• You are not logged in.
  • Index
  • General
  • What's wrong with the Dvorak layout?

What's wrong with the Dvorak layout?

  • Started by wibble
  • 62 Replies:
  • Reputation: 0
  • Registered: 01-Apr-2008
  • Posts: 6
wibble said:

From the FAQ at https://colemak.com/FAQ#What.27s_wrong_ … _layout.3F
It's not comfortable to use Ctrl-Z/X/C/V shortcuts with the left hand while holding the mouse with the right hand...

I read a study somewhere, either on BBC or from slashdot, that an experienced showed that using mouse on your left hand result in a more natural typing position since the ABCs are centred in front of you instead of aligned to your left.  If you use your mouse right-handed, the numpad and mouse both need right space squeezing the main typing area to the left.

But what is most convincing is that over half of the subjects keep using the mouse on their left hand side after the experiment ended.  70%? 80%? Whatever.  That's enough for me to switch.

Once you get used to it, it really feels more natural.

EDIT: But then, when you use those commands, shouldn't your hands be on the keyboard typing rather then holding the mouse?

Last edited by Sheepy (02-Apr-2008 05:13:38)
Offline
  • 0
  • Reputation: 0
  • Registered: 03-May-2008
  • Posts: 7
Sheepy said:
wibble said:

From the FAQ at https://colemak.com/FAQ#What.27s_wrong_ … _layout.3F
It's not comfortable to use Ctrl-Z/X/C/V shortcuts with the left hand while holding the mouse with the right hand...

I read a study somewhere, either on BBC or from slashdot, that an experienced showed that using mouse on your left hand result in a more natural typing position since the ABCs are centred in front of you instead of aligned to your left.  If you use your mouse right-handed, the numpad and mouse both need right space squeezing the main typing area to the left.

But what is most convincing is that over half of the subjects keep using the mouse on their left hand side after the experiment ended.  70%? 80%? Whatever.  That's enough for me to switch.

Once you get used to it, it really feels more natural.

EDIT: But then, when you use those commands, shouldn't your hands be on the keyboard typing rather then holding the mouse?

Even better for righties like me is to get a keyboard without the numberpad hanging off the right, and buy a separate keypad/calculator to place on the left -- much better. That way G is center, but mouse and numberpad each have wrist neutral positions, requiring only a simple pivot at the elbow if you have an ergo keyboard. Entering numeric data into  random cells in a spreadsheet is like lighting that way...

The diNovo, Goldtouch, and Happy Hacking keyboards are good for this (or Apple's wireless), IMHO. I always thought it strange that most keyboards are actually for left-handed/mousing people!

BTW, using the http://www.siteuri.ro/dvorak calculator shows me that with my own php or javascript/html code or pasting in the main Cocoa Objective-C source from OmniWeb Base, or Webkit shows that Colemak has better left/right balance than Dvorak (usually the best in everything except left-right alternation -- the design goal of Dvorak). A real eye-opener for me to try Colemak. The one-hand Control-ZXCV is less an incentive because it is Command on macs; it is really awkward to hold Command (thumb?!!) instead of Control no matter what, so no value there for me. Colemak wins in other ways.

Last edited by swarmofkillermonkeys (03-May-2008 23:40:15)
Offline
  • 0
  • Reputation: 0
  • Registered: 08-Mar-2008
  • Posts: 303
swarmofkillermonkeys said:

BTW, using the http://www.siteuri.ro/dvorak calculator shows me that with my own php or javascript/html code or pasting in the main Cocoa Objective-C source from OmniWeb Base, or Webkit shows that Colemak has better left/right balance than Dvorak (usually the best in everything except left-right alternation -- the design goal of Dvorak). A real eye-opener for me to try Colemak. The one-hand Control-ZXCV is less an incentive because it is Command on macs; it is really awkward to hold Command (thumb?!!) instead of Control no matter what, so no value there for me. Colemak wins in other ways.

I have tested Arensito, Klausler, Colemak, and Capewell on that calculator, and Colemak does the best. Actually DDvorak does better, but I'm sure that's because of the special characters layout. Colemak has better same finger than even Arensito, whose goal was to reduce same finger, and has more hand alternation than Arensito or Capewell. It also has the least home row skips. It performs better in every category. It's like some sort of magic super layout.

