I'm typing Dvorak since almost ten years now and from what I remember it's damn hard to re-learn the layout (was touch typing QWERTZ before). The hard part in not even the initial memorization of the keys, but the old reflexes that kick in once you stop actively focusing on each key and let your subconscious take over.
I did not regret my decision (far from it), as Dvorak is indeed nice to type and quite readily available out of the box.
Normally you don't get much benefit from switching layouts though. Especially as developer. As an author or someone who is writing prolonged texts in plain English, yes it's definitely an advantage. Yes, it's also more fun to type. But really, QWERTY is good enough in most cases. You don't really need to bother.
100% of alternative layout users i know are developers.
100% of the layouts still treat symbols as second class citizens.
my typing would vastly improve if a keyboard simply added another row for symbols so they are one key press only.
changing layouts only help in a few cases and turns using somebodyelses computer or a phone or a tablet or a virtual keyboard or remote vim sessions, etc a royal pain in the ass, because they will be qwerty.
I've been using a keyboard layout I made with Ukulele that swaps the symbols with the numbers and I couldn't be happier. You get used to it in less than a day of typing and it makes all the () $ @ ! so much easier. It makes so much sense as a programmer when symbols are typed so much more frequently than numbers.
I use programmer dvorak, which also has swapped numbers and symbols and it's a blessing. The good thing about this one is also changes layout of the symbols so it's really a lot nicer to use.
Whoah I never even thought of doing that. It seems like it would work, I mean I really don't type numbers all that often these days... hmmm. I might chuck it on my laptop and see how I dig it for dev. Tempted to swap [] and {} as well. It's a similar concept to swapping F-keys for media keys, which I have and much prefer.
This is exactly what I've done with KeyRemap4MacBook. It has built in options for switching around numbers and symbols, as well as changing {} and []. Makes so much more sense for a programmer, and you can make multiple sets of options so you can disable them if someone else needs to use your machine for a bit.
any idea how to combine this with vim/vimperator/pentadactyl?
it seems to work with ^ but with the vim plugins, following a link requires numbers. pressing shift every time is less ideal. and in vim you use numbers quite a bit too (for moving and editing).
I'm using Vimium and it uses from one to three letters as a label for a link. I used Vrome earlier where the links where numbered. I find Vimium scheme much faster. Especially because the labels are generated intelligently, from adjacent (and preferably home row) keys. I never missed an ability to write a bit of text from the link to select it - instead of text, which can be both longer and less convenient to type, I have short labels, displayed over the beginning of link when I press F. I don't need to look down on keyboard or move my hands to numpad.
So I suggest you try this mode and see for yourself if it suits you.
thanks. i actually used homerow instead of numbers for a month or two. for some odd reason, i find it quicker to discern what combination i have to press with numbers. 'dk' is harder to pick up then '18' for me. or on company websites you can just do 'fconta' and off it goes. you dont even have to look for the link
> I don't need to look down on keyboard or move my hands to numpad.
that does not happen for me anyways. i use my numbers too much (e.g. switching windows/tabs).
No, just use uppercaps letters for hintkeys :) That way, you can still type to narrow down the links, and then press shift and select whatever link you want with homerow.
The irony, of course, is that programmers like myself adopt the US layout (instead of the local one, German in my case). To access \[];'/ without breaking fingers, having 'easy' access to @{}|:"
I do understand that there should be room for improvement. The situation is just so much worse if you look at other layouts as base line..
What I do is I have a second layout just for symbols which I toggle with Alt Gr using my right thumb. This way I can stay in the home row and accessing symbols is almost as fast as letters. You'll need a wide layout with 12 home row keys for this to be comfortable though (my right index rests at the l instead of j).
As a bonus I map Shift to the right Win key so that I can press it with my thumb as well. Using the thumb for Ctrl is also doable but it's a little annoying when you need to use the same hand for the combo.
Take a look at programmer's dvorak [1]. It treats numbers as second class citizens in favour of symbols. It comes pre-installed on some linux distros, but it won't solve your querty problem.
It's not completely what you're looking for, but I use this[0]. I think it's great and I've been using this and only this while coding for about 8 years or something. The symbols are typed with Alt.Gr. + the keys specified.
It seems like an algorithm to design a custom layout should be possible. You keylog what you type for a few days, then the algorithm could give you a few layout suggestions based on your keypress stats.
I was touch typing QWERTY before I switched to Dvorak and then to Programmer Dvorak and I agree that old habits are hardest part while you are learning new layout. I think that Programmer Dvorak is really useful for developers. I don't have any regrets. I don't have any pain in my hands any more. I'm more productive. And I think that at least everyone should try something different. QWERTY is good enough but isn't perfect IMHO.
As a keyboard layout aficionado, I appreciate the thinking that went into this keyboard layout.
I've been a fulltime dvorak user for over 10 years. I also tried Colemak but found that many motions felt awkward while using it.
Anyways, I thought I'd give this one a try just for the hell of it. And... so far I am quite pleasantly surprised. I fully memorized the layout in about half an hour. It feels comfortable to type on, just like Dvorak does.
Typed up this comment with the Workman-P layout, and seriously considering switching to it now...
The Norman layout [1] beats Workman at it's own metrics (or so it says on the tin), and from brief experimentation makes more sense to me.
I've always wanted to try an alternate layout, and this post led me to find minimak [2]. I'm typing with it right now and really enjoying the similarity to qwerty (only 4 keys changed) with reduced movement, plus 99% of my shortcuts remain the same.
I'm tempted to try out minimak. My only concern is that... do the numbers that get thrown around about percentage improvement only apply to touch typists?
I've never been strict enough with myself to learn touch typing (I still hit 120WPM so speed isn't an issue) and worry that the benefit won't be as obvious.
The metric I lean on the most is the one published by Andong at http://www.andong.co.uk/dvorak/Default.aspx. I guess that's odd since I spent the least amount of time talking about it and put it at the bottom of the page. It is focused on metrics of efficiency and effort. If you wanted speed, you might want to look at things like finger distance and home row usage more closely, in which case Arensito or Colemak might have an edge.
I must confess, since I swapped R and H in revision 2, Workman usually gets the edge on Norman using Workman metrics, depending on the input text. I guess I'll hunt down whatever text revisions are appropriate on normanlayout.info.
