Hari's Corner
Humour, comics, tech, law, software, reviews, essays, articles and HOWTOs intermingled with random philosophy now and thenAn alternative view on the whole keyboard layout business
Filed under:
Software and Technology by
Hari
Posted on Fri, Jul 26, 2013 at 21:46 IST (last updated: Wed, Jul 31, 2013 @ 14:23 IST)
As I have explained in my previous blog entry, I was recently playing around with alternative keyboard layouts for their perceived and scientific benefits over the good old fashioned qwerty layout. In fact, I have read quite a bit on the topic, including some scientific analysis on keyboard usage and finger reach/strain while touch typing on various layouts. In all the tests, the qwerty comes out as the worst layout and is universally decried as the most inefficient of them all. And I have no doubt that it is true; at one level.
But here's the interesting part. Most of the studies assume that the typist is using the standard and the formal method of typing that is usually taught by typwriting institutes, touch-typing. If you are not aware of what touch typing is, Wikipedia can help you out. And undoubtedly many tests show that qwerty puts maximum strain on the hands and fingers of touch typists who utilize all their fingers. I am not getting into the scientific part of the discussion. There are several websites that have studied the problem of keyboard usage and have come up with interesting solutions. Dvorak was probably the first popular alternative to the qwerty back in the days when computers weren't around. But since then, there have been many new layouts invented to eliminate stress related typing injuries by analyzing finger usage and finger travel.
And all those studies seem to base their results on touch typing techniques.
And that, I think is a mistake. I have no statistics to back me up, but I believe that most people who have started their typing on a computer (and not learned typing as a profession) probably haven't learnt to touch type at all and instead hunted-and-pecked and developed their style from there. And for those who haven't learnt to touch type at a young age, it can be quite awkward to learn, especially because of the usage of the little finger which is by far the weakest finger. Even on the best functional layout, if you touch-type, you do utilize your little finger and ring finger and you do have to curl and move your fingers both horizontally and vertically.
I am a good example of somebody who has adapted a three-finger typing technique that can achieve relatively fast speeds on qwerty without losing much comfort over the years and not glancing down at the keyboard at all. Undoubtedly the awkward key placements on qwerty forces people to make such unscientific and ad-hoc adjustments. Using three fingers (my index, middle and occasionaly the ring finger) to travel all over the keyboard and then hit the correct keys by making the right hand movements, I entirely avoid the little finger. A different kind of muscle memory to locate keys without looking down at the keyboard than that used by touch typists has developed for me, using which I can actually type at an average speed of around 80 words per minute (occasionally peaking at about 90 wpm) without much stress and certainly by making the adjustments to reach the requisite keys by moving my hand rather than reach out with my fingers. This means that although hand travel is an issue, fingers don't reach out awkwardly for keys and less strain is put on the ring finger and none on the little finger. I am not aware of how many people use this kind of method, but I am guessing that a lot of people have developed their own style of typing on a qwerty keyboard that reduces finger travel and stress, compensating instead with a slight hand movement. I am also not sure of what medical drawbacks or benefits this method may have. What I do know is that the fluid hand movement is far less stressful for me personally than learning touch typing and actually utilizing the little finger at all. For those who have tried to touch-type after years of free flowing hand movement, the style feels cramped and awkward. I am sure that it can be got used to, over time, but it is a skill that is entirely new and learning to touch type with a whole new layout can mean added stress on the brain-muscle coordination. Overall the two main problems are the usage of the little finger and the resting of the hand on the home row and reaching out to keys with all the fingers.
My point is this: I think many of the stated benefits of the alternate keyboard layouts make some big assumption about typing habits and the way the stresses/pressures are distributed across fingers and hands. I am not even sure how many of them are completely scientific and how many of them are based on anecdotal evidence. But I have yet to see an analysis of alternate keyboard layouts that take into account the various ways in which people type other than touch-typing. There also appears to be a universal assumption that touch-typing is the best way to type. I also would love to be pointed to studies that prove this scientifically. I certainly find using my strongest fingers to type a lot less stressful than touch-typing. It is not entirely error free, but occasional use of the backspace key won't bother most people using computers rather than typewriters.
The poor qwerty layout receives little love from most people who have analyzed functional keyboard layouts and typing enthusiasts who have embraced alternate layouts; but it may be pertinent to note that there are millions of people across the world for whom qwerty is branded into their muscle memory after years of typing and who have learnt to work around its inefficiencies and disadvantages in spite of so many strong contenders to replace it with something better and more efficient over the decades. I think unlearning something as fundamental as typing technique is a bigger deal than people make it out to be.