Offline
  • 0
  • Reputation: 0
  • From: Switzerland
  • Registered: 21-Aug-2007
  • Posts: 176
swarmofkillermonkeys said:

The one-hand Control-ZXCV is less an incentive because it is Command on macs; it is really awkward to hold Command (thumb?!!) instead of Control no matter what, so no value there for me. Colemak wins in other ways.

Personally, I feel exactly the other way. I prefer using the Command key (with the thumb) rather than the Control key - which I feel is awkward with the thumb, and I'm not used to do it with the pinky, since I've been using Macs for years and Windows only occasionally. (And recently I've switched to a Kinesis Contoured keyboard, which makes even more use of the thumbs)

Anyway, you do know that you can easily swap the Command and Center keys in System Preferences > Keyboard & Mouse > Keyboard (Tab) > Modifier Keys... (Button) ?

Last edited by boli (06-May-2008 21:39:01)
Offline
  • 0
  • Reputation: 0
  • Registered: 03-May-2008
  • Posts: 7
boli said:

Anyway, you do know that you can easily swap the Command and Center keys in System Preferences > Keyboard & Mouse > Keyboard (Tab) > Modifier Keys... (Button) ?

Say... no, I did *not* notice that somehow. Thanks! After trying, I guess I'm halfway trained; I tend to use same hand thumb/command to hit Cmd-W, Cmd-S, Cmd-Q easily, but still retain my ancient same hand Ctl-ZXCV habit when working quickly. What a basketcase I am! :)

I'll have to give it all some thought so that I get this change right. Which reminds me that the keymap for colemak on the Mac appears to be something of a basketcase itself. The extended characters shown by option, and option-shift don't appear to match either the Apple-Roman or the Apple-Unicode sets (Apple's Dvorak looks a translation of Apple-Roman). There seem to even be many unmapped keys that default to ~.

Is there a reason for that or is it a work in progress? I'm sure that there are many Mac users that would help if that is what is needed. Including me, but honestly don't know what I'm doing. I downloaded a layout editor (ukelele) to see if the latest version would remap CAPS to BS (it won't) and quit playing at that point.

The third way of making a different character set than either Roman or Extended, optimized by either frequency or logical layout would be great, if a large undertaking. The new Extended unicode layout is overkill with the stacking diacritics, but is missing useful common things like π, ∆, Ω, ≈, etc. But the old roman wastes slots on (apple symbol) and NON-stacking accents in addition to the stacking ones. Not to mention nifty unicode that is missing from either that might come in handy were it there like all the fractions (½⅓¼⅕⅙⅛), all the major currencies, and the Cmd, Opt, and other symbols for user instructions. What a mess.

But I wonder if going that far to make a standard Colmak mac layout is too far a deviation? Or is my 10.4 not reading the layout I downloaded here correctly? (sorry for the tangent from the topic)

Last edited by swarmofkillermonkeys (07-May-2008 13:47:37)
Offline
  • 0
  • Reputation: 0
  • From: Switzerland
  • Registered: 21-Aug-2007
  • Posts: 176

Still off topic - sorry.

swarmofkillermonkeys said:

I downloaded a layout editor (ukelele) to see if the latest version would remap CAPS to BS (it won't) and quit playing at that point.

As you noticed this won't work. I used the DoubleCommand kernel extension to make CAPS work as BS. The link is at https://colemak.com/Mac
It will be a global change though, that is it'll still be active when switching to other layouts like QWERTY. It can be enabled/disabled any time in its preferences pane.

BTW I agree the characters typed with option and option+shift are quite a change from the Mac-norm, but as I use them little enough I didn't bother changing them, yet (might switch the locations of the umlaut characters some time). I also noticed that there are lots of unmapped keys with the option+shift combo, so I guess your experience with 10.4 is normal.

Last edited by boli (07-May-2008 22:12:48)
Offline
  • 0
  • Reputation: 0
  • Registered: 03-Jul-2009
  • Posts: 189

I apologize for resurrecting this old thread, but I'd like to share out some gripes I have about shortcomings of Dvorak stated in Colemak FAQ, which have not be mentioned in this thread.