Summary of that article:
The 'QWERTY was developed to slow down the typist' story is a lie. There is no serious evidence that Dvorak is a 'better' layout. If you were starting from scratch, there would probably be no advantage choosing one over the other, except that QWERTY is the standard. Retraining is a waste of time.
One of the key quotes:
> The study design directly paralleled the decision that a real firm or a real government agency might face: Is it worthwhile to retrain its present typists? If Strong's study is correct, it is not efficient for current typists to switch to Dvorak. The study also implied that the eventual typing speed would be greater with QWERTY than with Dvorak, although this conclusion was not emphasized.
I have no reason to believe that the Workman layout offers any other advantages.
What a terrible article. First off, I was frustrated by how they quoted research but did not actually cite it. For example:
>For example, A. Miller and J Thomas, two researchers at the IBM Research Laboratory, writing in the International Journal of Man-Machine Studies, conclude that "no alternative has shown a realistically significant advantage over the QWERTY for general purpose typing."
OK, great. What was the name of the paper? Was it even in a peer reviewed article, or was it an op-ed? And when did they say it? The rate of research into keyboard layouts has been a trickle. If they said that in 1975, then it's not particularly useful today.
In fact, there have been a few studies that have in one way or another demonstrated superior qualities of Dvorak:
(admittedly both of those were published after the Reason article, but I'm not going to spend all night looking for pre-1996 articles)
The article also claims that studies show only "a few percentage points" difference between QWERTY and Dvorak. How can this be, when the total typing distance for QWERTY was nearly twice that of Dvorak on Don Quixote?
So why are we seeing quote and number mining, more typical of political discussions, in an article about keyboard layouts? Well, to understand that, we need to understand Reason's bias, and then read the last paragraph.
And then all becomes clear. This isn't about uncovering the truth about ergonomics and keyboard layouts. It's about scoring a few points for the ol' free market. My guess is that one of the authors was having an argument at a party, and someone drunkenly suggested that QWERTY vs. Dvorak was a great example of the free market failing. "This will not stand!" the writer shouted at the sky, and then proceeded to write a five page article about how "nuh uh".
Of course I'm interested in reading it, or at least the parts relevant to ergonomics. And I did.
As it turns out, my suspicion was correct: the quote mentioned was from 1977. Unfortunately, that paper's brief mention of keyboard layouts is then based entirely on "Human factors in international keyboard arrangement", which was published in 1975 and does not seem to be available online.
The Fable authors mention fatigue early in the paper, but fail to provide any evidence on this point. Their section on ergonomics focuses exclusively on the question of whether Dvorak is faster.
For those unfamiliar with Reason, it's a libertarian publication.
You might then wonder why they are so concerned with typewriter ergonomics.
The reason is that it's a popular example of path dependence, where previous decisions affect the best possible current choice. This undermines their economic theories (the authors are economists). So don't expect them to be an unbiased source on this.
They also, for example, claim that Windows is the best OS, and that network effects have no impact on people choosing to use it
The article was from 1996, I think you'd be hard pressed to suggest an OS that was better suited for the average user than Windows 95 was at the time. I say this as someone who was still running OS/2 at the time (with a Dvorak layout of course), but that was due more to me being ideologically driven (go Team OS/2!) than Reason. Network effects are only one feature that appeals to end users, but they aren't the first.
Here is an "economic" explanation that suggests that both QWERTY and Windows are the worst available alternatives of their respective domain, simply because they are the most popular. Basically, it boils down to "You can survive by being popular, or you can survive by being good."
> There is no serious evidence that Dvorak is a 'better' layout.
The typing speed record, while it still existed, was held by a dvorak typist. I can type at 130-140wpm using qwerty and feel like it's almost impossible to go any faster, while it's not uncommon to see dvorak users typing 160wpm+. If anything, the reduction in total finger travel is enormous.
Without seeing some kind of properly conducted scientific test that examines a range of conditions (age, native language, current typing ability on QWERTY, typical typing material, etc) it doesn't seem worth the effort. If Dvorak (or Workman) is that much better, then it's surprising nobody has been able to conduct such a study.
All of the evidence out there, however limited, is in favor of Dvorak (that 1996 Reason article you linked not withstanding). Lower finger travel = less strain = less RSI.
The advantage of Dvorak is no more "theoretical" than the theory of evolution. Whether it's "worth it" for you to switch is aside from the inherent performance of either layout. I can say that Dvorak is better, while not being worthwhile for you to learn, though it was worthwhile for me.
In my experience typing on dvorak is much more comfortable. Speed wise i maybe reached the same level as my qwerty after almost a month but for me the biggest wow factor was the comfort. Your hands are really never twisting, most of the time you are just using the home row. It's like holding a gamepad where all the keys you need are designed to be in just the right position, not having jump and fire mapped to the start and select-button.
In the end i still switched back to qwerty because I'm using way to many other computers than my own throughout a day and need to be fast on all of them.
Yes, I tried, and I switched back for the same reason as you. The amount of time I spent trying to work out whether I was on QWERTY or Dvorak, and in that case where was the full stop, far outweighed any possible benefit in typing speed.
When people tell me that QWERTY is uncomfortable, slow, etc, I honestly have no idea what they're talking about. For me, the problem is sitting in a chair staring at a screen all day.
Look at the HN thread about this article [1], specifically the top comment. Among other things, that study failed to take into account the "plateau effect" that occurs in all types of exercise regimens after several weeks of the same type of training.
(EDIT: Downvoting doesn't make the claims in the references comment any less wrong. How about arguing the point instead)
I stopped taking that comment remotely seriously when it claimed:
> As a matter of fact, it is a well known and well documented phenomenon in the literature that if you train at close to maximal capacity on the same training regimen for longer than 3 to 4 weeks, you will stop making gains. You'll see big increases the first 3 weeks, and after that, nothing. This is a nervous system effect.
... which most people who have done heavy weight lifting for example can tell you is at best wildly misleading, and at worst, pure, unadulterated bullshit.
To flatten out after 3-4 weeks would take such extreme over-training, at stress levels way beyond what even most high level athletes would normally even able to achieve in training, that it is extremely unrealistic for most people to every experience it. Certainly not "just" by training at something until you feel you can't do any more. At that point most people are not anywhere near a real maximum.
