The Best Of All Possible Keyboard Layouts
February 4, 2020 11:07 AM   Subscribe

How was the Sholes QWERTY type-writer designed? Was it meant to slow typists down, or to speed them up?
I added up all of the possible word matches found in my word pair database, for all of the adjacent type-bars in his keyboard. I found 3877 total word matches for these type-bar pairs. This is out of a total of 1,239,045 found word matches, or about 0.3%. So I rewrote my simulation for this new standard, trying to find keyboards with this few total word matches of colliding type-bar pairs. With this new approach, it takes on average more than five million (about 5.9 million) random tries to find a keyboard layout that is as good as QWERTY.
By the constraints of the 19th century, the QWERTY keyboard is shockingly well optimized. But we're not typing on 19th century machines anymore. And QWERTY might not have been the best solution available even in the 19th or early 20th centuries (PDF). Why can't we give up this odd way of typing?

There's an economic theory that the best solution will always win out in the market. Indeed there is a paper by Stan J. Liebowitz and Stephen E. Margolis that argues that because the QWERTY keyboard layout is still the standard in English-speaking countries (and its variants like QWERTZ and AZERTY are the dominant layouts where German and French are the primary languages), and because studies that claimed to demonstrate the superiority of the Dvorak layout were suspect because Dvorak himself was involved in their execution, that QWERTY must still be superior (PDF):
This article examines the history, economics, and ergonomics of the typewriter keyboard. We show that David's version of the history of the market's rejection of Dvorak does not report the true history, and we present evidence that the continued use of QWERTY is efficient given the current understanding of keyboard design. We conclude that the example of the Dvorak keyboard is what beehives and lighthouses were for earlier market-failure fables. It is an example of market failure that will not withstand rigorous examination of the historical record.
This paper and a Reason article that relies on it have come up here previously and previouslier. [Market theory also supports Lucy van Pelt's argument that Beethoven wasn't so great, because he never got his picture on bubble gum cards. Well, at least not until 2009.]

Liebowitz and Margolis' market theory paper, however, relies on two pillars that themselves might not be as strong as the authors hope:
Yet there are many aspects of the QWERTY-versus-Dvorak fable that do not survive scrutiny. First, the support for the claim that Dvorak is a better keyboard is both scant and suspect. Second, studies in the ergonomics literature find no significant advantage for Dvorak that can be deemed scientifically reliable.
The second point is an argument by ignorance and can be dismissed as such. On the first point they are right … and wrong. The widely cited example of the US Navy tests is arguably questionable because Dvorak himself ran the tests, but it's also maybe less worthy of dismissal than the authors might hope. A meta-analysis of that study and others in a doctoral dissertation by Bradley John Lessley from 1978 notes:
The seven research studies reviewed in this chapter were designed to determine if retraining on the Dvorak keyboard would increase keyboard operator performance. Each of these studies found that operator performance had increased through retraining on the Dvorak keyboard. Other findings from these studies indicated that retraining on the Dvorak keyboard:
  1. Could be accomplished in a short period of time.
  2. Allowed for less operator fatigue and more enjoyable keyboarding.
  3. Improved keyboarding accuracy.
It's not all good news for Dvorak, though, because even Lessley's dissertation fails to prove that training everybody on Dvorak would be worth it:
Dvorak keyboard production rates, after 100 hours of skill development keyboarding, were not significantly greater than preretraining Qwerty keyboard production rates for both CRT and element keyboarding equipment. It should be noted that production performance increased 12 and 17 percent on CRT and element keyboarding equipment, respectively, as addressed in the Discussion of Findings.
Typists got faster, they just didn't get significantly faster. In this regard the dissertation validates the conclusion of the GSA study by Earle Strong that plays a central role in the Liebowitz/Margolis paper:
The newly trained Dvorak typists continued training and a group of ten QWERTY typists began a parallel program to improve their skills. In this second phase, the Dvorak typists progressed less quickly with further Dvorak training than did QWERTY typists training on QWERTY keyboards.
Even so:
Strong's study does leave some questions unanswered. Because it uses experienced typists it cannot tell us whether beginning Dvorak typists could be trained more quickly than beginning QWERTY typists. Further, although one implication of Strong's study is that the ultimate speed achieved would be greater for QWERTY typists than for Dvorak typists (since the QWERTY group was increasing the gap over the Dvorak group in the second phase of the experiment), we cannot be sure that an experiment with beginning typists would provide the same results.
Lessley's dissertation, however, made a recommendation that still hasn't been adequately addressed:
Research projects should be undertaken addressing reduction of operator fatigue and ease of keyboard operation resulting from retraining on the Dvorak keyboard.
In short, what has been proven by previous studies is not that QWERTY is objectively better than Dvorak (or other layouts), but that the cost of retraining experienced typists is greater, on an institutional scale, than the benefit derived by those institutions after retraining. What has not been adequately addressed is whether other key layouts are universally less fatiguing than QWERTY or could be more quickly or effectively taught to new users. As QWERTY has a century of market lock-in, other layouts are fighting an uphill battle.