Disclosure: I'm a Dvorak typist. I've tried Colemak, but I don't think it works well for my language (Italian) as Dvorak does.

Placing 'L' on the QWERTY 'P' position causes excessive strain on the right pinky. Colemak doesn't place frequent letters where the pinkies stretch.

If you are stretching your fingers to reach for keys, then you are doing it wrong. You should be moving your whole hand while typing. Watch videos about competent typists.

'L' and 'S' form a frequent same-finger digraph on the right pinky. Same-finger for the pinky is very rare in Colemak. In particular, Unix commands such as 'ls -l' are very uncomfortable to type.

We agree that Colemak has lowest same-finger ratio. However, why are Unix users fixing their layout when it would be easier to write an alias for "ls"? BTW, I think standard placement of punctuation sucks if you are an IT guy.

Another misunderstanding - I think - about Dvorak is that it is all about hand alternation. I don't think so. Dvorak did try to maximize rolls, but without having them backfire with same hand gimmicks (think "start" in Colemak).

I've much respect for Shai's effort, and for this community, and I'm happy that we have Colemak. I just wanted to point out some misleading opinions about Dvorak.

Thanks for reading.

Last edited by spremino (01-Apr-2010 14:17:28)

Dvorak typist here.  Please take my comments with a grain of salt.

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

Just, frankly don't agree about L, but then that's why after using Dvorak a couple times over the years, I'm now finally a touch typist on Colemak. 

It's not misunderstandings.  It's a difference of opinion.  Flapping my arms around for L isn't a solution, it's a work around to a problem.

Offline
  • 0
  • Reputation: 0
  • From: Berlin, Germany
  • Registered: 13-Mar-2010
  • Posts: 17
spremino said:

If you are stretching your fingers to reach for keys, then you are doing it wrong. You should be moving your whole hand while typing. Watch videos about competent typists.

Personally I learned Dvorak, but once I was using it all day long and reached 35 wpm, I noticed intense pains in my right pinkie.  I'm pretty sure it was related to the placement of L which is very frequently typed.  That's when I decided to admit that my month of training was worthless, throw it away and learn Colemak.

Learn Colemak, no download necessary: www.learncolemak.com

Offline
  • 0
  • Reputation: 1
  • From: New York, New York
  • Registered: 22-Nov-2008
  • Posts: 130

L and S on the right pinky broke the deal for me. Probably the only reason I learned Colemak. Glad I did so.

I can type 'start' pretty comfortably too now. Maybe worth noting.

Colemak typist

Offline
  • 0
  • Reputation: 0
  • Registered: 03-Jul-2009
  • Posts: 189
keyboard samurai said:

It's not misunderstandings.  It's a difference of opinion.  Flapping my arms around for L isn't a solution, it's a work around to a problem.

Yeah, it's a difference of opinions. I don't think it's a workaround though. "Workaround" means "working around" a flaw. But hovering your hands over the keys isn't working around anything, it's the right way to type. If someone prefers sloppy typing, that's fine, but please don't fault Dvorak.

EDIT: I'm not arguing Dvorak shares same-finger statistics with Colemak. Of course Colemak is better in some areas. I'm just saying the FAQ is exaggerating impact of some Dvorak's design decision. Maltron places L exactly in the same place as Dvorak does. That should say us something.

Last edited by spremino (06-Apr-2010 10:27:28)

Dvorak typist here.  Please take my comments with a grain of salt.

Offline
  • 0
  • Reputation: 0
  • Registered: 03-Jul-2009
  • Posts: 189
juice43 said:

I can type 'start' pretty comfortably too now. Maybe worth noting.

I believe you. "Start" was just an example of what I think Dvorak was trying to avoid. Maybe that depends on what kind of keyboard you are using (membrane, scissor-switch, ecc.).

Oh, and another gripe with Colemak's FAQ has just springed into my mind:

Even though the design principles are sound, the implementation isn't optimal because it was designed without the aid of computers.