I don't lift to a "training" 1RM (maximum weight I can at any time lift once) every week, much less manage/try to stress myself into a real max. (the reason for distinguishing these, is that it is normal that you can lift up to 10% +/- more in high stress situations like competitions), so maybe what I do don't count in that guys eyes. But I exercise large compound lifts to near failure (that is, until I have good reason to think that I will be unable to complete the next repetition) every workout - 3-4 times a week. So I'll hit my training 5RM, 3RM, 1RM for the various lifts on consecutive weeks. They are reasonably interchangeable. If I feel good enough, I may go for a 1RM to try to set a new personal record even if it's not the "right week".
And see slow but steady improvement for months on end before I shake anything up, and when I do change things, it is mainly for other reasons - some progressions may follow the same program for years. Changing too often does not help - it just confuses you tracking.
Most people are far more likely to run into limitations due to form/technique failures or lack of flexibility that they need to fix than due to any CNS limitations.
Gains do slow down as you get closer to your limits, and you certainly can over-train, but it takes a lot to over-train so badly that you stop seeing gains after 4 weeks even when you massively brutalise major muscle groups on a semi-daily basis. I can't even imagine what you'd need to do to overload your nervous system from typing in that short time-span.
The 'QWERTY was developed to slow down the typist' story is a lie. There is no serious evidence that Dvorak is a 'better' layout. If you were starting from scratch, there would probably be no advantage choosing one over the other, except that QWERTY is the standard. Retraining is a waste of time.
The last sentence isn't supported by the prior, assuming you want to improve your typing.
Let's say that dvorak and qwerty just as "good"/"bad" and that there would be no advantage choosing one over the other. Then retraining is probably not a waste of time because when retraining you don't pick up bad practices, like you did when you first started to type (not knowing better).
It probably is easier to start from scratch than trying to relearn some subtle parts of something you've done every day for for over a decade. Also, the motivation to learn qwerty better isn't as big compared to dvorak where you are forced to make progress to be able to do anything useful.
I found this to be true. I typed Dvorak for a few years while in University, but switched back to QWERTY when I entered the workforce. I was pleasantly surprised that the increased speed and comfort I had attributed to Dvorak was maintained when I went back to QWERTY. It may have been less efficient than putting the same time I to traditional typing training would have been, but it definitely did increase my general skills as a typist to practice a new layout "from scratch".
Laying in bed all day listening to the radio gives less 'finger distance travelled' but doesn't mean greater productivity. I'm being facetious to point out that 'finger distance travelled' is not a meaningful measurement (at least not to me).
Laying in bed all day gives you zero MPG. It doesn't mean that your bed is the most efficient vehicle in the world, it only means you fail at basic logic.
"Finger distance travelled" is a proper metric. If you have a better one, write an article and see how it does.
>QWERTY was supposedly designed for typewriters to solve a very specific problem–to keep the types from jamming against each other. The most frequently used keys were placed apart from each other to prevent them from jamming. This results in a non-ergonomic layout. However, there are alternatives.
That paper presents a good case that the layout was influenced by feedback from telegraph operators, but it doesn't really do much to refute the idea that changes were also being made for the purpose of preventing jams.
Obviously the claim that it was intended to slow typists down is unlikely. If the layout was rearranged in part to prevent jams, then it would have been rearranged so that nearby letters weren't hit in sequence. Without careful study of ergonomics, one wouldn't necessarily realize that that would slow a typist down. But it would probably prevent jams.
I also found some of the claims made in the paper very odd. They claim that because "SE" and "Z" are easy to mistake in Morse code, that those letters were placed near each other. How this would be helpful eludes me.
No, I'm pretty sure he got it right. That article's correct in saying it wasn't intentionally designed to slow typists down. It was however arranged so that letters frequently typed after one another weren't on neighbouring typebars, because that did cause jams in early typewriters.
Also, the correlation between the position of the bars and the position of the keys were not obvious in the first Sholes & Gidden typewriter. Some bars for adjacent keys were actually very far from each other.
By the way, the jamming problem basically went away with the Remington II. QWERTY was obsolete less than 20 years after its inception.
Although it's great that you shared this article so more people know about the history of QWERTY, I don't think it should be the top voted comment; what QWERTY was or wasn't designed for is interesting, but what matters is what it is like to use it in practice now. I am more interested in a discussion about whether or not this Workman layout works as advertised and is worth the effort of switching to.
QWERTY was designed so that all the characters in the word "TYPEWRITER" were located in the top row :) And this is quite easy to verify - coincidence? Does it improve its ergonomy? Well I'm not typing "typewriter" all day long, but first salesmen who marketed the device - did :) They're long gone, however, but suboptimal layout remained. As the man said, there's nothing more permanent than temporary solutions...
There was another that was truly optimized for use Coders, that is not Dvorak based. It had the Option key (⌥) as part of its modifiers. I assume it was called something like the coders layout? I don't recall. If any one can recall it, that would be awesome.
I used it for some time and I quite liked it, more than Colemak. I was reaching 80-90 wpm (I'm about 140-150 wpm on QWERTY). However English is not my native language and since QGMLWY is optimized for English, it wasn't a viable alternative for me in long run. Interestingly, even though I quit typing on QGMLWY long ago, when I occassionally switch to that layout (I still have it installed), I'm able to type on it. Muscular memory dies hard. I do make typos, but I feel that if I returned to practice, I'd pick it up again quickly.
I like the statistics of the different writings he writes at the end, however I would have liked a comparision of the different layouts while writing different programming languages (C, Ruby, JavaScript, etc).
I've seen two problems with alternative keyboard layouts: a) they are not done for programming (keys ~|!<>_-/"' are oddly laid out) or they are not made for international languages (Spanish in my case, for which ñ,á,é,í,ó,ú are difficult to find. Or worse yet (in case of Dvorak) when using an "international" version, the programming-related keys are horribly placed.
I had a programming layer on my ergodox, but due to some bugs in the firmware haven't bothered with it...
Ideally (when I get time to fix it), I'll have a thumb key to hold down and get any programming symbol (optimized to Python for me) on or nearby the home row.
