To home row or not to home row?

[Mirek028]

I can say for myself that I manage to touch-type. But I never actively *learned* how to do it, it just came by from years of computer use.

Since joining geekhack, getting a mechanical board and feeling the difference under my fingers I started being a bit more conscious about keyboard use - and what I do notice is that while I am touch-typing, my (acquired) techique completely ignores the home row position. My hands float around the board, mostly going with the flow of text. I do use most of my fingers, if not all ten - but I do wonder is it the right way to go or to continue.

What would be the advantages of relearning to touch type from the home row? Considering not only health issues but also typing speed. I tried to force myself to type from the home row but right now it slows me down considerably.

[daerid]

Home row. Hands down.


Hehe




But seriously, the main advantage that I find from using the home row and typing "properly" is that I end up with much less finger/hand movement, since I use the finger that's closest to any given key (generally). I still have my own little quirks though: I hit backspace and tab with the ring fingers, and only hit space with my right thumb.

[Oobly]

Home row on a normal horizontally staggered QWERTY board can be bad for your hands and wrists. Especially if you have broad shoulders (increases the angle between the arms coming in to the board).

Floating will be much healthier and if you are satisfied with your current speed and technique I recommend you stick with it.

I am fairly similar (more than 20 years of coding, but never learned to type "properly") and recently decided to try it. After a few minutes I realised just how bad "home row" typing on a normal board would be over a long period of time. The wrists are forced outwards, the fingers bunched together, each finger moves different distances to the upper and lower row keys, the left fingers move diagonally one way, the right ones the same way instead of mirrored (and different amounts of sideways movement between upper and lower rows), the character layout sucks with too many same finger combinations and weird twisting motions of the wrists, the little fingers do way too much work, etc..

The only way to type on a normal board and retain your health is to "float" around with fingers resting naturally where they happen to fall. I find I still have to look at the board now and then (which is what prompted me to try to learn to touch type properly), but I'd rather do that than ruin my hands and wrists.

In response to all this I built my own ergonomic keyboard with better physical and character layouts and can touch type properly on it at 20WPM so far, but improving all the time. I have retained my old speed on normal boards while learning due to it being so different, so I don't have to unlearn anything.

[Linkbane]

I can say for myself that I manage to touch-type. But I never actively *learned* how to do it, it just came by from years of computer use.

Since joining geekhack, getting a mechanical board and feeling the difference under my fingers I started being a bit more conscious about keyboard use - and what I do notice is that while I am touch-typing, my (acquired) techique completely ignores the home row position. My hands float around the board, mostly going with the flow of text. I do use most of my fingers, if not all ten - but I do wonder is it the right way to go or to continue.

What would be the advantages of relearning to touch type from the home row? Considering not only health issues but also typing speed. I tried to force myself to type from the home row but right now it slows me down considerably.

That's probably because you use Qwerty! 
Learning to type home row will speed you up because your hands will move less (in everything besides qwerty, because home row isn't even the most used).
The way I type Dvorak is with the right thumb off the keyboard, but since the left hand moves so little, it almost never moves from the home row. My right hand has to move all over the place, though. I feel like this is the most optimal configuration, but people will do as is most efficient or comfortable.

[Tony]

With two most frequent keys (E and T) are not on the homerow, Qwerty is not a good layout to try homerow.

For better typing experience, I suggest you switch to Colemak. It has all ten of most frequent keys on the home row.

[Linkbane]

With two most frequent keys (E and T) are not on the homerow, Qwerty is not a good layout to try homerow.

For better typing experience, I suggest you switch to Colemak. It has all ten of most frequent keys on the home row.

Show Image

It also has keys in Qwerty positioning, completely anti-ergonomic. ZXCVB/QW should not be where they are, as it's very difficult to hit the upper left and bottom left keys for any speed.
Colemak really has nothing to brag about.

[rowdy]

Always remembering that the QWERTY layout was designed to make typing slower so that the mechanical levers on the olden typewriters did not stick together.  Thus frequently used letters were placed further apart on the keyboards.

