a busy month.

Who should actually switch

Switching is a reasonable decision if:

It is a bad idea if:

If you do switch

Change the layout in your operating system rather than buying new keycaps. Every major system ships Dvorak and most ship Colemak, and remapping in software means the change follows your account to any keyboard. Leaving the printed letters wrong is an advantage, because it removes the option of looking down.

Go all at once rather than gradually. Maintaining both layouts keeps you at the start of both learning curves. Then use it for real work as soon as you can bear to, because drills teach the positions and real text is where the speed comes from.

Layout is not your bottleneck

Accuracy on your own weak keys almost certainly is. KEYSTORM measures each one separately.

Play KEYSTORM