4

I am a computer science student. I mostly work with computers and I am really wondering about the layout of keys in a keyboard. It is comfortable and easy. My question is this: what is the basic logic behind the key layouts in keyboard? Who presented this concept?


Clarification: Let's limit the scope of this question to the QWERTY keyboard layout. This is consistent with the two answers already given.

To avoid the question from being too broad, the history of AZERTY, Dvorak, and other commercially-sold layouts ought to be separate questions.

One could also design their own custom layout. The question of how one would make the choices to do that is primarily opinion-based, and therefore unsuitable for a StackExchange question.

  • 2
    (Hmm. This is from before the computer-keyboard era... so is it on-topic?) Welcome to Retrocomputing Stack Exchange. Please read the tour to familiarise yourself with how this site works; we're a little different to most forums. – wizzwizz4 Nov 28 '18 at 17:42
  • 1
    Welcome to Retrocomputing! I'm voting to keep this question open, as it is related to computer history and we can use the questions. Although wizzwizz's point that it occurred before computers is true, an asker wouldn't know that until the question is answered. However, please do follow Chenmunka's comment, and edit the question to specify which keyboard layout is of concern here. Thanks! – DrSheldon Nov 28 '18 at 19:59
  • 1
    @wizzwizz4 Somewhat - but I would attribute poor quality, as it's a topic that has been beaten to death in numerous places - not at least in the coresponding wiki entries: en.wikipedia.org/wiki/Keyboard_layout#History – Raffzahn Nov 28 '18 at 20:37
  • 6
    In an alternate universe, computers could have inherited Linotype keyboards, used for typesetting newspapers, books, which were sorted by letter frequency : ETAOINSHRLDU...QWERTY is allegedly optimised for the mechanical contstraints of early typewriters. – TEMLIB Nov 29 '18 at 0:20
  • 2
    Just voted this as on topic. It is about the really early days (that probably very few, if any here remember). A lot of things in modern computers comes from the time before computers. For instance the question earlier about paper width and 80 vs 132 characters per line. – UncleBod Nov 29 '18 at 7:30
8

An article from the Smithsonian museum suggests a more complex history, and describes the "avoiding common letter pairs" as debunked by the fact that ER is a very common pairing. It suggests that perhaps the profitable training courses helped lock the users to particular manufacturers, and also describes some 2011 research which suggests that Morse code transcribers, an important early set of users, drove some of the design.

https://www.smithsonianmag.com/arts-culture/fact-of-fiction-the-legend-of-the-qwerty-keyboard-49863249/?no-ist

The history of typewriters is well studied by industrial historians. For a trail into that world, this academic journal article covers the literature Kroemer, K. UAIS (2001) 1: 99. "Keyboards and keying An annotated bibliography of the literature from 1878 to 1999" https://doi.org/10.1007/s102090100012

  • 6
    Can you summarize the article in your own words? Link-only answers are discouraged on StackExchange. – DrSheldon Nov 29 '18 at 1:48
  • I had a bit of a laugh when reading that Smithsonian text - It tries to plant the idea of Sholes having invented the typewriter without explicitly saying so (and ignoring lots of previously existing machines). I'd have assumed a decent institute like the Smithsonian would leave credit where credit is due... – tofro Nov 29 '18 at 8:37
  • @DrSheldon thanks for pointer, added summaries – jonathanjo Nov 29 '18 at 16:49
  • I'm a long way convinced by this article. Some of the references it cites are somewhat dubious (e.g. a link to a slashdot article about Dvorak vs Qwerty, which itself cites a widely-debunked article published by researchers known to promoting a political agenda). It contains numerous non-sequiteurs: yes, it seems reasonable that optimizing the keyboard for Morse transcription was an important goal, but that doesn't mean it was the only or even the primary goal of the design. The adjacency of ER is irrelevant, as it is vertical alignments that would need breaking up, not horizontal. Etc. – Jules Dec 1 '18 at 4:17
  • @tofro, You're talking about the same Smithsonian Institution that, for forty years and, for political reasons, refused to acknowledge that a couple of bicycle builders from the midwest had built the world's first powered aircraft that was capable of carrying a human. wright-brothers.org/History_Wing/History_of_the_Airplane/… – Solomon Slow Dec 3 '18 at 3:07
7

The keyboard layout was invented before computers - it was already used with mechanical typewriters that did not even use electricity!

I have read that such mechanical typewriters could get problems (maybe even get damaged) when certain keys were pressed quickly after each other.

Let's say the first two keys ("Q" and "W" on modern keyboards) are such a pair of keys which may cause problems if pressed quickly after each other.