[Oobly]

Always remembering that the QWERTY layout was designed to make typing slower so that the mechanical levers on the olden typewriters did not stick together.  Thus frequently used letters were placed further apart on the keyboards.

And the staggered layout is because of the levers from the keys going up to the mechanism. Both completely redundant for an electronic keyboard, and both very bad for ergonomics.

Both also retained due to people being "used to" them, which is not as relevant now as it was when keyboards were used primarily for data entry by typists. It's time for a change.

[Mirek028]

Always remembering that the QWERTY layout was designed to make typing slower so that the mechanical levers on the olden typewriters did not stick together.  Thus frequently used letters were placed further apart on the keyboards.

And the staggered layout is because of the levers from the keys going up to the mechanism. Both completely redundant for an electronic keyboard, and both very bad for ergonomics.

Both also retained due to people being "used to" them, which is not as relevant now as it was when keyboards were used primarily for data entry by typists. It's time for a change.

Indeed. But the "mechanical" reasons for this were obsolete practically at the dawn of personal computing - which was more than 30 years ago. I guess that at that time, the key point was moving over the existing trained typists to the new machines smoothly. And the problem is that that installed user base still exists, I see people type on a computer keyboard as they would on a mechanical typewriter.

My concern about Dvorak is whether it is optimal to type in languages other than English - and the interoperability given that I also frequently need to move around and change machines. And I cannot pick up any random machine and install a new keyboard layout just because. How easy it is to switch back and forth for you dvorak-typing people?

[Linkbane]

My concern about Dvorak is whether it is optimal to type in languages other than English - and the interoperability given that I also frequently need to move around and change machines. And I cannot pick up any random machine and install a new keyboard layout just because. How easy it is to switch back and forth for you dvorak-typing people?

It's pre-installed on all three major operating systems (MS, Mac, Linux) and can be made to be switched to via standard ctrl+shift or alt+shift with or without a number in about ten seconds in the control panel, or keyboard options for Mac. Alternately, you can download DVAssist and put it on a flash drive, you can activate it on any computer and remove it, which will be the same as using Dvorak. One of the many benefits over other layouts.

[theMANtonio]

My concern about Dvorak is whether it is optimal to type in languages other than English - and the interoperability given that I also frequently need to move around and change machines. And I cannot pick up any random machine and install a new keyboard layout just because. How easy it is to switch back and forth for you dvorak-typing people?

It's pre-installed on all three major operating systems (MS, Mac, Linux) and can be made to be switched to via standard ctrl+shift or alt+shift with or without a number in about ten seconds in the control panel, or keyboard options for Mac. Alternately, you can download DVAssist and put it on a flash drive, you can activate it on any computer and remove it, which will be the same as using Dvorak. One of the many benefits over other layouts.

Dvorak has always interested me... but it's also one of those things that will probably won't work for me for practical reasons.
QWERTY may not be optimal, but at least it's 'universal' in the sense that if I use any computer or smartphone (in the US), it's there for sure.

[PointyFox]

I play lots of games, so I keep my hand on WASD

[Oobly]

I play lots of games, so I keep my hand on WASD

Can always switch back to QWERTY for gaming, just Alt-Shift in Windows to change layout. On the other hand, for most games you don't even have to, you can reassign the keys. ESDF is actually better for most games anyway 

[rowdy]

I play lots of games, so I keep my hand on WASD

The number of times I've sat down to check email and my left hand automatically goes to WASD.

I've considered writing an email client that uses WASD to navigate.

[Linkbane]

Dvorak has always interested me... but it's also one of those things that will probably won't work for me for practical reasons.
QWERTY may not be optimal, but at least it's 'universal' in the sense that if I use any computer or smartphone (in the US), it's there for sure.

It's hardly important talking about the myriad of reasons why people don't use it. Nothing new, nothing noteworthy. The same thing comes from 99.99% of other people, too.
I don't personally care about having to use Qwerty for on reason or another. Any computer I ever use immediately has Dvorak because I can put in a flash drive and double click a .exe file. You're doing a disservice to your fingers, but then again, it doesn't really matter that much to most people.