Operator fatigue is the primary reason the QWERTY standard still regularly comes into question. And it hasn't been addressed adequately, except as passion projects. There is no one model of typing effort or fatigue, and if you use a different model you may find a different "best" key layout. A new layout for France (en français) recently made the news, though, so there's still hope for change based on new research.

Much more work has been done on physical keyboard design, from Keyboardio (previously, also MeFi's Own) to Kinesis, from Maltron (PDF) to Microsoft (PDF). Not to mention previouslies on buckling springs or ALPS action. But that, really, should be its own post.

This post was inspired by this question on the Green and typed using QGMLWY.
posted by fedward (44 comments total) 37 users marked this as a favorite
 
I will say, however, that having learnt typing on the big manual typewriters in 1983 and being reasonably fast and accurate, the advent of lovely clicky keyboard that are well spaced (Lenovo/Thinkpads for eg. in laptops, and a basic external keyboard that clicks) meant that my fingers could fly across the keys since the resistance was far less than that of the big manuals.

My penchant for long sentences, however, remains unchanged.
posted by Mrs Potato at 11:18 AM on February 4, 2020 [15 favorites]


I used a Dvorak keyboard for a couple years in college, but when I interned at a place with Bloomberg terminals, and realized I'd probably be having to use those once I graduated, I figured it wasn't worth replacing with another Dvorak keyboard when mine stopped working.

I think my favorite part of it was watching people struggle to use my computer.
posted by Grither at 11:18 AM on February 4, 2020 [1 favorite]


A digression on touch-based keyboard layouts like 8pen (previously) and KALQ didn't stand a chance because both projects now seem to be dead. It's a pity, because KALQ in particular is really intriguing to me.
posted by fedward at 11:45 AM on February 4, 2020 [1 favorite]


Differences for experienced users are marginal. In most cases the cost of switching is higher than the benefits that might be gained, especially if evaluated from an annual return basis.

We have a mania in our culture for a singular best. Beset by choice, it's not enough to be good enough, we are told we can only accept "the best". The reality is that for most things, good enough works for almost everyone. There are many, many good enough answers. There's no real need for a best.

The same is true for QWERTY vs whatever other "best" there is. It's good enough for most people. The keyboard layout isn't the thing slowing most typists down or making them less accurate. On the other hand, there are large benefits to a standard layout. A minor (if even real) cost, outweighed by a major benefit. It's not hard to understand.
posted by bonehead at 11:58 AM on February 4, 2020 [14 favorites]


Having worked with people who used Dvorak, and having had to occasionally sit down at their computer to help them debug things: please don't do that. Make it easy to switch keyboard modes. Especially if you're using qwerty keycaps. [steadies shaking hands]

I can't find a link right now, but the PLUM keyboard layout always struck me as pretty neat.
posted by phooky at 11:59 AM on February 4, 2020 [1 favorite]


Differences for experienced users are marginal. In most cases the cost of switching is higher than the benefits that might be gained, especially if evaluated from an annual return basis.