I am really happy and thankful for people working on improving the keyboard usage. I knew about Dvorak and Colemak but it's the first time I am discovering the workman layout. The author went to a length of analysis and effort to come up with this layout - including trying dvorak and colemak - the end of that page contains many comparisons between qwerty, dvorak, colemak and workman.
As well by reading this article I learned about the TypeMatrix keyboards [1] - I must say I am very intrigued by this idea - I would definitely order one if it contained a good trackpoint.
I myself learned touch typing first QWERTZ (German layout), then after moving to France had to learn AZERTY (the worst layout ever, especially for programmers). In the end because I had to travel and work on keyboards in different countries a lot - I decided to learn QWERTY US International, I configured it with AltGr Dead-Keys and now I am able to use the same layout where ever I go (even Russia and CIS countries) and on my computer I have all the accents for French, German and Turkish. So yes, I know it is not optimal for the Finger usage, but I need a keyboard layout that (a) is available on ALL stock computers, (b) on my computer can be enhanced to allow writing ALL the accents of European languages by using key-combinations.
I think alternative keyboard layout researches should take into account international usage.
You might also want to take a look at the ergodox: http://ergodox.org/
It's not so easy to buy as its just a DIY project. Massdrop organizes some group buys [1], next one should be due anytime this year.
I'd really like a keyboard layout optimized for phone usage, particularly with swype or predictive keyboard styles in mind. In particular I think you actually want to go the opposite direction of things like dvorak with such a keyboard because you want to make common motions distinct from each other. With swype, dvorak would leave most common words having a very similar motion profile. Qwerty is perhaps better, but I suspect non-optimal from how often I have to guide it.
While not exactly what you're looking for, on Android I've had a good experience with the MessagEase keyboard. The key idea is to have fewer, bigger buttons, and use swiping motions within those to distinguish other characters. Even after getting used to predictive typing, I quickly came to prefer the MessagEase keyboard, in part because I tend to type weird characters, which are much more bother with the regular keyboard. With all the times I would have to correct swype, I might be as fast or faster with MessagEase. Oh, it also has cursor movement built in.
Fellow MessagEase user here. It's absolutely fantastic. It's great at non-dictionary words and special characters (there's no need to press a magic option key to switch to the symbol set), as mentioned above. Two other things I like: a) It has built in arrow keys, perfect for going back to insert a skipped letter or similar. b) Typing errors occur at the character level, not the word level. And are hence easier to fix. (Or, if left unfixed, are usually still legible.) No damnyouautocorrect moments.
Edit: I forgot to mention one of the most important parts: it's incredibly fast once you know what you're doing.
I think ti could still be a good keyboard for phones but they seem to have died off and the layout is patented or something and so there don't seem to be any versions of it for android.
Can't access the article due to database error, but since we're discussing keyboard layout effectiveness for developers:
If we were indeed limited by input, why wouldn't this come up in discussions about languages like CoffeeScript (or even Python/APL) that save keystrokes to begin with? In the case of Coffee and Python, the keys saved are actually the hard-to-press symbols as well. And yet I've never heard the point come up.
Perhaps this is more about our need to feel superior by adopting a routine of cargo-cult efficiency than anything else. In the broader view of productivity, optimising keyboard layouts is more like optimising memcpy() in C. While it may bear some results, it's rarely the actual bottleneck.
> since we're discussing keyboard layout effectiveness for developers.
We're not, at least not in the way you think.
> In the broader view of productivity, optimising keyboard layouts is more like optimising memcpy() in C. While it may bear some results, it's rarely the actual bottleneck.
This article is a solution developed by a someone in response to his own RSI problem, optimised for writing English with minimal strain, and presumably tested based on "how much do my arms hurt." So it's a solution/optimisation to a different problem altogether (which may or may not be relevant to you.)
It's an impressively in depth and well reasoned analysis, and takes into account common bigrams which is something I'd never considered being important for keyboard layouts.
I tried switching to Dvorak many moons ago, I'm feeling it may be time to have another crack at an alternative layout, both out of interest in improving efficiency and reducing strain.
I realized [my fingers] were moving too much laterally [...] Just ask yourself, how often do you type ‘the’, ‘these’, ‘them’, ‘when’, and ‘where’, etc. on a day-to-day basis?
All of these words are typed in Dvorak without any lateral finger movement, and only two letters outside the home eight.
'ls' is, however, awkward to type. That's why I alias it with 'd'; problem solved.
It is much more efficient to ride the momentum of a single arm or wrist stroke and type a combo rather than just one key.
This is not my experience:
An example of this is the word OPERATION. If you were to type this in Dvorak, you could type it as o-pe-r-a-t-io-n where each grouping is a hand stroke–a total of 7 hand strokes.
The letters 'pe' are the slowest typed for me when typing 'operation'. I strongly prefer alternation, as the finger on the opposite hand can be lined up on the upcoming key just as the current key is being typed. Having to move the hand around for keys that are on one side slows things down.
I don't see what the point of changing your keyboard layout is.
I don't know if QWERTY is even a good layout but what does it matter? Isn't the best keyboard layout the one you already know?
Thinking about it now, I can't think of any common digraphs or trigraphs that are inconvenient to type on a qwerty keyboard. Don't fix what ain't broke.
QWERTY is deeply, gut-wrenchingly broken. (Kind of like British electric outlets, or Christian death metal.)
However, just switching keyboard layouts doesn't fix things. I once switched to Colemak for a couple of months. My hands felt better, and typing on my own machine was way more comfortable. But every other keyboard I interacted with made me look like a dude twice my age with some kind of senility disorder.
To really get the benefits of a better layout, you have to not switch, but rather add a new keyboard fluency, while maintaining QWERTY proficiency. That made the cost-benefit equation very different for me... and I went crawling back to my old abusive partner, QWERTY.
Hey now, British sockets are the best: integrated switch, grounded, fused plugs, bomb-proof, cable is flush to wall and its much harder to accidentally pull out the plug.
I've read a lot of reports of people who maintained dual fluency. They say you just have to switch back to it for about half an hour a day and that will keep it fresh. I think that you can do much less if you're concerned with basic competence rather than actual proficiency.