My belief is that typing stress is relieved more by using ergonomic keyboard designs combined with mechanical keyswitches for tactile feedback and less force per keypress rather than the functional layout itself. Of course a lot of touch typists will disagree and I actually think both viewpoints are right at different levels.
So if you are going to learn an alternate functional layout, I think it is important to consider whether you are a touch-typist or not. I think many of the stated issues of qwerty are not fully relevant to people who do not touch type and that learning to touch type is a separate and non-trivial skill in itself, which combined with learning an alternative layout (a layout that takes full advantage of touch typing techniques). can be a stressful, daunting task with a high learning curve. For non-touch typists alternate keyboard layouts may not be beneficial enough to make the switch.
Comments closed
The blog owner has closed further commenting on this entry.
19 comment(s)
Comment by Sudipta Chatterjee (visitor) on Fri, Jul 26, 2013 @ 21:53 IST #
Comment by Hari (blog owner) on Fri, Jul 26, 2013 @ 21:59 IST #
I can notch out 122 wpm peak (100% accuracy for several minute passage) and over 100 wpm on average on qwerty on my mechanical Das Keyboard unless the text being typed is filled with crazy symbols and other non-standard/frequently occurring keys, which even on other keyboard layouts like Colemac is going to cut down your speed significantly.
I utilize all 10 fingers and obviously touch type. The biggest strain for me is not the movement of the fingers but the pressure needed for each keystroke. As I mentioned earlier, a good mechanical keyboard easily fixes this condition.
Since I use all 10 fingers, I have a decent coverage of the most frequently used keys and don't need to travel my fingers over a longer distance except for special symbols and the occasional backspace key, (although for backspace I tend to use my ring finger and not the little finger).
While layouts like Colemacs are no doubt more efficient in terms of key placement for English, I doubt I can physically go much beyond 122wpm and sustain such speeds even if I shift to Colemac.
Therefore for me, Qwerty is okay. It's not great but it's something I can live with given a good keyboard.
Comment by Ramshankar (visitor) on Tue, Jul 30, 2013 @ 20:22 IST #
It appears that speed is not the important consideration for some alternate layouts like colemak - comfort is. And that is where most of the studies I've seen assume touch-typing.
Comment by Hari (blog owner) on Tue, Jul 30, 2013 @ 20:49 IST #
That, I can believe.
That, I cannot. As far as I can tell, the majority of people still use primitive hunt-and-peck. You're assuming that people actually improve but:
1. As you found out yourself, changing a habit is not so easy.
2. Most typists trying to improve themselves would probably have been seduced by touchtyping anyway!
And the second one is huge. There's a lot of resources, both online and off, in formal and informal training, about how to type "well", and all of it is touchtyping. In order to get to where you hypothesize, a typist would have had, not only to improve himself, but to avoid the one of the most canonical ways of doing so that history has ever seen. The people left to hear your argument, far from "most...non-touchtypists", are the 1%.
From this standpoint, statements like "For non-touch typists alternate keyboard layouts may not be beneficial enough to make the switch" are utterly misleading. Only a bare minority of non-touchtypists would have persisted long enough to get that 1% status and overcome the disadvantages of QWERTY. The rest of them can still improve from other layouts just like everyone else (as long as they change their ways and start touchtyping, of course
Anyways, I guess this is an old post (I hope my suggestions on the colemak forum have since changed your mind somewhat), and good luck in your search.
Comment by lalop (visitor) on Wed, Jul 31, 2013 @ 02:44 IST #
For that reason alone, I state that the people who have learnt to touch-type "formally" must be few in number and only keyboard enthusiasts and professional typists take the effort to learn such things. Most of the people who otherwise state that they touch-type probably mean that they can type without seeing the keyboard by using some improvised version of two or three finger methods.
I go with my brother's theory that the force per keystroke plays a bigger role in finger/hand strain than the functional layout itself.
Thanks for sharing your views, though. It's always interesting to understand the experience of others who have been successful in learning an alternate layout and touch-type at the same time. I am confident that both view-points are right at different levels. The only disagreement we have is probably in the proportion of computer users who actually touch-type.
Comment by Hari (blog owner) on Wed, Jul 31, 2013 @ 10:37 IST #
Uhh, touchtyping is a lot easier than you're making it sound. The only essential part of touchtyping is your fingers' default position, and that's easy enough to remember (even if you don't, most keyboard manufacturers just happen to put bumps to help you do so). Once there, there is only so much choice for which finger to use with which key. Even if you use the adjacent one for some keys (as I pointed out on the colemak forums) that doesn't really make much difference.
To keep all the discussion organized, I'm going to reply to the related from the colemak forum:
To be more precise, I was disagreeing about the proportion of non-touch-typists who'd, in fact, overcome QWERTY's shortcomings.