I think the reason I am so fascinated by this history (and disappointed by the market theory argument) is that I believed the difference would be marginal, learned a new layout anyway, and increased my speed by more than 25%. I did not expect that outcome. I spent the effort because I was curious about comfort. I believe that the improvement in comfort is valuable enough to have been worth the learning curve, and gaining extra speed was just gravy.

QWERTY has a lock-in advantage but without that advantage the differences aren't marginal at all. At this point you can buy a Dvorak keyboard or even a programmable one (not just the Keyboardio I have but OLKB, Clueboard, or ErgoDox EZ, to name a few) and put whatever layout you want on it. I won't say that the advantage of lock-in is overstated, but I do think the cost of switching is.
posted by fedward at 12:14 PM on February 4, 2020 [1 favorite]


Dunno about Windows but Macs come with Colemak and four different Dvorak variants if you want to try them. System Preferences → Keyboard → Input Sources → click the plus sign.

I tried Colemak for a while but I have to be on different people's computers all the time and switching from Colemak to QWERTY all the time got annoying quickly.
posted by Ampersand692 at 12:41 PM on February 4, 2020


I think the reason I am so fascinated by this history (and disappointed by the market theory argument) is that I believed the difference would be marginal, learned a new layout anyway, and increased my speed by more than 25%.

How "correct" of a typist were you on a standard QWERTY keyboard?

That is to say, I am a pretty fast typist, but because I am self-taught, my hands wander all over the keyboard and basically never hit the "home row".
For example, when typing this, I used my right hand to press "f" , my left hand to type "y" and both index fingers for the space bar.
My spouse on the other hand, who took typing while in school. is a much more fluid and "correct" typist and probably gains a measure of speed from that, regardless of layout.

I'd imagine that if I put in the practice to learn proper keyboard usage, I'd increase my speed accordingly.
I'm not sure I could break 30 years of bad habits on a QWERTY keyboard, but an alternative keyboard would force me to start over from scratch.

(As a side note, I corrected so many typos in this post. Paying attention to where your fingers go while typing really messes with the automatic part of your brain.
Kudos to you for managing to retrain yourself entirely!)
posted by madajb at 1:03 PM on February 4, 2020 [2 favorites]


Keeping people off my computer is a big plus. With the layout (and a script which locks my screen if someone still manages to type "free donuts") I'm the safest person in the office if I forget my screensaver.

Also, marginal improvements in hand and wrist pain are golden for me. Even if it was *slower* I'd still type in a layout with easier motions.
posted by Anonymous Function at 1:13 PM on February 4, 2020 [2 favorites]


How "correct" of a typist were you on a standard QWERTY keyboard?

About 93% correct? I mostly hit the 6 with my left forefinger, hit the 1 with the third finger of my left hand, and hit the 0 with the third finger of my right hand, not my pinky. 100% "correct" if you only count letters, though. I took a typing class in high school and hit about 65 WPM at the time, but hadn't been that fast in years. As a professional computer user my speed had gone down to about 45-55 WPM depending on whichever online test was making the rounds. I'm above 70 WPM now.
posted by fedward at 1:18 PM on February 4, 2020 [2 favorites]


I can touch type with five fingers and a thumb at 100 WPM. Whee. Since this metric is about how fast a person can transcribe something given to them, and so not exactly a useful metric for my actual work which involves writing while thinking, the real speed is 10-20 WPM as I sit and ponder, type a little, then sit and ponder some more. Then I start working on a graph or table, and at that point WPM is totally pointless.

The touch typing part is valuable though, and probably why I'll stick with QWERTY to the end.