I don't agree. Designing a keyboard layout is not rocket science, once you have put down statistics about your language and you are able to do your math. Computers just allow you to employ "dumb" algorithms and get the same results quickly. Peter Klausler tried to design a layout following Dvorak's principles, but with "the aid of computers", and here is his own conclusion about his "better" layout:

But (perhaps not surprisingly), it sure looks a lot like Dvorak, too, and is not quantifiably all that much better, and that advantage is probably less than the level of error in my work estimation function.
(source: http://klausler.com/evolved.html)

Last edited by spremino (06-Apr-2010 10:50:09)

Dvorak typist here.  Please take my comments with a grain of salt.

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

If you'd look at the source code for the third experiment, you'll see that Peter Klausler's research is incredibly flawed, based on very bad statistics. My conclusion is that after collecting extremely biased statistics of timing data from a single Dvorak typist typing on the Dvorak layout, that a computer program couldn't produce a better keyboard layout.

e. g.
    28, /* h -> i -> p ms */
    1033, /* t -> i -> p ms */

So according to his data t-i-p takes 36 times longer to type than h-i-p.

Offline
  • 0
  • Reputation: 0
  • Registered: 03-Jul-2009
  • Posts: 189

Shai, thanks for replying.

My quote from Klausler was about his second experiment, which - according to his description - was not skewed by data collected from his typing. However, your observation shows that having computers doing number crunching isn't enough; you have to make them chew the right numbers.

Happy typing ^_^

Dvorak typist here.  Please take my comments with a grain of salt.

Offline
  • 0
  • Reputation: 0
  • Registered: 01-May-2009
  • Posts: 68

This kind of back and forth is exactly what we need.  The fact is, we do not have the tools to easily figure out which metrics are worth a damn, despite how possible they are and how much is riding on what they would give us.

Look at it this way.  As far as keyboards go, I think the most credible research goes to Dvorak, despite being from so damned long ago.  I mean no insult to Colemak, just that Dvorak was created by observations of tons of real typists, doing real academic research.  We can't do that--it would cost too much money.

So every other layout seems to be based on naked hunches about various "metrics."  Basically, you pick metrics and pick how much they weigh compared to each other and then, combined with your design restrictions, you end up with an optimal layout.

But the hunches about metrics have to be swallowed whole over the credible academic research of Dvorak.  This just doesn't make sense.

Instead of sniping back and forth about whose hunches are better, or how we need something better than Qwerty but not Dvorak (because of some design restriction that Dvorak didn't follow--such as keeping more keys in place or shortcuts in place--or worse, such as something NOT a design restriction but because your hunches contradict Dvorak's research) ... instead of that, if we are going to be creating these community projects what we ultimately need is a toolkit that makes it possible to choose "hunches" in a way that can ultimately approach the credibility of Dvorak's work.

For instance, programs that gather typing data on hundreds or thousands of users and compile it together in a meaningful way would be a start.  Only with such data could we start realistically approaching some of the "big questions" about metrics.  Only once we have credible answers can we start to truly approach the goal of surpassing Dvorak (in fact) while having different design goals than Dvorak.

You shouldn't replace Dvorak's science with pseudoscience.  Write real tools that bring us towards a more scientific approach than crunching corpuses against arbitrarily weighted metrics.  Metrics should be rules based on observations of real typing--from many users who use many different layouts.  They should translate into speed.  Minimizing injury will have to be based on older research still though, I guess.

Offline
  • 0
  • Reputation: 0
  • Registered: 03-Jul-2009
  • Posts: 189
klalkity said:

As far as keyboards go, I think the most credible research goes to Dvorak, despite being from so damned long ago.  I mean no insult to Colemak, just that Dvorak was created by observations of tons of real typists, doing real academic research. [...] You shouldn't replace Dvorak's science with pseudoscience.

Well said.

Dvorak typist here.  Please take my comments with a grain of salt.

Offline
  • 0
  • Reputation: 0
  • Registered: 01-May-2009
  • Posts: 68
spremino said:
klalkity said:

As far as keyboards go, I think the most credible research goes to Dvorak, despite being from so damned long ago.  I mean no insult to Colemak, just that Dvorak was created by observations of tons of real typists, doing real academic research. [...] You shouldn't replace Dvorak's science with pseudoscience.