My point is not entirely orthogonal to yours', because one has to consider the base rate as well. Even if you show up at the doctor with all the symptoms of a 1 in a billion disease, for example, it's still pretty unlikely that you actually have that disease just because its base rate is so rare. Similarly with the 1% typing style.
It's possible that any particular reader might be in the 1% of non-touchtypists who'd overcome QWERTY's shortcomings, but it's not all that likely. When compounded with humans' tendency to rationalize what we're comfortable with as "probably" being good enough (similarly to the blub paradox), I felt the blogpost's last sentence, for example, was much more likely to mislead than inform.
Comment by lalop (visitor) on Wed, Jul 31, 2013 @ 12:46 IST #
While adhoc may not be ideal, I think my point about non-touch qwerty typists not necessarily facing the same problems described by studies showing higher strain of hand and fingers in qwerty touch-typists and that the concerns may not be wholly relevant, holds.
As far as touch typing being easier or more difficult to learn, I think it is neither trivial nor easy and requires quite a bit of dedication not to slip back to old ways, especially for long time users.
P.S. I'll look into the BBCode tag problem. Thanks!
Comment by Hari (blog owner) on Wed, Jul 31, 2013 @ 13:22 IST #
I agree that (the scant) analysis between layouts mainly considers 'home row touch typing'.
Touch typing was invented after the Qwerty layout and typewriter was invented. It's relatively modern!
Dvorak focused his attention on home row touch typing. And as such, other layouts are heavily influenced by this. Colemak is heavily influenced by Dvorak.
Also layout analysers and optimisers, rarely mess with all 100 or so key placements - as I guess there are just too many possibilities- Factorial(100) permutations at least! So the optimisers need to start somewhere to narrow the scope of the problem. And as you say they don't necessarily consider the hunt and peck style, or alternative blind typing layouts or alternative physical layouts!
In short it's a difficult problem to crack, which isn't helped with a lack of rigorous good scientific study. It's been suggested that Dvorak's claims lack good scientific backing, and sadly we haven't got any decent justification or studies to back; even modern layouts.
So you are relying on anecdotal evidence, and a handful of algorithms and your own common sense I'm afraid.
I for one have wasted a lot of time putting off learning how to touch type. I believe I have injured myself with my previous half blind hunt and peck methods - the reason I sought touch typing and alternative layouts in the first place.
Personally I really wish I'd learnt how to touch type (10 or 6 fingers) from an early age. There are loads of young people that are notching up rapid respectable WPMs under Qwerty. I'll wager that the earlier you learn the easier it is. For me I've gained enormously from being able to touch type, and touch typing is worth it if you can be bothered to learn it, but it's a skill that takes time and attention.
Learning the layout map took me less than an evening and was cemented in my brain in a matter of days. Proficiency and speed take far longer.
I found 10 finger home row touch typing very alien, like you, because it was a very different style to what I was accustomed too.
Still I feel the main resistance I meet when typing is from the crappy ergonomics of existing keyboards, that have barely changed in design in the last 20 or so years. The standard computer keyboard is antiquated, and the alternative layouts are desperate hacks. That's not to say that Dvorak wouldn't lend itself well to something like the Datahand or other digit based input devices.
Text entry via the keyboard is a laborious barrier to interfacing with computers and other people. I give them less than a decade.
Comment by Archie Pink (visitor) on Wed, Jul 31, 2013 @ 17:51 IST #
Very interesting views, though I tend to disagree with your final thought that text entry via the keyboard being a laborious barrier. I think good keyboards make typing a pleasure and that alphanumeric keyboards offer the most fine-grained input control on any computer or electronic device. It will be hard to replace the keyboard for stuff like programming and long data entry, in spite of speech recognition and other technologies.
Comment by Hari (blog owner) on Wed, Jul 31, 2013 @ 18:42 IST #
I don't think keyboards are particularly elegant at all. You may be fast at keyboard entry. Probably because you've spent many hours honing that skill.
I know many people that didn't grow up with the luxury of access to a typewriter or computer. I work with people that have been working with computers on a daily basis for at least a decade and they don't touch type, further they can barely type well or fast at all!
These people fail to communicate via email, they fail to take good notes via the computer, and are pretty lousy when it comes to data entry. They struggle, even if they aren't aware that they are struggling.
Glance at social media sites and forum posts and you'll see erroneous texts circumscribing scant content. I prefer to think that this is a limitation of text entry rather than that of mind. New skill sets may evolve from those very limitations. For example some people are very impressive at distilling thoughts and ideas.
A conversation between two people is very rich compared to text input. Any text input method slower than human thought and speech will stifle communication.