You don't need to justify staying in a layout, I doubt that any alternate-layout user really minds. If you don't have something constructive to say, though, try to refrain from expressing it unless it's highly important.

[Burz]

With two most frequent keys (E and T) are not on the homerow, Qwerty is not a good layout to try homerow.

For better typing experience, I suggest you switch to Colemak. It has all ten of most frequent keys on the home row.

Show Image

It also has keys in Qwerty positioning, completely anti-ergonomic. ZXCVB/QW should not be where they are, as it's very difficult to hit the upper left and bottom left keys for any speed.
Colemak really has nothing to brag about.

You seem to have forgotten that the problem with QWERTY is mainly on the home row, not the ZXCVB/QW positions. But I suppose its fitting to take a purist attitude towards a layout created from circa 1920s theory.

I'll take the product of 21st century analysis, thank you.

Incidentally, Wikipedia has Colemak listed alongside Dvorak under "non Qwerty-like layouts".  :'(

[Burz]

Home row use can seem bad to a QWERTY typist because that layout has "E R T" and "I O" all on the top row. So for QWERTY its a "nice idea" that only works to make things simple for instructors.

On improved layouts the home row advantage becomes a reality.

[Linkbane]

You seem to have forgotten that the problem with QWERTY is mainly on the home row, not the ZXCVB/QW positions. But I suppose its fitting to take a purist attitude towards a layout created from circa 1920s theory.

I'll take the product of 21st century analysis, thank you.

I'd rather take the idea that worked for a hundred years than one that's pretty useless for speed or accuracy made ten years ago.

[Burz]

You seem to have forgotten that the problem with QWERTY is mainly on the home row, not the ZXCVB/QW positions. But I suppose its fitting to take a purist attitude towards a layout created from circa 1920s theory.

I'll take the product of 21st century analysis, thank you.

I'd rather take the idea that worked for a hundred years than one that's pretty useless for speed or accuracy made ten years ago.

And you know this how...

[Linkbane]

You seem to have forgotten that the problem with QWERTY is mainly on the home row, not the ZXCVB/QW positions. But I suppose its fitting to take a purist attitude towards a layout created from circa 1920s theory.

I'll take the product of 21st century analysis, thank you.

I'd rather take the idea that worked for a hundred years than one that's pretty useless for speed or accuracy made ten years ago.

And you know this how...

Lack of any good speed typists on any typing site using Colemak.
I've never seen a Colemak user typing over 150, despite my repeated searches. I've learned Dvorak for less than a year, and I am already able to exceed 140. I doubt any Colemak typist anywhere can say the same, even after two or three years.

[Oobly]

Similarity to qwerty means you have to retrain a skill. Different means you learn a new one. Retraining is a lot harder and can cause confusion while you switch, meaning you most likely can't keep using qwerty at work while you learn Colemak at home without working very hard at it, whereas learning Dvorak or other really different layout is easier and you can keep you qwerty skills intact while you do so.

If you want a "product of 21st century analysis" instead of a layout that's been tried and tested by actual users for a long time, how about AdNW? It's based on mostly the same criteria as Dvorak, but analysed and adapted by modern methods. It also puts U and L in better places.

If you really want to get into the "21st century analysis" debate, you have to ask yourself if they're using criteria that are relevant or not, which criteria should get what weighting, etc. Most of them are based on incorrect assumptions or incorrect weighting of factors for many typists. Just saying.

Almost anything is better than qwerty, though, and Colemak is decent (although it overloads the index fingers and is optimised for rolls which go the wrong way for my hands).

[Linkbane]

Haven't heard of that layout before, I should go take a look. At this point, I already feel as if further changes, unless minute, would be difficult to relearn at my current level. However, I would be apt to swap a key or two. I'm not a fan of the U, but I actually don't mind the L location.

[angelic_sedition]

You seem to have forgotten that the problem with QWERTY is mainly on the home row, not the ZXCVB/QW positions. But I suppose its fitting to take a purist attitude towards a layout created from circa 1920s theory.

I'll take the product of 21st century analysis, thank you.