After years of using Colemak (and not practicing QWERTY regularly), I've found that QWERTY is actually getting easier again. I'm not fast with it, but I no longer have to think hard to recall positions or look down at the keyboard. It feels like my brain said "Oh this is how keyboards work now" and rearranged the QWERTY structure in my brain to do Colemak instead. Now after years of having to occasionally use QWERTY, my brain seems to have built a semi-competent second typist, so to speak.
I switched to Dvorak seven or eight years ago to see if it would help with RSI issues I was having. It helped a lot, and I never switched back. Sometimes I have to type QWERTY (at kiosks at the library, for example) and I just laugh at how ridiculous my hand motions become when using it. For people experiencing wrist pain while typing, I recommend trying it. A lot of people use it (compared to other minority layouts), so support is widespread.
I also have RSI from years of piano, typing, and then (this is almost embarrassing to say) high level starcraft.
Changing your keyboard layout isn't going to fix bad habits and the things you have to do to stay healthy (stretching, taking breaks, and strengthening).
Good point. About 15 years ago I had the beginnings of RSI, and discovered the immense benefit of just getting up and walking around every now and then (good for the eyes, too). Also there are a few wrist/forearm massages that helped, basically I just squeezed wherever it felt necessary.
I also changed my typing style to use my arms to slide my wrists around the keyboard a little rather than overextending fingers, and I rotated my hands inward about 30 degrees (maybe less, never measured it). I still keep about 60-ish WPM with this method, even at 42 years old.
Anyway, I like the Workman layout a lot, especially that the common move for index fingers is down, and the common moves for middle and ring are up. Great thinking!
I also changed my typing style to use my arms to slide my wrists around the keyboard a little rather than overextending fingers, and I rotated my hands inward about 30 degrees
Yes, this. I've been typing this way my entire computing life (22 years) and have yet to develop symptoms of RSI from typing. (Note: I have developed RSI symptoms from mousing -- and relieved them by using a trackball.)
The reason I switched to Dvorak was that I wanted to learn to touch-type properly. When changing an ingrained habit, there's always that transition period where the bad old habit is the better short-term choice than the good new habit, and if I'd tried to learn touch-typing QWERTY I'd have fallen back on my bad habits probably without realising it. By introducing a new keyboard layout into the mix, falling back on my bad old habits had an immediate negative consequence (wrong letters appear on screen) so I knew I'd messed up and could immediately rectify the mistake.
Of course, now I know how to touch-type and I'm quite comfortable with Dvorak, so I'm not sure I could muster the mental energy to switch to this Workman layout, pleasant-looking as it is. The biggest drawback of Dvorak is indeed the L and R keys... but I guess my finger muscles have strengthened or I'm just more used to it now, so it no longer bothers me.
There are many measurable metrics to determine the "best keyboard layout", such as whether common letters in your language are on the home row, or the total finger distance travelled when typing typical text, etc.
The three most common english digraphs are TH, HE, AN, and they all require moving off the home row to type in qwerty. I consider this an inconvenience, one that doesn't exist in other layouts.
Of course there is a cost to learning a new layout, but if you're probably going to be typing for much of the rest of your life, the cost is very small.
Keyboard layouts are a fun topic to be sure, and retraining your brain to use a different one is an awesome adventure. Where it always falls down for me is the whole experience of switching machines. It is just a massive fail when I need to remap my brain back and forth like that.
I enjoyed learing Colemak and I would probably enjoy Workman, but it isn't the most practical thing for me to do long term.
Also, a lot of the keybinds in vim or emacs are designed around certain ergonomics and being efficient in an editor is often a more useful thing for me than being efficient when I am typing.
The problem with this keyboard and others (like Dvorak, Colemak) is nobody else uses them.
I have one of those old indestructible IBM model M keyboards with removable keycaps. I love that keyboard. I decided I'd try to learn DVORAK and work with it regularly, so I did so. And I realized something: every time I'd work with a keyboard, I _will_ be working with QWERTY. There's no way I would request to an employer that I have a special keyboard just for me, or try to reconfigure software to support it.
So I went back to QWERTY. May not be ideal but I know everyone supports it.
> There's no way I would request to an employer that I have a special keyboard just for me
Why not?
> or try to reconfigure software to support it.
Why not?
If there's a much better alternative out there that can make you more productive - and Dvorak has been one of those things for me - your employer should be jumping up and down in joy at throwing however many $10 keyboards at you as you're asking for.
There are also plenty of USB devices that convert between the two, or keyboards with hardware switches. Or, of course, in software. I use AutoHotkey, because keyboard layout switching in Windows is completely broken, and it works pretty well.
If that's your excuse for not using Dvorak rather than anything else, you're doing both yourself and your employer a huge disservice, especially if you're going to take that attitude to other things.
I've worked several programming jobs since switching to Dvorak, and it has never been an issue---it takes < 10 seconds to change the layout OS-wide. No need to request a special keyboard (and I leave my keycaps as QWERTY because I touch-type anyways).
I learned dvorak by printing out the layout and glancing over at the paper while typing back in high school. It took me about a month to get comfortable and I'm now basically a touch-typing bilinguist (though I'm slower at qwerty, it's more mental effort to do the conversion on the fly since my muscle memory is dvorak).
I'm relatively sure that every OS supports changing the keyboard layout without being root and for just one user, so unless your employer specifically disabled that, then that shouldn't be an issue.
The point about the keycaps is valid, but ideally you shouldn't be looking at them anyways. Additionally, I learned Dvorak an a QWERTY labeled keyboard, and found that eventually I just associated the labels I saw with what I was actually typing.
I don't get the point why someone would learn a new keyboard layout.
- It is very hard to learn since muscle memory is very difficult to change.
- I think it would take me a while to be as fast as on my current layout QWERT[ZY].
- I see no real advantage for learning a new keyboard layout. None of my friends and coworkers ever complained about keyboard layout and the urge to change it.
- When you want to get things done, learning a new layout would be the last thing you want.
- You may not have your new keyboard layout available at different workplaces.
So I think this is just a hipster thing. Prove me wrong :)
I don't get the point why someone would learn to ride a motorcycle.
- It is very hard to learn since your muscles are trained for driving a car
- I think it would take me a while to be as confident as in my current Monte Carlo LS
- I see no real advantage for learning a new automobile. None of my friends and coworkers ever complained about driving a car and the urge to change it.
- When you want to get from point A to point B, learning a new vehicle would be the last thing you want.
