It was designed to put diphthongs on opposite sides of the keyboard so that a mechanical typewriter wouldn’t jam as frequently. I think avoiding “jams” with my thumbs is definitely the way to think about it.
All of this is speculation because I haven’t had the chance to try it our yet ( ͡° ͜ʖ ͡°)
https://news.ycombinator.com/item?id=7110619 , but the article is broken, so see https://web.archive.org/web/20150320170000/http://minuum.com... .
Having all the vowels next to each other, and so many words in English that differ just by vowels, is actually really hard on the swiping algorithm. As I mention in the comments of that article, it is likely that QWERTY is not optimal for that use case either, but it's a lot closer. Despite using Dvorak for almost everything else, including typing this comment right now, I have no use or desire for it on mobile phones. YMMV, of course.
I wonder if the same could be done for a swipe based phone layout. Seems like you'd end up with something very different than Dvorak and probably quite different than Qwerty as well. Perhaps part of the metric could give a bonus to layouts that were similar to some other layout, thus creating Qwerty-phone and Dvorak-phone alternative layouts.
As someone who types with Dvorak on the computer, I'd happily attempt to learn a new layout for phones. My error rate is extremely high with swipe keyboards, even though I use Qwerty.
[1] https://web.archive.org/web/20220926205948/https://mkweb.bcg...
There are others, but the one that comes to mind for smartphones and other small devices is MessagEase. [1]
They offer apps for Android, iOS and Windows. I use the Android version. I love using it although it is sometimes buggy (display gets stuck on letters/digits and you have to remember where the digit/letter you want to enter is located) and backspacing over a lot of text (instead of selecting and cutting it) is always very slow.
It faced the same problem as all new keyboard designs, getting people to use it.
If you are on Android, you can find plenty of research project keyboards that are indeed more efficient to use, after you get over the learning curve.
Your _muscle_ memory makes you good at moving your fingers (or thumbs) to an absolute position on your keyboard (or screen) - but to do that you must first decide what absolute position your finger (or thumb) should move to (where is the J key?).
Qwerty is useful for "swipe typing" on a touch screen and Dvorak/Colemak is great for touch typing on hardware and the way "muscle memory" works those are such different media/muscle movements that they have separate "muscle memory".
I don't know, I'm not an iOS user but it would hugely inconvenience me if the only way I could use Dvorak on Android is to use a keyboard with a hardware Dvorak layout.
Edit: Indeed, it appears that iOS supports Dvorak layout for external keyboards already; this is supporting the native on-screen keyboard
This is a little ambiguous, I assume you mean a (rare) hardware Dvorak keyboard, not that iOS supports remapping a standard ANSI external keyboard to Dvorak - but correct me if that's wrong.
A better layout for smartphones, if one had to design it, would try to put all characters in two circles around the thumbs, while putting symbols and the least frequent ones towards the centre, requiring some extension of the thumbs to reach.
I've been learning to use a split ortholinear keyboard, and it exercises a different type of muscle memory than a regular keyboard, which is even different than on-screen keyboards. I have a different layout in each and I haven't lost any speed whatsoever.
I quickly realized that my brain would use dvorak for my mechanical keyboard at home but would only use qwerty on the soft keyboards at school.
Available on iPads, but not on iPhones at this time.
One downside was that it seemed to have memory issues but this was years ago on a Samsung GS5.
There are 9 base consonants (K S T N H M Y R W) and 5 vowels (A I U E O), and every syllable is either vowel or consonant + vowel. Digits 0-9 on the number pad correspond with the consonants (including "no consonant"), and then you swipe in a different direction for a different vowel (4 vowels for swiping in a direction, 1 vowel for not swiping).
Nice, big chunky buttons you can press with one thumb on one hand, and each time you press one of those big, chunky buttons, you get an entire syllable. After a little bit of practice, I can type on this layout much faster than I can type in English, even though my Japanese is not very good.
Words that use unvoiced, full-size kana are super fast (especially if it has a bunch of あ vowels like あたたかい, which takes 4 taps and 1 swipe instead of 8 taps in romaji), but words with lots of voiced consonants and small kana need you to tap the modifier button a bunch (e.g. ざっぴ requires seven taps+swipes, instead of just 5 taps on the romaji input).
You can consistently type without looking at the keyboard once you learn the layout, which I've found to be great for note-taking
Some years ago, I set up my Android phone to use Dvorak layout but I quickly reverted to QWERTY, without being able to use my muscle memory, it was actually a very frustrating experience.
Because most Dvorak users do not use a keyboard labeled with the dvorak layout, you never really learn visually where the keys are and you only know the layout in your muscle memory which does not translate at all to using a touchscreen keyboard on a phone.
Now maybe the next generation will find that Dvorak on iOS is no worse, and if they use it they can avoid learning a second layout. But for existing Dvorak users who also were forced to learn QWERTY at some point, it's hard to see many switching over.
https://userinterfaces.aalto.fi/typing37k/resources/Mobile_t...
> Finger usage: Participants who reported to use two fingers were significantly faster than those who used only one finger (M = 37.7, SD = 13.2 versus M = 29.2, SD = 10.7, p < 0.001, d = 0.66). A closer look at the reported typing posture shows that the use of different hands and fingers had a significant impact on performance. Over 82% of participants typed using two thumbs. Confirming the findings of prior work [3 , 7], this was the fastest way to enter text
Turns out I'm weird it seems you should and indeed most people do use two thumbs. Given that fact your theory seems extremely plausible.
Not the home row thing, but the fact that vowels are on the left hand side, so that two thumbs will alternate as much as possible (alternating left and right hand fingers being an explicit goal of Dvorak).
QWERTY if anything seems to be about being convenient for transcribing Morse code. Letters with a similar Morse code are grouped etc.