Should you switch keyboard layout?

Every typing forum eventually arrives at the same argument. Here is what the evidence actually supports.

The QWERTY origin story is half a myth

The version everyone repeats is that QWERTY was designed to slow typists down so mechanical typewriters would not jam.

The truth is close but importantly different. The layout was arranged to reduce jamming by separating letter pairs that commonly follow one another, so their typebars approached from different angles and were less likely to clash. That is not the same as slowing the typist — it was an attempt to let them go faster without the machine failing.

The distinction matters because the “deliberately slow” version implies the layout is arbitrary and therefore easy to improve on. The real story implies it was optimised for a constraint that no longer exists — which is true, and a much weaker argument for switching than it sounds.

The famous Dvorak study is not trustworthy

The evidence usually cited for Dvorak’s superiority is a set of wartime US Navy studies reporting dramatic retraining gains.

Two problems, and they are serious. They were conducted under the supervision of August Dvorak himself, who held the patent. And the methodology has been criticised repeatedly since — most notably in a 1990 economics paper that went back to the records and found the comparisons did not support the conclusions drawn from them.

Later independent work, including studies run by the US General Services Administration, found much smaller differences: enough that Dvorak may be slightly faster and considerably less finger-travel-intensive, nowhere near enough to justify the retraining for most people.

What switching actually costs

This is the part usually skipped, and it is the deciding factor.

• Several weeks at roughly 20 words per minute while the new layout becomes automatic. If you type for a living, that is a real cost with a real number attached.
• Slowness on every machine that is not yours. A colleague’s laptop, a library computer, a friend’s desk.
• Keyboard shortcuts scatter. Cut, copy and paste are placed where they are because of QWERTY. On Dvorak they are no longer under one hand.
• Your QWERTY ability degrades. Most people who switch fully do not keep both.

Set against a speed gain that independent evidence puts somewhere between small and indistinguishable.

Colemak, and the case that is actually reasonable

Colemak was designed later and deliberately as a compromise: it changes 17 keys rather than nearly all of them, and it leaves Z, X, C and V — the shortcut cluster — where they are.

It is the sensible option if you are switching, and the honest case for it is not speed. It is finger travel. Colemak and Dvorak both keep far more typing on the home row than QWERTY, which means less movement for the same text.

Whether less movement means less strain is plausible and less well established than its advocates suggest. If your hands hurt, that is worth investigating — but investigate it with someone qualified rather than by changing layout and hoping.

What to do instead, if speed is the goal

Everything here beats a layout change on evidence and on cost:

Fix accuracy first. Every error costs about a second — noticing, backspacing, retyping — which at 60wpm is a whole word. Slowing down until you are consistently above 97% accuracy usually returns higher net speed within a fortnight.

Stop looking at the keyboard. The single most common ceiling for an intermediate typist, and uncomfortable to break for about a week.

Drill your specific problem pairs. Your consistency score on the typing test points at them: a decent average with low consistency almost always means a handful of combinations are stalling you, not general slowness.

Sort out the ergonomics. Wrists straight, screen at eye height, shoulders down. It will not make you faster this month; it is what lets you keep typing for years.

What the layouts actually change

Worth being concrete, since the debate is usually conducted in adjectives.

QWERTY places several very common letters away from the home row — E, T and R all require reaching, and the left hand does noticeably more work than the right for English text. Alternative layouts move the commonest letters onto the home row, so more typing happens without the fingers leaving their resting position.

That reduces total finger travel by a substantial proportion on English prose. It is a real, measurable difference, and it is the honest basis of the ergonomic argument.

Whether it makes you faster is a separate question, and the answer appears to be barely. Speed for a proficient typist is limited more by how quickly you can read ahead and by how often you have to correct than by how far your fingers move — which is the same conclusion the typing speed guide reaches from the other direction.

The path-dependence argument, and why it is overstated

QWERTY is the standard example in economics of a technology that persisted because it arrived first rather than because it was best — the lock-in story. Everyone learns it because everyone knows it, so a better layout cannot displace it.

The 1990 paper that examined the Navy studies also went after this, and its argument is worth knowing: lock-in only matters if the alternative is meaningfully better, and the evidence that it is turns out to be thin. If Dvorak offered a large gain, the story of an entire industry foregoing it would need explaining. If it offers a small one, no explanation is needed.

The honest position is somewhere in between. QWERTY almost certainly is not optimal, keyboard layouts genuinely are path-dependent, and the practical gain available from fixing it is small enough that the theory is more interesting than the switch.

What about split and mechanical keyboards?

A more useful question than layout, and a different one.

Split and tented keyboards change wrist and forearm posture rather than finger travel. If you have discomfort, the ergonomic case here is stronger than the case for any layout change — you are altering the geometry of the joints that hurt.

Mechanical switches change how the keys feel and sound. Typing speed differences between switch types are small and highly individual; what people mostly report is preference. That is a legitimate reason to buy one and not a performance argument.

Key travel does affect speed measurably. A laptop’s shallow travel and a full-depth keyboard produce genuinely different numbers for the same person — often ten words per minute apart — which is why comparing a score across machines tells you about the keyboards.

If you are going to spend money on typing, spend it on posture before layout.

Who should actually switch

A short list:

• Your hands hurt, you have taken advice, and reducing finger travel is part of a plan rather than the whole of it.
• You are learning to touch type from scratch and have no QWERTY habit to unlearn. The cost is then close to zero, and you may as well pick the better-designed layout.
• You find it genuinely interesting. This is a perfectly good reason and more honest than most of the ones offered. Just do not expect it to pay for itself in speed.

For everyone else, the answer is no — and the reason is not that QWERTY is good. It is that the gain is small, the cost is certain, and the same weeks spent on accuracy would produce more.