- You may not have a motorcycle available at different car rental places when you travel
Didn't really prove you wrong, and maybe I am a hipster :-)
I found it enjoyable learning a new layout. Maybe it was just for the sake of trying something new. My productivity didn't suffer since I never really abandoned QWERTY. I didn't utilize colemak in my work until I was relatively fluent. Learning workman might be an interesting weekend excursion, but I have no plans to use it for work until I'm fluent, if it is as fun to type on as colemak.
Assuming that everything you don't personally understand is "just a hipster thing" is such a shitty attitude!
You offer five cons, which _to_you_ outweigh the pros – which are so obvious it seems redundant to even mention them, but let's just say, to sum it up, ergonomics.
It's very well known that full-time IT work takes a heavy toll on your body. That's why reasonable people make sure to work with decent posture, to take frequent breaks, and to exercise. It's why decent companies make sure to provide high-quality tools to help their employees stay healthy and avoid injury.
And it's why some people are interested in improving the ergonomics of typing, which – as everyone knows – is fraught with the dangers of repetitive stress.
That doesn't mean that keyboard layout choice is the most important thing in IT ergonomics. But for some people, switching to Dvorak or Colemak or whatever else seems like a reasonable choice, and many of them, myself included, are very satisfied with it, even while working in a collaborative environment where others use QWERTY.
For you to call this "just a hipster thing" is embarrassingly dumb! Do you say the same about ergonomic keyboards, rollermice, and saddle chairs?
I'm not convinced either, but your arguments basically are: I don't get it, it's hard, and I don't know anybody doing it. If you read it like this, you should see such arguments don't make much sense :]
Okay, as I wrote in another comment, I once learned Colemak (I didn't like it very much) and subsequently, CarpalX (QGMLWY) which I liked and I soon reached a reasonable speed of about 80-90 wpm on it (which is about 2/3 of my QWERTY speed).
What were my reasons of trying new keyboard layout?
- "The hipster thing" (so that one doesn't count).
- Curiosity. I wanted to find out what the fuss is about - whether the talk behind it isn't just that, a hipster thing itself
- I like to learn and I love the feel of improving, of being better than the day before
- It's very comfortable once you get used to it.
Adressing your objections:
"- I think it would take me a while to be as fast as on my current layout QWERT[ZY]."
1. The time will pass anyway.
2. I never typed faster than on QWERTY, but I'm not really a typist. I'm not a court reporter. My productivity is not directly proportional to my typing speed. It's the thinking that takes most of the time anyway. So I can sacrifice some of the speed for the sake of comfort.
"None of my friends and coworkers ever complained about keyboard layout and the urge to change
This I believe, but as Henry Ford remarked: "if I had asked people what they wanted, they would have said faster horses." :)
"You may not have your new keyboard layout available at different workplaces."
That's obvious, but I, for one, don't change workplaces that often. And I only use two computers: one at home, one at work. Installing a preconfigured keyboard layout on a new machine takes a minute.
I mainly returned to QWERTY (I'm repeating myself again, but so be it) because English isn't my native language. All these alternative keyboard layouts are optimized for English. I use a QWERTZ mutation (PN-87 - rare and somewhat forgotten even in Poland) for typing in Polish. So sticking to QGLWMY could only ever make like 1/3 of my typing life better :) If I used no other language than English, I probably wouldn't have looked back.
The most serious problem with learning a different keyboard layout for hipstering purposes is not only that it requires a certain degree of effort, but that nobody can tell you've done it unless you tell them. Which you absolutely never must. Because you'll never hear the end of it. As this thread is demonstrating amply.
The advantages are speed and comfort. Yes, it's hard to learn and you'll be slower than with QWERTY for a while, but it pays off in the end. Consider it a long term investment in productivity.
I've been typing on Workman layout for almost 2 months. I've not been very disciplined about doing practice every day.
I'm only using it on my Microsoft Natural 4000 ergonomic keyboard, and still using a regular QWERTY bog-standard-dell keyboard at work, which means I can still type pretty effectively on colleagues' computers.
I quite like it. It does seem to make a lot of sense.
Switching back and forth between a normal QWERTY board and Workman-on-a-special-board takes at most one sentence of brain-mush, but then it's fine.
The thing which slows me down the most I think is programming in VIM. I'm not going to remap anything, as all the commands I think of as sentences (c)hange-(i)nside-(t)ag, (d)elete-(t)il, etc. And there's a lot of muscle memory to overcome. Also with sh commands.
Still, I do like it.
One of the reasons I decided to go for Workman rather than one of the alternatives is that the whole column-layout for physical keyboards seems like a better design in general, so one day, if I can afford it, I will try and get one.
I have been really struggling with RSI for a years now, so along with changing layout, I'm also trying to take more breaks, sit with a better posture, etc. It does seem to be helping.
After 2 months of very irregular typing on it, this also being my first time doing 'formal' touch typing, I'm around 50wpm typing normal prose, which still feels a bit slow to me, but I am improving.
It sounds like all of these layouts are just people taking guesses and reporting on their anecdotal experience.
Has anyone ever tried determining a fitness function (travel distance, priority finger use, sequential characters on nearby fingers) and running random layouts through a genetic algorithm?
For sample data use english, romance languages, open source code, etc, for a good general purpose layout.
If you're going to relearn a layout it might as well be the optimum one.
He definitely did quite a good job with the fitness function (bigrams, finger priority, etc), I just didn't see quite how he came up with the workman layout specifically. If he manually evaluated a few layouts and made tweaks, then surely there is further optimisation to be done for any given body of text. For something as central as a keyboard layout it seems like even those small gains could have huge effects.
It's not just guessing. The Carpalx does have algorithms for scoring various characteristics like finger bigrams, finger travel, and outputs the optimal results based on scores for key presses and bigrams, etc.
From the article: 'However, I believe that the way that alternative layouts focus on just the home row for optimization is somewhat misguided. We should optimize the keys inside the hand’s natural range of motion and not just strictly the home row.'
While alternative layouts strive to make typing easier, they invariably come up against the constraints of a grid-based key placement.