Well said.

At risk of detracting from the main point of my prior post further, I would like to add some further remarks to take away from the argument you have cherry picked from me and misrepresented me as putting forth.  I will do this below, but let me first say that my original post basically claimed that we need to develop better tools so that we can have better data and truly have credible data that can add to, or question Dvorak's work.  I believe this is within reach of an open source free community type project.  As it stands now almost every hobbyist who seeks to design a keyboard rolls their own tools and compiles their own data--we should consolidate this work so that it can be more credible so that it may have the credibility to question Dvorak's research (and the couple other typing research projects that have gone on).

As I said in my prior post, Dvorak's research seems to be the ultimate.  This does not mean that his keyboard layout is, however.  One may use the principles he discovered and used and create a keyboard layout with different design constraints.  I believe that for the most part, Colemak appears to do this.  If Colemak was more "open sourced" and less "marketed" I believe we would see this.  As it stands now, a lot of Colemak's design feels like a bit of a black box to me.  Its design choices and principles are not completely explicit.  We can't repeat his results, so to speak--and consequently we can't see to what extent it is built on real research vs. hunches.  But even in the "marketing" material of Colemak, and in what data we can gather from it, clearly it seeks to do some of the things Dvorak does as well.  Others it clearly rejects.  And it has different design constraints (learnability from QWERTY and common shortcuts remaining the same).  It also meets demands that Dvorak never tried to, such as internationalization.  Colemak might be somewhat a mystery, somewhere between a mixed bag and an improvement on Dvorak (depending on how he designed it, which is the black box I referred to above).  However, even as a mixed bag, many people have good reasons to switch to it, choosing it over Dvorak.

:P

Last edited by klalkity (08-Apr-2010 22:51:41)
Offline
  • 0
  • Reputation: 0
  • Registered: 03-Jul-2009
  • Posts: 189

Dear Klalkity, I meaned what you meaned.

I did not state that Dvorak is sound while other layouts are not. Even amateurs are able to come up with sound designs. I have already stated my respect for Shai's work. Yes, people may prefer Colemak, that is fine. But please, use solid arguments when criticizing Dvorak, because he maked his homework.

What has striked me is that critiques against Dvorak - here and elsewhere - lack foundations. "You must stretch your pinky to reach for L" is bogus: you should not be stretching your fingers at all. Why do not you just say: "If you are a sloppy typist, then Colemak will work better"? "Dvorak jeopardizes XCV shortcuts" should be: "If you are neither able nor willing to learn alternative shortcuts - which by the way seem better because they keep your mouse nearer - then Colemak will work better". And: "If you don't know how to make an alias for the 'ls' command, then Colemak will work better.". And so on. That would be an honest comparison, I think so.

Moreover, I have seed many designers judging a layout only from their own experience. For example, I have readed that alternating hands is tricky. Indeed it is... but only at the beginning, then your coordination  improves with practice (and with usage of a decent keyboard, which is not stressed enough).

Happy typing ^_^

Last edited by spremino (09-Apr-2010 00:26:39)

Dvorak typist here.  Please take my comments with a grain of salt.

Offline
  • 0
  • Reputation: 0
  • Registered: 05-Jan-2010
  • Posts: 91

I would suppose that many of the points of critisism aimed at Dvorak here on the site are formulated from a personal viewpoint and as a way of promoting Colemak.  I have tried Dvorak, and while I don't see the point in making a fuss about ZXCV shortcuts (I rarely use them anyway, plus what about the S[ave], O[pen], F[ind] and other common shortcuts?) I do generally agree about the L issue, but then again, sometimes I'm what you seem to prefer to call a ``sloppy typist''.  Right now I'm half lying in my bed, with my laptop at my side, and I rest my left elbow on the bed, and the right hand on the computer.  This is a situation the layout should cope with as well, if you ask me.  But then again, that's a personal opinion.  Like a lot of other things about layouts.