Meanwhile, my handwriting has gotten slower and worse, and that's the real shame.
posted by linux at 2:16 PM on February 4, 2020 [3 favorites]


Great post! This has inspired me to try the Colmak layout (which comes with macOS and is supposed to be easy to learn as well as ergonomic). It's slow going so far because I don't have muscle memory yet but not nearly as daunting as I'd assumed it would be.
posted by treepour at 2:57 PM on February 4, 2020 [1 favorite]


If we're going to optimize the keyboard layout for speed, why settle for Dvorak? Let's have a computer crunch through 10,000 hours of simulated typing and produced some even weirder layout that will inconveniences qwerty and Dvorak users alike.
posted by Pyry at 3:08 PM on February 4, 2020 [2 favorites]


At this point I’d settle for just getting Mac and PC to standardize the copy and paste short cuts?
posted by interogative mood at 3:26 PM on February 4, 2020 [6 favorites]


Let's have a computer crunch through 10,000 hours of simulated typing

That's kind of what the "full optimization" layouts are on the Carpalx site. If you trust the effort model, and if you're willing to let the semicolon move, you end up with the simulations converging on:
Q G M L W B Y U V ;
D S T N R I A E O H
Z X C F J K P , . /
If you look at that and say "why not keep all four of ZXCV in place like Colemak does" the simulations converge on:
Q G M L W Y F U B ;
D S T N R I A E O H
Z X C V J K P , . /
You will note the similarities between these two. The biggest reason I ended up with the latter instead of Colemak is that it separates Q and W. In its desire to keep keyboard shortcuts, Colemak actually perpetuated a problem for me: I often meant to close a window (cmd-W) and accidentally ended up quitting the whole app instead (cmd-q via fat-finger).

A valid criticism of Carpalx is that it doesn't consider moving other punctuation characters. Arguably by doing so you could get something even more efficient, but counting punctuation as part of the corpus used for analysis increases the scale of the problem tremendously and probably doesn't increase efficiency enough to be worth the extra computational cost. And while professional software developers are probably a sizable percentage of the total number of people willing to consider switching layouts, non-letter character optimizations are highly dependent on the language and syntax of one's code (a Perl programmer's needs are different from a C programmer's needs, and both of those use more semicolons than Python, to name just three).
posted by fedward at 3:40 PM on February 4, 2020 [3 favorites]


Based on the "new research" link, which concludes that the best indicators of high performance are "1) unambiguous mapping (a letter is consistently pressed by the same finger), 2) active preparation of upcoming keystrokes, and 3) minimal global hand motion", I wonder if one of those hyperminimal, steno-like, chorded keyboard layouts would be the most efficient. I'd bet they're tops for those three criteria.

But could the mental overhead of remembering chords undercut some of the benefit of barely moving your fingers at all? Stenographers seem to make it work, but what if you're coming up with the text yourself?

In the meantime, I'll be trying out a split, ergonomic (Redox) adaptation of QWERTY and this was typed on an IBM Model M buckling spring beauty aww yiss...
posted by Grimp0teuthis at 3:49 PM on February 4, 2020 [1 favorite]


BTW if you didn't follow the link to Paul David's piece (PDF) in the stuff above the fold, he mentions an early competitor to QWERTY:
Freed from the legacy of typebars, commercially successful typewriters such as the Hammond and the Blickensderfer first sported a keyboard arrangement which was more sensible than QWERTY. The so-called "Ideal" keyboard placed the sequence DHIATENSOR in the home row, these being ten letters with which one may compose over 70 percent of the words in the English language.
Blickensderfer called it the scientific keyboard. If you rearrange DHIATENSOR you get the DSTNRIAOH home row of the two Carpalx optimized layouts above and the ARSTDHNEIO home row of Colemak. Dvorak has a U instead of an R, but 9/10 ain't bad.
posted by fedward at 3:51 PM on February 4, 2020 [1 favorite]