Using a genuine 'anatomic'/'ergonomic' placement would put reset the whole layout argument: how to optimize for a given language, with human-friendly placement of keys (say, using the Kinesis Advantage keyboard or equivalent). Now that would be an interesting experiment in layouts! (Would there even be a home row?)
And, BTW, I've 'forked' my usual Dvorak layout to make it easier for Lisp programming. We (particularly Linux users) tweak our window managers and screen accessories to individual perfection; why not our keyboard layouts too?
At my desk, I use a Dvorak keyboard. However, I often need to rip my laptop away from my desk for meetings, emergencies, etc. When I'm typing on the laptop, I use Qwerty. It's not a problem at all. I don't need to think about it; my hands just do the right thing depending upon where I'm typing.
I've been pretty consistent about only using dvorak on split keyboards, and only typing QWERTY on non-split keyboards. This has done a pretty good job of keeping my muscle memory for both separate. My QWERTY typing is a bit rusty, but I can still manage up to 60wpm when using my laptop, and I don't have to pay too much attention to what my fingers are doing. Though when I use QWERTY on a split keyboard I have to watch my fingers.
(These environmental cues can be funny sometimes - if I use emacs for editing C++, which I do only very rarely, I keep missing out semicolons. Because most of the time, if I'm programming, and my eyes can see emacs, my fingers know I'm using python.)
It's good to hear that. Being afraid of forgetting how to touch type on a qwerty was one of the reasons I never made the plunge into alternative keyboard layouts.
Though, now that I think about it, alternating between keyboard layouts is really no different than using a modal editor like vim -- after a while it just becomes second-nature.
I originally thought this two. Trick: when typing QUERTY, look at your keyboard. Anything else, try to just look at the screen. Helps your brain differentiate layouts.
At least on your computers (where it matters) you can change the mapping. If you can touch type you can simply ignore the labels on the keys.
I'm building a custom mechanical keyboard (with blank keycaps) and I have changed the various function layers several times and had no problems with adjusting to that.
It all comes down to the question what you do mostly and how easy you can get used to new layouts both physical and in software.
It's reasonable to be a bilingual typist. I use dvorak on my personal and work computers, but qwerty for other people's computers and iphone/ipad. It's really not too bad.
I like the research that went into all this before deciding on the layout. I don't think it would fit many people's use pattern, instead anyone seriously typing for a living could take the same approach and come out with a personalized layout. Now, if only it could be easier to remap a set of devices keyboard with an arbitrary layout.
That's a kind of pet peeve, but if you are a programmer, depending of the language you use the most at a given time, the most used keywords and special characters won't be the same (i.e optimizing for lisp or perl would make wildly different layouts). For people also speaking non english languages, the most used letters won't be the same depending of the language as well. To try to come out with a new layout supposed to be efficient for everyone is a fool's errand.
The Workman layout sounds interesting, though I don't have the brain-cycles to try switching. The Shift + Capslock = Escape seemed interesting though (especially for a vim user).
If you want to use it with KeyRemap4MacBook this should save you a bit of time:
The QWERTY layout has been around for so long that it is difficult to just change to something else. People are used to the well-known keyboard interface to the computer/smartphone/tablet/... and actually expect it that way, no matter how inefficient it may be.
Of course, you may switch personally to an alternative layout, but the overall design used by everyone else will not change. It's just a burden for yourself to be able to type two different layouts with comparable speed as you cannot change the layout everywhere you use computers.
I'm a happy user of the Neo layout. It's very different from QWERTY (or rather German QWERTZ) but optimised for English and German bi- and trigrams as well as programming.
I used to use dvorak, but ironically, I switched to querty because I was having RSI issues and I reverted to two finger hunt and peck for a while (not a fun time). Now that things are better I've considered going back, but the sticking point for me is ctrl-c/ctrl-v. These were always awkward for me under Dvorak, and at the time almost impossible to remap.
This has perhaps inspired me to give colemak (which retains the usual c/v positions) a shot.
Can't speak for other OSes but Mac OS X has a Dvorak layout that switches to Qwerty for hotkeys. I leaned on this pretty heavily when I first switched to Dvorak.
I feel like improvements over Colemak are going to be pretty incremental at this point. In this case, there's a pretty big trade-off on leaving home row for the sake of not reaching for the H.
What I'd like to see is some proper innovation like good programmer-usable chorded layouts that work with existing inexpensive hardware (I don't mind buying a good non-ghosting keyboard, but I don't want to pony up a few grand for a Velotype).
I love the idea of alternative keyboards, and I spent a few months trying to learn Dvorak, but no matter what I stop because it screws all my keyboard shortcuts.
OSX's "Cmd Qwerty" feature was nice, but didn't help in Inkscape (since it uses Ctrl) or Vim.
I wish the Programmer's Layout of this would take HJKL into account at least. That's and :wq being a nice "roll" down the top left of my keyboard are what keep me on QWERTY.
In terms of vim, I actually like it a lot more. j and k are typed with the left hand (coincidentally j is still on the left side - same as qwerty). The right hand does h and l (again, h is on the left side and maps to going left).
On the other hand, Eclipse is super annoying because it seems half the bindings are cmd qwerty and half are dvorak but that's another issue. I've started using Vrapper to get vim bindings in eclipse and that seems to be working out okay so far.
Think of it this way: typing is a distraction from thinking, and minimizing it is a good thing. When I type much slower than I think, make mistakes or otherwise have to think about the keyboard, it pulls me out of flow.
I switched to Dvorak a few months ago. It's much better than I expected. The switch was painful and it took like 3 months to get fast enough.
Compared to Qwerty writing Dvorak is so much easier on the hands. I would say the strain is reduced by 50% at least and it's potentially faster to type (less movement, less way).
I will never switch back to Qwerty unless someone forces me to do so.
One thing that struck me is that this layout is immediately intuitive. Unlike Dvorak and of course QWERTY or any other popular layout, I can already picture myself typing in it (of course I learned on QWERTY, so I wasn't in-tune with key familiarity before that).
I can't explain why exactly, but this layout just makes sense. Can't wait to try this out.
I'm surprised no one has mentioned the norman layout here. I consider it to be an improvement over colmak and workman.
Using it on my ergodox (and I know another relatively prominent geek doing the same), and really enjoying it.
Or learn QWERTY with your hands placed diagonally.