It might be true (I don't want to say it is, even though I'm pretty sure) that the issues regarding Colemak on the site are those taken in consideration by Shai when designing Colemak, intentionally or not (some might be a side-effect in the process).  And, as every user of Dvorak would be able to say, not every point might be true for every individual.  In fact, anyone, Dvorak typist or not, should be able to say that.  I do agree that they are stated in a somewhat objective fashion, as if they were objective, but as you, me and most of us know, they are not.

If someone would start a website promoting Dvorak, I bet you it'll manage to find deficiencies in Colemak too, and probably formulate them in ways close to how it is here.  Of course, I'd be happy to be proven wrong here, but I wouldn't care if I were right either.

...And then I had some more interesting point but it got lost somewhere in the to-write-stack of my mind.

Cheers.

Offline
  • 0
  • Reputation: 0
  • Registered: 03-Jul-2009
  • Posts: 189
kqr said:

I'm what you seem to prefer to call a ``sloppy typist''.  Right now I'm half lying in my bed, with my laptop at my side, and I rest my left elbow on the bed, and the right hand on the computer.  This is a situation the layout should cope with as well, if you ask me.  But then again, that's a personal opinion.  Like a lot of other things about layouts.

I didn't mean "sloppy" in a dismissive way. I sometime type while in my bed, too. I just think that criticisms against Dvorak look stated in an objective way, while they are not, as you agree later:

kqr said:

I do agree that they are stated in a somewhat objective fashion, as if they were objective, but as you, me and most of us know, they are not.

That's what I meaned when I quoted Klalkity about "pseudoscience".

kqr said:

If someone would start a website promoting Dvorak, I bet you it'll manage to find deficiencies in Colemak too, and probably formulate them in ways close to how it is here.

Maybe. But then Colemakkers would have their right to retort.

Of course if someone chooses Dvorak over Colemak, that's because she thinks it will work better. What she will not think will be that Dvorak will work better for her. As I have already stated, I choosed Dvorak over Colemak because it feeled better for my native language. Your mileage may vary.

"Imagine there's no Qwerty,
It's easy if you try
No RSI in front of us
Just ten fingers which fly
Imagine all the people,
typing their own way..."

Happy typing ^_^

Last edited by spremino (09-Apr-2010 09:18:46)

Dvorak typist here.  Please take my comments with a grain of salt.

Offline
  • 0
  • Reputation: 0
  • Registered: 05-Jan-2010
  • Posts: 91
spremino said:

As I have already stated, I choosed Dvorak over Colemak because it feeled better for my native language.

Out of curiosity, which is your native language?

Offline
  • 0
  • Reputation: 0
  • Registered: 03-Jul-2009
  • Posts: 189
kqr said:
spremino said:

As I have already stated, I choosed Dvorak over Colemak because it feeled better for my native language.

Out of curiosity, which is your native language?

Italian.

If you're interested, here are results I've got by running the evaluating applet on a classic Italian novel ("I Promessi Sposi" alias "The Betrothed"). Dvorak's finger load is more even than Colemak's one.

Dvorak: 12% 12% 12% 14% -- 10% 10% 15% 12%
Colemak: 12% 6% 10% 16% -- 15% 16% 9% 11%

Dvorak's travelling distance is 25% more than Colemak's one, but I reduced that to 17% by swapping U and I in Dvorak.

Dvorak's usage of bottom row is 7%, Colemak's is 14% (swapping C and F would reduce it to ~10%, but that would make RC and CR uncomfortable, I feel).

Same finger is similar (< 3%).

Then we come to personal perceptions. Dvorak's same hand is 20%, Colemak's is 34%, and since Italian's words are usually longer than English' ones, I felt them more awkward to type. I felt placement of L in Colemak uncomfortable on desktop keyboards (LE and EL are frequent digraphs) which I prefer. Most words in Italian end with a vowel, so consistently hitting the space bar with my left hand on Dvorak felt better.

I tried customizing Colemak, but my customizations always felt worse than the original. It's obviously a finely tuned layout ;-)

Cheers.

EDIT: Hacked the sample text to get more statistics.

Last edited by spremino (09-Apr-2010 15:55:26)

Dvorak typist here.  Please take my comments with a grain of salt.