I'd rather take the idea that worked for a hundred years than one that's pretty useless for speed or accuracy made ten years ago.

And you know this how...

Lack of any good speed typists on any typing site using Colemak.
I've never seen a Colemak user typing over 150, despite my repeated searches. I've learned Dvorak for less than a year, and I am already able to exceed 140. I doubt any Colemak typist anywhere can say the same, even after two or three years.

I'm not sure why you think this was relevant in the first place, but I did point out to you that on hi-games there are scores that meet these arbitrary qualifications. 156wpm for the 30 second test (same as the Dvorak high score). 148wpm for the minute test (higher than the Dvorak high score). Of course this is just a single site, but I'm not sure why you're still saying this unless you have some reason to believe that these scores are not legitimate.

[Linkbane]

I'm not sure why you think this was relevant in the first place, but I did point out to you that on hi-games there are scores that meet these arbitrary qualifications. 156wpm for the 30 second test (same as the Dvorak high score). 148wpm for the minute test (higher than the Dvorak high score). Of course this is just a single site, but I'm not sure why you're still saying this unless you have some reason to believe that these scores are not legitimate.

Those are still nothing high at all. Thirty second test is not indicative of typing ability, just sprint. As you can see, vonunov (i.e. Jack) has gotten 200 wpm (I'd assume on a short test), and scores in the 150s often. Again, 140 is not an impressive score for typists who have been using layouts for years and years. Colemak has been out for quite some time, and the standards are by NO means arbitrary. 120 can be reached by typists with poor technique, but anything above that requires technique.

Again, no good typists. Even in the 1970s, typing schools showed off their best at able to type over 150, on a typewriter. 140 is a rookie's score.

[keymaster]

I feel that without learning to correctly use and position my hands on the home row, I've saved my hands/wrists from injury. I type around 85-90WPM on average (100WPM max), but I let my hands float around the keyboard. I think the loss of potential WPM is worth it to reduce any kind of injury.

[Linkbane]

I feel that without learning to correctly use and position my hands on the home row, I've saved my hands/wrists from injury. I type around 85-90WPM on average (100WPM max), but I let my hands float around the keyboard. I think the loss of potential WPM is worth it to reduce any kind of injury.

Heh, QWERTY home row doesn't save you any effort. Any other format, just not that one in particular.

[angelic_sedition]

Those are still nothing high at all. Thirty second test is not indicative of typing ability, just sprint. As you can see, vonunov (i.e. Jack) has gotten 200 wpm (I'd assume on a short test), and scores in the 150s often. Again, 140 is not an impressive score for typists who have been using layouts for years and years. Colemak has been out for quite some time, and the standards are by NO means arbitrary. 120 can be reached by typists with poor technique, but anything above that requires technique.

Again, no good typists. Even in the 1970s, typing schools showed off their best at able to type over 150, on a typewriter. 140 is a rookie's score.

The point is that you claim to have made repeated searches, but it seems you haven't. Within a minute of searching I found scores that would suggest otherwise. It seems sort of dishonest to claim that there are no fast Colemak typists when you have not searched everywhere on the internet and especially not everywhere in the world.

Whether the speed of the typists I mentioned is considered "fast" is subjective and not relevant to my point.

If you're going to make a claim, I think that you should at least give the sources of your information and the basis for your standards of measurement. Yes throwing around 150wpm is an arbitrary standard unless you give the reasoning behind it. What is the base you are looking at? It's certainly not hi-games. Typing for how long? Punctuation included and actual sentences? Are we talking about fastest wpm or average? You haven't given any standards.

Okay 30 seconds is a sprint. Note that there are no Dvorak scores over 150 on hi-games except for the 30 second test. For the 1 minute test, the highest non-QWERTY score is from a Colemak user (148wpm). For the 2 minute test, the highest non-QWERTY score is from a Colemak user (140wpm). For the 5 minute test, the highest score is 134wpm from Jack.. is this the Jack that you are referring to? His highest score on any test is 145wpm.. clearly a rookie by your standards. There are tons of QWERTY scores over 150wpm. Where are the fast Dvorak typists? If the fastest typist is somehow a measure of the worth of a layout, then it would seem that QWERTY is this best. Of course other things matter, and I think you put too much emphasis on the layout itself when it comes to speed.