If the keyboard layout was "QUERTY" instead of "QWERTY" it would happen very often that the keys "Q" and "U" are typed quickly after each other (because in European languages every "Q" is followed by an "U") while it nearly never happens that a "Q" is following a "W".

So using a mechanical typewriter you would have problems very, very often using a "QUERTY" layout but a "QWERTY" layout does not cause problems.

I've read that typewriter manufacturers investigated which letters often occur after each other and designed the keyboard layout in a way that such letters are explicitly placed on keys which could be pressed shortly after each other without causing problems.

Later on - when electrical typewriters and computers were introduced - they did not change the keyboard layouts because people were already used the existing layouts.

  • 3
    The ancestor to the modern typewriter used half of a circle to hold type bars for the top two rows of keys, ordered in zig-zag fashion, and the other half of a circle for the bottom two rows. Of all the pairs of consecutive letters, only two appeared in any common words: "SC" [since the positions of X and C were the opposite of today's keyboards] and "ZA", the latter appearing only in "pizza". The positions of "S" and "C" were swapped very soon after that early prototype, perhaps for "SCientific" reasons if you want to get into "specifiCS". I think that late swap is rather... – supercat Nov 28 '18 at 18:47
  • 1
    ...strongly indicative of why the keyboard is as it is. – supercat Nov 28 '18 at 18:47
  • @supercat The order of levers is the same as the order of keys projected to a horizontal line. The levers for Z and A would be separated by the lever for 2. The adjacent keys on the same row are separated by 3 levers. Examples of frequent pairs that correspond to adjacent levers are ED and FR. – Leo B. Nov 29 '18 at 1:55
  • @LeoB.: According to widespacer.blogspot.com/2015/11/… the earliest typewriters sent the top two rows to back half of a circular basket, and the bottom two rows to the front half, so each half would have its own independent set of potential jamming combinations. Note that S-C are not adjacent on later typewriters, but were on the original Sholes keyboard, and some common letter combinations like E-D are adjacent on modern keyboards but not on the original Sholes design. – supercat Nov 29 '18 at 15:54
  • @LeoB.: I do find it somewhat curious that when the type bars were re-ordered, the keyboard layout wasn't changed to accommodate that. I also wonder if the strict finger usage of touch typing evolved to make it impossible to type adjacent keys on the middle two rows too quickly,which could otherwise cause "FRequent" jams. – supercat Nov 29 '18 at 17:22
0

Reason is simple: this layout had shown the best typing performance using typewriter.

As typist is measured by the speed of typing and low error rate, the best layout which was selling best at that time is the one modern keyboards are having.

I read about it somewhere a time ago, I am sure you can find this research on the internet.

That is totally logical. And totally wrong.

@manassehkatz - I suspect that you are 100% totally right, and I am totally wrong! Thank you for comment!

In the "defense" of my answer I would say that originally I said about "speed of typing and low error rate", thus maybe this layout allowed achieving balance of both to some practical extent.

  • 1
    That is totally logical. And totally wrong. The QWERTY layout was originally designed to slow things down in various ways to prevent mechanisms from jamming. It is far from an optimal layout. It is the "fastest" layout only because we have been using it - generation after generation - for so long. It is far from the best layout. Inertia is the most powerful force in the universe. – manassehkatz-Reinstate Monica Dec 2 '18 at 15:20
  • @manassehkatz Actually, it's not true. The QWERTY layout was possibly designed to make it possible to type at speed without jamming the mechanism, or possibly not. The whole subject is accreted with urban myths, of which your version is just one. – JeremyP Dec 5 '18 at 10:12
0

My understanding was that typewriters where originally laid out as ABC....etc but the typists became very quick at doing this. On a mechanical typewriter very quick sequences of key strokes would end up with letter "heads" or "stamps" (not sure what they are called) being swung towards the page at nearly the same time which would end up with them jamming together.

The QWERTY style keyboard was made to slow the user down, of course this did not work as muscle memory does not care about such things and users quickly adapted to using the new layout - but by the time this mattered things had moved on.

To put it another way, it was simply not possible to type as fast on an old style mechanical swing arm typewriter as it is with a modern keyboard so they tried to find ways to slow them down. I also believe that the QWERTY layout was actually picked to be as awkward as possible and NOT (as has been suggested) because it made it quicker or more sense.

I did not research my answer so this may be myth but would be nice to see if right or wrong.

  • 2
    I believe (but don't have a reference to hand -- possibly those in earlier answers cover this) that the design wasn't "to slow typists down" (as is popularly believed) but to allow them to type at full speed without jamming. In other words, with QWERTY, they didn't need to slow down because (a trained) typist would no longer cause jams. – TripeHound Dec 3 '18 at 11:48

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.