Offline
  • 0
  • Reputation: 0
  • Registered: 05-Jan-2010
  • Posts: 91

Okay I was mainly wondering because I felt exactly the opposite about my native language, Swedish.  Unfortunately I don't really have any fancy statistics to show off.

Offline
  • 0
  • Reputation: 7
  • Registered: 21-Apr-2010
  • Posts: 818

I've post edited my first post to include a little background (sorry if the tense is all over the place).  I had only been touch typing with Dvorak for a short while at this time.

Background and motivation for learning an alternative layout
--

I experience and have experienced a lot of neck and shoulder pain.  This I mostly attribute to typing and using a computer.  I did technical support work before web programming, and that meant a lot of desk swapping in unfamiliar and uncomfortable work environments.

The pain got so bad that at times I could barely use a computer.  I looked into ways that could possibly alleviate stresses and strains.  Touch typing was a candidate idea.

I'd heard about Dvorak as an alternative to Qwerty.  And figured that if the positive claims about it being more modern and easier to learn were true, then I may as well learn how to touch type using this alternative layout.

At the time I also discovered Colemak.  It was difficult to decide between the two layouts (I later discovered many other layouts).  I chose Dvorak, mainly because most operating systems had Dvorak support baked in.  And as I machine hop often it had better native support (it is more portable).

I'd always wanted to learn how to touch type, but I'd been putting it off for years.  I decided to give it a shot during a hiatus.

Initial impressions of Dvorak
--

The layout was quite easy to pick up, as in which letters related to which keys.  My initial typing rate was very slow.  I thought that keyboard short-cuts could be a pain, but after a while I acclimatised.  (It's almost easier to think of short-cuts in terms of letters.  You read them out in your head and your hand naturally finds them.)  Initially I kept closing windows when I was trying to paste.  W is next to V under Dvorak.  Qwerty short-cuts: Cut, Copy, Paste, Undo appear designed with right hand mouse users in mind.

After using Dvorak for a while, you pick up on some of the subtleties of its design.  The most frequent letters that occur in the English language are placed on the home row.  It's a layout designed for (home row) touch typing.  The left hand under Dvorak has the vowels on the home row and the rest of the left hand gets infrequently used letters.  This results in the left hand not being worked as much as the right.  It also encourages hand alternation.  In contrast my right hand feels like it works harder.  The right pinky does a fair bit of reaching (backspace, enter/return, and some punctuation).  My right pinky feels over worked (partly because of the type of work I do).  This was quite a shock as I had never used it that much to type previously.  It was quite work shy.  Qwerty is left hand heavy so Dvorak is initially quite odd.

The downsides of Dvorak for me
--

* Switching back to Qwerty feels very awkward (perhaps less so with Colemak?)
* I am blind to my layout (this is also a plus point), one handed lazy typing is difficult!
* I have to rest both hands in the home position to initiate typing now.   It can feel less relaxed.

Moving between computers with different keyboards and operating systems can feel awkward:

* It takes a while to get used to other keyboards (this is probably layout agnostic).
* The method for layout switching varies.  And it can be quite opaque.
* There is a subtle difference between layouts: e.g. UK and ISO Dvorak.  These small changes can be overcome, but can be a nuisance.

Computer related injuries alleviated with touch typing?
--

I was hoping my pains might lift after the switch to touch typing.  In theory it should cut down on neck strain (the hunt and peck nod).  Unexpectedly I soon gained aches in my fingers and wrists.  Touch typing was a totally new way for me to hold my hands and fingers and wasn't comfortable for me.  After heavy practice my fingers would lock up (usually the day after).  I had never had a problem with my hands, wrists and fingers in the past from typing.  Perhaps touch typing can lend to RSI?

You read a lot of articles saying that you can reduce RSI with touch typing as it should reduce hand and wrist movement.  My fingers feel more strained as a touch typist.  Perhaps my technique isn't right.

The standard keyboard, is it flawed?
--

The number row feels awkward on the standard keyboard, the layout is not symmetrical.  It is not balanced is this deliberate?  Why model the computer keyboard on the typewriter?  The stagger doesn't feel right.

