• You are not logged in.
  • Index
  • General
  • Seeking justification: "QWERTY-Colemak is easier than QWERTY-Dvorak"

    Seeking justification: "QWERTY-Colemak is easier than QWERTY-Dvorak"

    • Started by wibble
    • 20 Replies:
    • Reputation: 0
    • Registered: 27-Dec-2007
    • Posts: 21

    Let me first state that I am not anti-Colemak. If I'm anti-anything in respect of keyboard layouts, I'm most definitely anti-QWERTY. It's clear that Colemak is superior to QWERTY (let's face it, almost any rationally-designed layout is).

    What I'm having trouble with is the assertion that Colemak is easier to learn than, say, Dvorak (note that it's https://colemak.com/Easy_to_learn that itself makes several comparisons with Dvorak, almost inviting counter-argument).

    Apart from the shortcuts issue, which I agree is a valid point (although it can be addressed in other ways), the main justification for the 'easier to learn' claim appears to me to be due to the assertion that "Memorization isn't linear: fewer changes from the de-facto standard improves the ability to switch from it."

    It seems to me that this assertion, that Colemak>Dvorak because "Colemak moves only 17 keys, while Dvorak moves 33 keys," while logical-sounding, is flawed. A layout has to be learnt as a whole - a gestalt. It's inappropriate to make the comparison with learning a linear sequence of characters, since each key in the layout has a dynamic relationship with every other key.

    Even minor changes can throw things out of kilter. Consider the 'a' and 'o' keys on the Dvorak layout. Michael Capewell points out that these two letters should be switched, and I tend to agree, as I personally find it troublesome typing 'inboard stroke flow' and 'keyboard' on the ANSI Dvorak. (I'd love to know why Dr. Dvorak apparently broke the 'inboard stroke flow' rule on these two).

    As an experiment, I tried switching just these two characters around on my own (ANSI Dvorak) layout. Immediately, my typing speed dropped by a third, while my error rate soared by a factor of three. These two characters had only moved a single position (cf. "The keys in Colemak are just a key or two away from their QWERTY position"), and yet this simple change caused chaos, with knock-on effects as my fingers, that had learnt certain combinations, now fumbled almost constantly, and the comfortable flow was completely lost.

    I didn't persevere with the revision; I do acknowledge that it would take me less time to learn the revision than it did to learn Dvorak itself - but I'm only talking about two adjacent keys here, and it was clear that this apparently minor change had a substantial knock-on effect.

    This experiment convinced me of two things:

      1. Fine-tuning a '90% perfect' design isn't worth the effort (whereas switching to ANY sensibly-designed layout from QWERTY definitely is)
      2. If there's this much difficulty with simply swapping the positions of two *adjacent* keys, then it's going to be no more difficult to learn a new layout that moves half the keys than one that moves all of them.

    In a similar vein, it's claimed that the "hardest keys to learn are those that move between hands" (with another Dvorak comparison). This does not seem to me to hold water: I may be wrong (I quite often am). However, justification for this assertion would enhance credibility.

    I hope that this post invites debate on this topic - if indeed it has already been addressed elsewhere, I'd be grateful if you would point me to that.

    Offline
    • 0
    • Reputation: 210
    • From: Viken, Norway
    • Registered: 13-Dec-2006
    • Posts: 5,343

    I can only throw in my own experience: Colemak was much, much easier to learn than Dvorak. This proves nothing, but it's what I can do. You're really asking for a scientific study, and nobody I know of can afford that.  :(

    *** Learn Colemak in 2–5 steps with Tarmak! ***
    *** Check out my Big Bag of Keyboard Tricks for Win/Linux/TMK... ***

    Offline
    • 0
    • Reputation: 0
    • From: Köln, Germany
    • Registered: 01-Apr-2007
    • Posts: 264

    The edge colemak has over Dvorak concerning easy learning is that it's so similar to QWERTY... I would say it's actually quite obvious that Colemak is easier to learn than Dvorak if you have QWERTY experience. If we have a scientific study we get to pay loads to get told what we already know.

    Offline
    • 0
    • Reputation: 0
    • Registered: 27-Dec-2007
    • Posts: 21
    vilem said:

    The edge colemak has over Dvorak concerning easy learning is that it's so similar to QWERTY... I would say it's actually quite obvious that Colemak is easier to learn than Dvorak if you have QWERTY experience.

    It *sounds* obvious, but I don't believe that it's correct: it's my contention that a layout is a gestalt.