Well obviously I was only looking at one website for the last paragraphs (I specified which one it was though). What other sites have you looked at that list the highest typing speeds of users and their layouts? Looking at typeracer's vonunov, his record may be 200wpm, but his average is 129. His most recent scores are 162, 145, 133, 130, 123, 129, 131, 154, 124, and 105. That's quite a range with only 2 scores above 150.. so is he just another rookie? A good typist for sure, but he's also just a single person.

Another factor is the legitimacy of scores. Let's look at the rest of the best of what typeracer has to offer. mrcam1987, ranked 4, types at an average of 197wpm on Dvorak with a best race of  305wpm. This guy's average and highest race actually destroy Sean Wrona (who has an average of 189 and a high of 254). Seems a bit ridiculous to me. Is it a legitimate score? I don't know. Maybe, but we don't really have a way of knowing for sure (unless you know more about this person than me; he lives in Cambodia apparently and I couldn't find out anything about him). He does have an alternate account though with a full average of 61wpm.

Continuing down the list, it took about a minute to find a Colemak typist who types over 150wpm. So much for your repeated searches? shrimpneck has an average of 151wpm and a best of 203wpm on Colemak. His latest races: 101, 202, 188, 195, 178, 195, 200, 199, 202, and 197. This user has much better stats on typeracer than Jack. Are they legitimate? I don't know. I honestly have no idea how difficult it would be to fake high scores on a site like typeracer or hi-games.

That's the top 20. 1 Colemak and 1 Dvorak typist. In the top 50 I found a Dvorak typist with roughly the same stats as shrimpneck called rebel2000. I stopped at 50. The way typeracer organizes the top racers is admittedly annoying, but everyone else in the top 50 either specified QWERTY as their keyboard or did not specify a keyboard. Where are all the fast non-QWERTY typists? I think at 150wpm, things depend more on the typist and not the layout. There seem to be very few 150wpm+ average typists for these two websites for any alternative layout, so I do not see why you give that as a standard.

I think it's also important to note that Sean Wrona's high score is 254 on typeracer and 205 on the thirty second sprint on hi-games. This is just one example, but obviously there are huge differences between typing at 150wpm+ on hi-games vs on 10FF.

Like you've said before, the number of people using the layout and the age of the layout matter. Dvorak was created in the 1930s. Colemak was released in 2006. I think that it hardly can be said that 7,8 years is long enough to have as many fast typists as layouts that are so much older. I don't think it really matters anyway. To call Colemak useless for speed an accuracy based on no data and only the assertion that there are no fast Colemak typists based on only your limited observations seems ridiculous. If someone was actually going to conduct a study on how "good" a layout is for speed and accuracy, they would first set standards for what is meant by that and then devise some test to measure the relative performance of a similar group of people over a long period of time switching to the different layouts controlling for differences. They certainly wouldn't just look at the anomalies.

I'm just saying I don't think you have enough information to draw these conclusions

[Oobly]

I feel that without learning to correctly use and position my hands on the home row, I've saved my hands/wrists from injury. I type around 85-90WPM on average (100WPM max), but I let my hands float around the keyboard. I think the loss of potential WPM is worth it to reduce any kind of injury.

I agree with this. The physical layout is bad for extended home row use no matter what character layout you use.

[angelic_sedition]

I feel that without learning to correctly use and position my hands on the home row, I've saved my hands/wrists from injury. I type around 85-90WPM on average (100WPM max), but I let my hands float around the keyboard. I think the loss of potential WPM is worth it to reduce any kind of injury.

I agree with this. The physical layout is bad for extended home row use no matter what character layout you use.

I position my hands on the home row. I'm back up to averaging in the 90s at the moment on colemak and after switching to a wide mod and adjusting my desk, I haven't felt any pain recently. If your layout has the most frequently used keys on the home row, it makes a huge difference.         

[yasuo]

This is bottom row