I switched to Colemak mid-last year and, after something like 3 weeks of muscle memory reworking, I got back up to normal speed. I'm honestly not sure that I ever got radically faster under Colemak (I don't remember what my typing speed was before I switched) but I will say that my hands feel more comfortable and less painful. I can distinctively notice the fact that my hands move from the home row less often and that translates into less wrist movement.
posted by Inkoate at 4:32 PM on February 4, 2020 [2 favorites]


I feel like so much of it now, at least, is that young brains are far more elastic than old brains, and if you learn to type as a young kid, no matter what the layout is, you'll never get as good as an adult with a new layout. That's probably a GenX-Millenial thing at this point (I'm at the tail-end of GenX and learned to type on Mavis Beacon; I think kids how just learn by typing on their phones endlessly). So, any new keyboard layout with have to be released for young kids only, then grow in market share with them. And, well, nobody in capitalism is that far-looking with the exception of Disney and nicotine-delivery device manufacturers.

At this point I’d settle for just getting Mac and PC to standardize the copy and paste short cuts?

As someone with a Mac at home and a PC at work, I think 90% of my typing slowdowns and mixing up CMD and ctrl. CMD is a bigger button (and closer to the center of the keyboard), so Windows should just adopt CMD. Also, I get a lot of utility out of the Windows key+typing, so Mac should adopt that key for Spotlight as well instead of the stupid CMD-space that works about 20% of the time.
posted by General Malaise at 4:46 PM on February 4, 2020 [3 favorites]


Also: my work keyboard has a weird left ctrl key (it seems to angle just to the left), which means that instead of typing Ctrl-V (pasting is the fastest way to type for me at this point in my life), I just hit v endlessly. So, instead of new layouts, maybe we all just need better keyboards.
posted by General Malaise at 4:48 PM on February 4, 2020


A friend of mine will only use Kinesis keyboards. The boards are stranger in person that pictures can do justice: huge deep bowls of keys. He likes 'em, and no-one steals 'em, so all good.
posted by scruss at 6:16 PM on February 4, 2020


Best April Fools joke I ever pulled was in college when I switched my roommate's keyboard layout to Dvorak when she was out. She was an English major, and early April is when all the final papers were due.

#sorrynotsorry
posted by basalganglia at 6:30 PM on February 4, 2020


One of my friends tried switching to Dvorak on his phone, and his typos went way up. It turns out that clustering the most common letters together makes the job of autocorrect all that much harder because there's less space between valid words than on QWERTY.
posted by ckape at 6:30 PM on February 4, 2020 [5 favorites]


At this point I’d settle for just getting Mac and PC to standardize the copy and paste short cuts?

Actually, most of the time this isn't an issue--it's comes down to command-versus-control. I'm even an EMACS user, so bouncing in and out of that adds a bit to the mix, but I manage.

What really frustrates me is laptop keyboards:
  • Do the function keys work as function keys, or as brightness/volume controls?* Or do you have a touch bar?
  • Are Home, End and the like their own keys, or cords? Where are they on your keyboard?
  • On a big enough laptop, do you have a number pad, or do is the main keyboard centered on the machine?
Every system seems different in these regards.

With respect to layouts: It seems to me that QWERTY benefited from a combination of getting out front early, then the Network Effect: the more people trained in that keyboard, the greater the demand. I dare say interest in alternate layouts is a bit of a twenty-first century problem, where the cost of a different layout by user is zero.

That said, I've yet to see any indication that any alternate layout provides significant benefit. I always regard it as somewhere between people who lecture about the efficiencies of a recumbent bicycle (usually while being dropped on a climb) and people who learn Klingon.