    Last edited by wibble (29-Jan-2008 18:09:25)
    Offline
    • 0
    • Reputation: 0
    • Registered: 27-Dec-2007
    • Posts: 21
    DreymaR said:

    I can only throw in my own experience: Colemak was much, much easier to learn than Dvorak.

    A question occurs: were you originally a QWERTY touch-typist, rather than a 'hunt-and-pecker'?

    I'm wondering whether the learning curve might in fact be substantially different for the two.

    Me, I was a QWERTY hunt-and-pecker for some thirty years. I finally (before encountering Colemak or any other variant) took the plunge and learned Dvorak late last year; and I learned how to touch-type it at the same time. I recall trying to learn how to touch-type (QWERTY, on a manual typewriter) as a teenager, and remember how boring that was; I only picked up a keyboard again when computers came along, and hunt-and-pecked merrily from then on.

    Offline
    • 0
    • Reputation: 210
    • From: Viken, Norway
    • Registered: 13-Dec-2006
    • Posts: 5,343

    I'm afraid I must contest that your gestalt theory sounds like balderdash to me. It's my personal opinion and not much more, but your hypothesis just doesn't ring true at all with me. I must ask what support you have for that?

    While a piano with many keys is a gestalt to me, it is only because the piano has a simple arrangement of those keys. A system of 26+ letter keys is more complex than a system of half as many, and no statement without solid backing can change that in my perception. Once a keyboard is fully learned it does become something else - but in the learning process the parts become very pertinent.

    It's already been well contested (the 'digits of Pi' analogy) in my opinion. Your argument that a key block isn't a sequential entity holds some merit, but not enough to break the analogy I feel.

    Your moving of two keys is a lesson in the effects of going "out of flow" as you say. It will indeed break the "gestalt" and put you in a mode where you become aware of the whole process. It's a bit like becoming conscious of how you walk I think. You'll have that whatever you do to your keyboard, but it doesn't really tell a lot about the whole learning process.

    I made a switch at the lower bottom half (as you've all been told a bajillion times by now, heh) and it immediately crashed my typing flow/speed even if it mostly affected rare letters. But in a good day of typing I was back in the flow, mostly because no fingerings had been changed and the number of keys changed was small. You cannot mean that a small and a large change are basically the same thing?

    I can chime in and support the hand-switching claim: When changing QWERTY -> Dvorak -> Colemak, I felt strongly that the keys moving between hands were harder and the keys staying on the same finger were simpler.

    I do feel that the number of keys moved alone isn't a full indicator of switching difficulty. For instance, some characteristic movements of much-used di-/trigraphs are a substantial part of the feel of a new layout. Luckily for me, Colemak is replete with happy moments in that respect. I actually feel that even when typing Norwegian there's less going "against the grain of the board" on Colemak than what I felt when typing Norwegian Dvorak. It's got a little to do with the appalling placements of the Norwegian wovels on the Norwegian Dvorak (a disheartening example of following a principle rather than sense), but more often it's about nice rolls. I feel that the constant hand switching on Dvorak was taken too far and there should've been more inward rolling. But I digress. The fact that the number of keys switched isn't a sole indicator doesn't mean that it doesn't feel important.

    *** Learn Colemak in 2–5 steps with Tarmak! ***
    *** Check out my Big Bag of Keyboard Tricks for Win/Linux/TMK... ***

    Offline
    • 0
    • Reputation: 210
    • From: Viken, Norway
    • Registered: 13-Dec-2006
    • Posts: 5,343
    wibble said:

    A question occurs: were you originally a QWERTY touch-typist, rather than a 'hunt-and-pecker'?

    I was a 60 WPM QWERTY touch typist, then managed to get Dvorak to a similar speed after much hardship, and now my record at Ryan Heise with Colemak is 65 WPM. So it seems to me that I've learnt all three layouts to a similar level of proficiency, more or less.

    *** Learn Colemak in 2–5 steps with Tarmak! ***
    *** Check out my Big Bag of Keyboard Tricks for Win/Linux/TMK... ***

    Offline
    • 0
    • Reputation: 2
    • From: Houston, Texas
    • Registered: 03-Jan-2007
    • Posts: 358
    wibble said:

    2. If there's this much difficulty with simply swapping the positions of two *adjacent* keys, then it's going to be no more difficult to learn a new layout that moves half the keys than one that moves all of them.

    Your personal example though only addressed the magnitude of disruption and not the rate of recovery. 

    The question of learning being easier would need to be answered by examining measures of recovery. 