Is it really an issue moving your fingers?  I'm sure there would be more gains in totally redesigning the keyboard as we know it for ergonomic benefit. Perhaps less keys and more modifiers and chords would be better.  How about a radial layout?  Or splitting the keyboard?

You'd think someone could do an independent study in this field.  If we fail to replace a keyboard with touch devices, it would be great to unify and simplify a comfortable input devices.  With a design that suits everyone.

You can tell how unwieldy a keyboard is by watching a newcomer try and use one.

Implementing layouts and layout switching
--

I beg the UI guys (and developers) to simplify layout switching.  It currently is not and I am computer literate.  OSX (10.4) is easier to use than Windows in this regard.  I would like to be able to switch layouts at any time, and to be able to select a layout at boot time / sign in.

Learning and switching to another layout in the future
--

A concern for me is, that if Colemak is 'better' than Dvorak and I decide to change later on will that be a difficult transition?  Is it possible to hold multiple layouts in your head and switch between them painlessly?

Last edited by pinkyache (08-Aug-2014 19:04:50)

--
Physicians deafen our ears with the Honorificabilitudinitatibus of their heavenly Panacaea, their sovereign Guiacum.

Offline
  • 0
  • Reputation: 0
  • Registered: 03-Jul-2009
  • Posts: 189
pinkyache said:

I thought shortcuts would be a pain.  But you pick them up, it's almost easier to think of shortcuts in terms of letters, your hand naturally finds them.  I sometimes hit close instead of paste (wv), which is a royal pain, shame on my accuracy.

Every layout has its own difficult keys. I remember having issues with "wv", too. Some Colemakkers complain about "rs".

pinkyache said:

The Qwerty shortcuts, cut, copy, paste, undo, are designed surely for right handed mouse users?

Yes.

pinkyache said:

Left hand on Dvorak lends to vowels on the home row. The other alphabetic keys are pyjkx, and they don't get used that much, the rest is left for punctuation. My left fingers don't have to do much.

Contrastly my right hand and right pinky does alot of reaching. I actually quite enjoy moving my fingers.  But it certainly feels like the right hand is working more, despite the fact that typing is spread evenly between hands.

Yes, right hand works more, that's by design, since most people are right-handed.

pinkyache said:

My right pinky is over worked, but that might be because I am a programmer.  And it's work shy.

Yes, standard placement of punctuation sucks for programmers. Have you checked out Programmer Dvorak?

pinkyache said:

Is it really an issue moving your fingers? I'm sure there would be more gains in totally redesigning the keyboard as we know it. Less keys and more modifiers and chords, would be an improvement.  The number row feels awkward, there still feels like there is an imbalance. Staggered keys why?  Why not a radial layout? How about splitting the keyboard. Some real ergonmic improvement.

Don't hold your breath. We'll have to cope with standard staggered keyboards for a long time.

pinkyache said:

Personally I have had neck and shoulder issues, and I was hoping that they might lift with touch typing, less strain on my neck. I gained aches in my fingers and wrists. Initially after heavy practicing my fingers would lock up.  I never had a problem with my hands or wrists in the past.  Perhaps touch typing lends to rsi?

You read alot of articles saying that you can reduce rsi, having less hand and wrist movement.  My fingers feel more strained now.  However my technique might be wrong, I couldn't find many articles on correct finger positioning, it would be nice to see a good typist up close.

Just keep your wrist straight, slow down and relax. While learning, your hands will develop tenseness because you are controlling them too much.

pinkyache said:

One downside about switching to Dvorak is that going back to Qwerty feels awkward, maybe less with Colemak?

Yes, Colemak has been designed to ease transition from Qwerty. Many Colemakkers report easily switching back and forth from Qwerty.

pinkyache said:

My concern is, that if Colemak is better than Dvorak, and I decide to change later on, will that be a difficult transition, or is it possible to hold multiple layouts in your head and switch between them painlessly?

It depends. Some people have no issue when switching layouts, some others find it difficult.

Learning your first layout is hard. Learning another layout will be much less effort.

Dvorak typist here.  Please take my comments with a grain of salt.

Offline
  • 0
  • Index
  • General
  • What's wrong with the Dvorak layout?