I never learned how to type "properly", i.e. "home-row" typing: my hands rest naturally somewhere between the middle and upper row at maybe 30° angles from the vertical, and some keys I strike with either hand depending on the word (e.g. t/y g/h b/n).
Yet I type faster than most people I know, and have had no RSI issues due to typing almost daily for 22 years. (I have had RSI issues due to using (gripping) a mouse; replacing the mouse with a trackball / no mouse eliminated this.)
This "nontraditional" method of typing leaves my wrists straighter and reduces repetitive motion, to which I credit my fortunate freedom from RSI.
The downside is that I am unable to use so-called "ergonomic" split keyboards, as my fingers are trained to work both sides of the keyboard.
(And I say "ergonomic" in quotes, because while most "ergonomic" keyboards, such as the Kinesis, are designed to reduce movement, my personal experience is that ergonomics is all about making morevaried movement.)
Interesting. I have a split ergodox and switch to a laptop keyboard a bit. My fingers don't seem to mind the angle too much. Though it is crazy how keyboard layout has been status quo for nearly 100 years.
I'd love to learn something like this but my problem is that I'm scared I'll disable myself from being able to use a regular QWERTY keyboard without looking like an idiot.
Changing keyboard layout of library or university computers or basically any other computer than your own comes with a lot of trouble.
I've been using Colemak for a couple years now - but instead of learning it with my laptop's keyboard, I learnt it with a new Kinesis Advantage Pro. I now find that if I'm using a standard keyboard, I can type qwerty without much more difficulty before, but if I'm using the Kinesis, I have no trouble touch typing Colemak. My guess is that the massive difference in the actual keyboards stops my brain from getting confused.
Maybe what really needs to happen is for someone to make a "software keys" keyboard, where the characters are not imprinted on the keyboard, but are displayed on the keys instead. That way when you change stuff around, as is variously suggested here, WYS remains WYG, as it were.
Rad or dud?
I've only ever been able to hunt and peck QWERTY, not having learned it through touch typing, nor did I ever pick it up.
So when I learned Dvorak, through touch typing, it affected my QWERTY skills exactly not one bit. The only way in which my QWERTY skills are now bad is through attrition, but it's still no problem to accomplish a reasonable percentage of my previous speed. It just comes up pretty rarely that I have to type on QWERTY these days, and if I do for a few minutes it's no problem.
It's as if the two are in completely different areas of my brain, and for anyone wanting to learn Dvorak, then this is a very easy way indeed.
Just started using this and can say that it is awesome. I used to use Turkish Q keyboard previously, because it is the layout of my laptop; albeit I write in English primarily, and have already reached ~5 wpm, in fifteen minutes. I strongly suggest giving it a try.
Switched to Dvorak after plateau-ing in typing speed and experiencing pain in my hands and wrist. Helped immediately, but what also hleped almost as much was:
- Aliases for frequent shell commands
- Remapping 'caps lock' to 'control' (on US keyboards)
What about tracking everything you type for a month to see which letters you use the most as well as their frequency and then generating a keyboard layout that will actually work for you?
These keyboard layout wars seem just too subjective to me.
as a dvorak user i find his measurements very interesting.
has anyone noticed that while the distance traveled in dvorak is higher than both other alternatives, it alternates hands much more frequently (23% vs. 31%+ same hand frequency), and moves towards the top row more frequently than to the bottom one. I don't know about you but for me it's easier to extend my fingers top than it is to flex them towards the bottom row.
there is also less time spent on the number row, but repetitions with the same finger are more likely.
i still think there ought to be a better way to deal with special characters than shift keys
sadly all alternative keyboard solutions are spoiled by riding the health bandwagon and are utter crap.
and the ones that aren't turn out to be so expensive they die (datahand) or get bought out and turned in non keyboard replacement products (fingerworks)
Although I completely agree that QWERTY is poor and Workman looks like a very good system, I'm in my late 50s and think it would be a painfully slow process to get used to a different key layout. And I don't seem to get RSI.
But what does really steam me about keyboards in general, is the numeric keypad. I'm right handed, so the mouse sits to the right of the keyboard. And of course the mouse gets used a lot. Which means my right hand is always swinging back and forth from RH keyboard home and the mouse.
Now it happens that I _never_ use the numeric keypad. Never have to enter large blocks of numeric data, and so am quite happy using the numeric top row. Used to it that way, and don't see anything wrong with it. So to reach the mouse I'm moving much further than I should have to, across the wasted space of the numeric keypad. Also if desk space is a bit tight I keep hitting the mouse on the RH side of the keyboard - again because of that extra length to the right.
It's a workspace centering thing too - the main area of the keyboard should sit on the centerline of the screen, to avoid small but persistent twisting of the spine and neck. But then the keyboard R end sticks way out, and the mouse movement field gets pushed further to the right.
It's been bugging me increasingly for years. I've found a few 'small' format keyboards without the numeric pad, but the designers always go overboard and think that 'small' means everything on the keyboard ought to be small - smaller keys, thin base therefore short key travel and terrible touch feedback, flat keytops (another pet hate - you get no tactile feeling of centering, and so have to keep watching your fingers type), compressed layouts of the existing keys (arrrgh!), and so on.
What I really, really want, is something like an old klunky and reliable IBM (like I'm typing on now) but with no numeric pad. No other change.
I'd long ago have simply taken a hacksaw to a standard keyboard and cut the numeric pad off, except guess where the controller IC _always_ is in keyboards? Right above the numeric keypad, of course.
It's so annoying! Why does no one, that I've ever been able to find, make a full size, heavy, long-key-travel, concave key-tops, easy-typing, standard layout (or Workman layout!) keyboard without that stupid numeric keypad?
In these days of USB it's especially dumb, since if someone really wanted a keypad as well they could just buy a separate keypad and plug it into a USB port. There's NO reason why it must be included in every damned keyboard in existence.
If anyone knows of such a thing for sale, I'd very much like to hear it. I'd buy several.
I did not regret my decision (far from it), as Dvorak is indeed nice to type and quite readily available out of the box.
Normally you don't get much benefit from switching layouts though. Especially as developer. As an author or someone who is writing prolonged texts in plain English, yes it's definitely an advantage. Yes, it's also more fun to type. But really, QWERTY is good enough in most cases. You don't really need to bother.