    Obviously the magnitude of disruption forms a nonlinear relationship to the number of keys moved (saturation must occur).  You could imagine one explanation of your experience is that the magnitude of disruption rises very fast and saturates quickly.  In other words, that moving a few keys has as nearly as big effect on the initial loss of speed as a moving many keys.  The real question that needs addressing though is not the magnitude of this loss which might very well be about the same but how quickly does the recovery occur.   To someone making a switch, that big initial loss might be very disconcerting in either case causing them to abandon the change and conclude perhaps wrongly that either option  because of such a big loss will take too long to recover from.  It's possible to think that if the underlying perturbation were significantly smaller (layouts much closer alike) that the big loss would have a more rapid recovery to a new equilibrium (not far from the old equilibrium in some high dimensional space, sorry thinking in terms of nonlinear dynamics - branch of mathematics for complex systems ).

    So the real question is not about how much of a disruption there is to the gestalt or whatever but how fast does one recover from the disruption ?

    Some of the arguments Shai makes from a perspective of the neuromuscular system make some sense.  Cognitive neuroscience really isn't up to tracing exactly what happens when I recognize or think about the letters/words and then translate that to patterned movement of my fingers. A lot we don't understand.  I don't think any hard data exists on whether the rate of recovery for the switch of  Qwerty-Colemak is significantly shorter than Qwerty-Dvorak.  What we have either way is anecdotal. 

    It is clear that Dvorak gave no thought to this, he assumed his layout would be embraced and replace Qwerty and be  the first touch typing experience for the student.   Shai has given some thought to the switch in ways that make sense and until some fairly sophisticated study is done, I don't see  it going much further than that.

    Last edited by keyboard samurai (29-Jan-2008 22:50:41)
    Offline
    • 0
    • Reputation: 0
    • Registered: 07-Aug-2007
    • Posts: 69

    It looks like DreymaR at least -- and maybe some others -- switched from Dvorak, rather than QWERTY, to Colemak.  That would mean he had to re-learn almost all the keys afresh and nevertheless found Colemak easy to transition to.  So maybe closeness to QWERTY is not really significant: Colemak hitting a sweet spot may have nothing to do with how similar it is to QWERTY.    That's good news, I should think.   Similarity to QWERTY is just a bonus?

    Offline
    • 0
    • Reputation: 0
    • From: Bristol, UK
    • Registered: 08-May-2007
    • Posts: 12

    I switched from touch-typing QWERTY to Colemak and found it slower and more frustrating to learn than I'd expected, especially in the middle of the course of lessons. But as I got near the end i certainly found it helped a lot that many letters hadn't moved (QWZXCVBM,.).  I doubt that it made any *long-term* difference to my typing speed six months later, but it certainly helped my morale at the time, and I'm pretty sure it made the later stages quicker.

    I found the letters that had moved a short way such as S and G caused the most mistakes early on, so I'm not convinced that it's hardest to learn the keys that switch hands.

    That's just my personal recollection of my personal experience of course.

    Background info:
    I touch-typed QWERTY (not using completely standard fingering) for about 20 years at at least 60WPM (with quite a lot of errors) before switching to Colemak. I learnt Colemak using the TypeFaster typing tutor, taking a period of 3 weeks or so to learn all the letters.

    Offline
    • 0
    • Reputation: 0
    • From: Horsham, West Sussex, UK
    • Registered: 11-Jun-2007
    • Posts: 86

    Well I can't say much about the theory but I can give some thoughts based on experience.

    I first tried to go Dvorak in June 2000, I managed to get up to about 20 words per minute after three weeks or so. The first week was excruciatingly painful, by the end of the week I could still barely manage 10 wpm. My second attempt was in February 2006, it lasted a week and at the end of it I was still not much more than 20wpm. My third and fourth attempts, last summer, lasted about three months and in the end I managed to get up to about 35-40 words per minute on my Kinesis keyboard.

    At the time I had heard about Colemak, but I didn't take it all that seriously as anyone who came across my "noisy fanboys" foot in mouth outburst on my blog will be aware. However, when I did eventually give it a try, it took me about three days to see more progress than I had seen on Dvorak in three weeks. I actually managed to switch to Colemak completely using a "qwerty by day, Colemak by night" approach in the early stages, which meant that my productivity didn't suffer at all. I think that all told I was up to about 50wpm on it within three weeks.

    You simply can't do that with Dvorak -- everyone who says anything about it whatsoever agrees that you need to go cold turkey from day one.

    What I found made the biggest difference is that Colemak leaves the least frequently used keys in the same place. The fact that keys such as Q, W, Z and X move so dramatically on Dvorak was extremely irritating -- because these keys are used much less frequently, it takes much longer to get used to them. In Colemak, they stay in the same place.