I introduced a 70 year old to their first laptop, and they were completely debilitated by the computer keyboard for a protracted period of time. They had never even used a typewriter and barely saw any rhyme or reason in the layout and find them very awkward.
On the flip side, there are many people that can type fast who have yet to suffer from typing related injuries.
The computer keyboard (and layouts) are a throwback to the days of typewriters (and their own physical limitations). There's a certain charm in skeuomorphism, but when we have software and 3D printers why limit ourselves?
Other physical, touch, kinetic and voice devices, and even nervous system implants could easily surpass what we have today in little time, if we had the motivation to do so.
Take MessagEase - a touch based input method - people seem to reach an impressive WPM with that, I've even read accounts of people logging into their servers over SSH with it. Part of the attraction of mobile computing (as in phones and tablets), is that they are somewhat easier to use for the masses. Even infants can use them!
Comment by Archie Pink (visitor) on Thu, Aug 1, 2013 @ 15:50 IST #
> Any text input method slower than human thought and speech will stifle communication.
I was really hinting here at real time communication. (You could potentially write an amazing book via morse code - e.g. The Diving Bell an dthe Butterfly.)
Comment by Archie Pink (visitor) on Thu, Aug 1, 2013 @ 16:01 IST #
Comment by Hari (blog owner) on Thu, Aug 1, 2013 @ 18:13 IST #
Also the first quote is the one that usually fails, so avoiding bad formatting.
That was the same thing as what I meant. Advanced multi-finger hunt-and-peckists can "overcome"/"work around" QWERTY's questionable design features by, essentially, using a dynamic rather than fixed positioning. My contention, however, is that this is so hard (and the population of would-be improvers so vulnerable being seduced by touchtyping) that one cannot expect more than the barest minority to have succeeded.
I'm not saying touchtyping is easy (it certainly requires practice), but it definitely is easier than you were making it sound. The way you described made it seem like people can't even understand touchtyping without "formal training", when in fact, the home row positioning pretty much dictates everything (modulo some small finger choice variation that's not really important anyway).
120wpm is insufficient for recording speech, much less thought. Court reporters, for example, are required to hit 225wpm in order to be certified.
If I were to redesign the computer from scratch, it would probably be a tablet with a stenotype taped at the back edges.
Comment by lalop (visitor) on Sat, Aug 3, 2013 @ 02:57 IST #
Comment by Luca (visitor) on Wed, Jan 8, 2014 @ 16:29 IST #
I was a hunt and pecker who tried to learn touch typing (with Dvorak) in my mid-thirties.
My two mechanical keyboards I too find stiff. My laptop has a keyboard that doesn't feel great, and I'm also used to an Apple keyboard that I don't loathe or like.
Half blind hunt and peck was a pain, but when I typed like that, I never felt it in my fingers (I did feel it elsewhere).
Some haven't a problem and take to touch typing like a duck to water. For me it has been a struggle. I learnt how to touch type in reasonable time, but I can't say it has ever been particularly comfortable, cramming my paws into the grid. It still feels alien and not pleasurable, but it has gotten easier.
I find keying phone buttons and using chip and pin machines difficult also, so it could just be my hands and fingers (or technique). I was more enamoured with keyboards and keyboarding before I began touch typing.
In short: I _kind_ of agree with your initial post.
Comment by Archie Pink (visitor) on Thu, Apr 24, 2014 @ 23:11 IST #
Personally I now use a tactile clicky mechanical keyboard (Cherry MX Blue switches) and find it much more comfortable to type on. A tactile light switch definitely helps because you don't need to bottom out the keys to register them (you need to practice typing light, so as not to bottom out the keys).
So if you're finding a particular mechanical switch quite stiff, I suggest looking at an alternate lighter switch. Or else, you should practice not bottoming out keys.
Comment by Hari (blog owner) on Fri, Apr 25, 2014 @ 07:56 IST #
When touch typing, I couldn't use those old keyboards as it felt harder on my fingers. Perhaps that's a load issue, or maybe I used my arms more, to throw some weight behind them.
Not bottoming out is possible on my mechanical, but it's still stiff and requires a dead on vertical strike.
Annoyingly I haven't got a comfortable seating position for the full sized keyboard, so end up using the laptop most of the time (stiff scissor switches).
The cold brings about aches and pains. I am very sore this morning.
Comment by Archie Pink (visitor) on Fri, Apr 25, 2014 @ 14:11 IST #
What is your mechanical keyboard, by the way? On my laptop also I use my mechanical keyboard whenever I have to type big documents because it is more comfortable.
Comment by Hari (blog owner) on Mon, Apr 28, 2014 @ 13:15 IST #