    I'm now pretty much full time on Colemak now -- I do have occasional forays into qwerty when I have to rdp into various servers, but in general I tend to avoid it. As for Dvorak, forget it.

    The difference really is quite dramatic if you are an existing qwerty typist.

    Last edited by jammycakes (30-Jan-2008 22:45:49)
    Offline
    • 0
    • Reputation: 2
    • From: Houston, Texas
    • Registered: 03-Jan-2007
    • Posts: 358
    jammycakes said:

    At the time I had heard about Colemak, but I didn't take it all that seriously as anyone who came across my "noisy fanboys" foot in mouth outburst on my blog will be aware. However, when I did eventually give it a try, it took me about three days to see more progress than I had seen on Dvorak in three weeks. I actually managed to switch to Colemak completely using a "qwerty by day, Colemak by night" approach in the early stages, which meant that my productivity didn't suffer at all. I think that all told I was up to about 50wpm on it within three weeks.

    This seems to be a path that a number of people have taken in successfully adopting Colemak. I think it's noteworthy because in the last year I have seen a few newcomers pop on this site announcing they have gone cold turkey to Colemak and then within a few days or week say they can't afford the loss in productivity and quit.  So while I know Shai strongly recommends cold turkey, it would be interesting if it was easier for someone to learn Colemak why maintaining proficiency in Qwerty for productivity reasons at least as a transitional phase.  That would certainly be attractive to some as an option.

    This is a quote from a NY Times blog,

    http://freakonomics.blogs.nytimes.com/2 … vs-dvorak/  And don’t even think about using a QWERTY keyboard during your Dvorak training. You’ll only undo all the progress you’ve made.

    My question: Is there anyone who cares enough about a few extra words-per-minute who would be willing to give up the ability to type for a whole month? Learning the Rubik’s cube or baking a kitty litter cake might be time better spent.

    from a pro-dvorak website,

    http://www.dvortyboards.com/about.html  The downside to learning Dvorak is that QWERTY must be suppressed. There is a difficult time during the transition in which both sets of keyboard layouts are held in the brain, and may be confused while typing. Practice and commitment will overcome this temporary confusion. Some typists are able to maintain accuracy in both layouts, but we do not recommend that this be an expected outcome.

    Offline
    • 0
    • Reputation: 0
    • From: Horsham, West Sussex, UK
    • Registered: 11-Jun-2007
    • Posts: 86

    Perhaps the relevant pages in the website (e.g. the tips for learning on the "Learn" page) should be updated to indicate that this approach is an option? I'd also mention it on the "Easy to learn" page too. It's certainly a possible selling point for anyone who has tried Dvorak and given up because of weeks of lost productivity.

    Offline
    • 0
    • Reputation: 210
    • From: Viken, Norway
    • Registered: 13-Dec-2006
    • Posts: 5,343
    ds26gte said:

    It looks like DreymaR at least -- and maybe some others -- switched from Dvorak, rather than QWERTY, to Colemak.  That would mean he had to re-learn almost all the keys afresh and nevertheless found Colemak easy to transition to.  So maybe closeness to QWERTY is not really significant: Colemak hitting a sweet spot may have nothing to do with how similar it is to QWERTY.

    Oh, I wouldn't say that. It's not as if my QWERTY skills were fully unlearned at any point, even if I couldn't readily touch type in QWERTY without looking anymore by the time I learnt Colemak. (I'm not certain it'd be possible to unlearn something like that.) I felt that the old QWERTY motor memory helped well enough when learning Colemak. And it's easier to type on a QWERTY board now than it was when I was using Dvorak, too - it feels like a less dramatic transition.

    *** Learn Colemak in 2–5 steps with Tarmak! ***
    *** Check out my Big Bag of Keyboard Tricks for Win/Linux/TMK... ***

    Offline
    • 0
    • Shai
    • Administrator
    • Reputation: 36
    • Registered: 11-Dec-2005
    • Posts: 423

    Perhaps the relevant pages in the website (e.g. the tips for learning on the "Learn" page) should be updated to indicate that this approach is an option? I'd also mention it on the "Easy to learn" page too. It's certainly a possible selling point for anyone who has tried Dvorak and given up because of weeks of lost productivity.

    I've updated the 'Learn' and the 'Easy to learn' pages to include the "QWERTY by day, Colemak by night" approach.

    Offline
    • 0
    • Reputation: 0
    • Registered: 18-Mar-2008
    • Posts: 19

    When I was doing the ktouch typing lessons, my speed jumped by about 30-40 characters per minute (6-8 wpm) when I started doing lessons on keys did not move from qwerty. It seems to me that the less you have to fight muscle memory, the better. One possible disadvantage is that the overlapping makes it harder to separate the qwerty and the colemak memory, which could make it harder to maintain skills in both compared to qwerty vs. dvorak.

    Offline
    • 0
    • Reputation: 210
    • From: Viken, Norway
    • Registered: 13-Dec-2006
    • Posts: 5,343

    I don't think memory works quite like that. It's very "fuzzy-networky" meaning that you wouldn't quite keep separate Dvorak/Colemak/QWERTY muscle memories no matter what. They will flow into each other, compete for focus and do all sorts of interesting things. For instance, I can't remember my Dvorak anymore - but I suppose that if I started re-training Dvorak it'd come back and start to flow again. I have no idea what that'd do to my Colemak or QWERTY, but I suppose they'd hurt a little since the Dvorak would get more focus then. That may be part of what you mean by separation, but I don't think it's necessarily linked with key overlap.

    Maybe it's easier to separate two layouts that have only a few differences, instead of harder? That seems to be the message from all the successful "QWERTY by day, Colemak by night"-ers. All I've heard from the Dvorak people is that combining Dvorak and QWERTY shouldn't be attempted while learning Dvorak.

    *** Learn Colemak in 2–5 steps with Tarmak! ***
    *** Check out my Big Bag of Keyboard Tricks for Win/Linux/TMK... ***

    Offline
    • 0
    • Reputation: 0
    • From: Australia
    • Registered: 22-Oct-2007
    • Posts: 47
    DreymaR said:

    Maybe it's easier to separate two layouts that have only a few differences, instead of harder?

    My theory is that this is true when the alphabets are the same. It is the same problem as trying to remember two phone numbers. If they are rearrangements of the very same digits (i.e. the same alphabet), then just hearing the second number may interfere with your memory of the first number. But if the digits of the second number do not intersect with the digits of the first number, then it is less likely that interference will occur.

    I like to think that this theory explains why learning Dvorak interfered with my QWERTY memory, but learning the Korean layout did not interfere with my QWERTY memory. The alphabets did not intersect.

    Offline
    • 0
    • Reputation: 0
    • Registered: 18-Mar-2008
    • Posts: 19

    Not to throw in even more anecdotal evidence, but  on some article about the dvorak layout the writer said that learning dvorak had no effect on his/her qwerty speed.  Unfortunately, in order to do a real study of this, we would have to take large samples of people, seperate them into groups, and teach one colemak, and the other dvorak, and then study their qwerty speeds.

    Offline
    • 0
    • Reputation: 210
    • From: Viken, Norway
    • Registered: 13-Dec-2006
    • Posts: 5,343

    You're right, we're getting awfully hearsaying-based here I'm afraid. I'm sure some people will have a problem juggling both Colemak/Dvorak and QWERTY in their heads and some people won't, and as you say only a proper study could really determine the difference with any certainty. But I at least, have heard a lot more Colemak users tell about dual proficiency than I've heard Dvorak users say the same. Which could theoretically be a matter of memes on this board, underrepresentation from the Dvorak users etc etc. Yeah. All we can really do is make (hopefully) intelligent guesses right now.

    Last edited by DreymaR (08-Apr-2008 17:25:21)

    *** Learn Colemak in 2–5 steps with Tarmak! ***
    *** Check out my Big Bag of Keyboard Tricks for Win/Linux/TMK... ***

    Offline
    • 0
    • Reputation: 4
    • Registered: 08-Dec-2010
    • Posts: 656

    I am a Qwerty typist with 60wpm speed and now I am learning Colemak for a week, 22wpm, 95% accuracy.

    Right now I have the same confusion exprience: funny and sometimes exhausting errors while transition. P, R, S, U, O, Y, are often mistyped. I agree with wibble that a mistake's still a mistake, no matter the mentioned key moves within same hand or to other hand.

    Bet this is an assurance for the switchers, but you can only realize its merit when you went through the transition yourself.

    Hopefully, with practice things will improve. The brain works wonders, it just needs sufficient time. To reach 60wpm in Qwerty in the past I needed a full year.

    I think we should have practical expectations that the transition takes longer than we expected. At least six months or a year.

    Last edited by Tony_VN (13-Dec-2010 11:10:05)
    Offline
    • 0
      • Index
      • General
      • Seeking justification: "QWERTY-Colemak is easier than QWERTY-Dvorak"