*Yes, I realize they do both, depending on if you hit Fn or the like, and, on most systems, you can select which is the corded versus non-corded behavior...assuming your company has the BIOS unlocked for you to do it.
posted by MrGuilt at 6:33 PM on February 4, 2020 [1 favorite]


METAFILTER: somewhere between people who lecture about the efficiencies of a recumbent bicycle (usually while being dropped on a climb) and people who learn Klingon.
posted by philip-random at 6:42 PM on February 4, 2020 [4 favorites]


I hope you realize I can't mark your post best answer if it's in the wrong thread.
posted by pwnguin at 12:09 AM on February 5, 2020 [2 favorites]


The real elephant in the room is whether all these newfinagled designs allow you to get biscuit crumbs out from between the keys more easily.
posted by scruss at 4:47 AM on February 5, 2020 [1 favorite]


A friend of mine will only use Kinesis keyboards.

I'm one of those people as well, and they really are the best.

I've been using a Kinesis as my primary keyboard since ~1998. I can still type fluently on a normal keyboard -- learning typing on actual typewriters in a class taught by a nun with a ruler apparently stuck. In a way, I think because I learned on a typewriter and also used standard keyboards and then later switched to a weird keyboard, I've got a pretty good knack for coming up to speed with e.g. a new laptop keyboard if needed, since I'm used to the typing equivalent of code-switching on a pretty regular basis already.

That said, I never bothered with an alternate layout. Don't get me wrong, I actually do have a few changes in my key settings (Kinesis keyboards are hardware reprogrammable, and there's a couple things I changed about certain control/meta keys that just makes more sense to me and is more comfortable). However, the reality is that I type (even on a normal keyboard) like 120+ WPM with QWERTY, which is about 5x faster than I really ever need to type when doing my job (software development).

Is it useful in short bursts, like when writing documentation or emailing? Sure, but in most cases whatever I'm doing requires enough stopping and thinking that input speed of new keystrokes is not the limiting factor. The argument for alternate layouts, for me, seems like a thing that only makes sense if your idea of who uses keyboards is limited to transcriptionists, which is not really a representative situation these days.
posted by tocts at 5:41 AM on February 5, 2020


I love QWERTY keyboards.

I learned touch-typing at nine from an aunt who taught high school business class and used an antique manual Royal through high school. As I typed my homework, my folks fielded calls from disbelieving teachers— neither parent used a typewriter, so they had to explain that there was no way they could have typed for me. At 12 I could do 120 wpm, no problem.

I burned through four Smith-Corona portable cartridge electric machines during college, as an English major who paid expenses by typing for others. After that I worked as a phototypesetter using many, many different keyboards. Never tried Dvorak but did try the so-called ergonomic layouts. Ugh.

The QWERTY layout is part of my brain and taking a keyboard out for a spin has always been a pleasure. The finest iteration was for something called the Quadex system, a high-end 1980s phototypesetter with a keyboard that had an incomparable touch. Just enough springy give and bounce, with a satisfying click when a keystroke registered. Lightning-fast monitor screen, too. Shame it was all in code with no preview. Absolutely gorgeous machine with the Rolls Royce of keyboards. I enjoyed every minute spent with my fingertips on those keys. *sigh* It was like playing a Steinway grand piano (yes, I also played piano).

I switched my roommate’s keyboard layout to Dvorak when she was out. She was an English major, and April is when all the final papers were due.

Had my roommate done this, I would surely have gone to prison for homicide. This is pure evil... like disarranging the keys on a piano. The stuff of nightmares.
posted by kinnakeet at 6:06 AM on February 5, 2020 [6 favorites]


In my eyes, DVORAK fails for a few reasons yet mentioned. First, there is no DVORAK in a row on the keyboard. Already that's enough to make the whole thing a joke in my eyes. Second, that name itself is horrible. D'oh-vore-ack. Homer and Cathy aside, it kind of sounds like a corny fantasy/scifi version of "divorce" or maybe just a reptilian alien who goes by name D'vorak.

The most pathetic thing about DVORAK, though, is their version of "WASD." It's shit! "<AOE!?" What kind of crap is that? WASD rolls off the tongue and through some serendipity the letters also relate to the directional action.
posted by GoblinHoney at 8:50 AM on February 5, 2020


I tried Dvorak out in college since I had plenty of time to retrain my muscle memory. It took time, especially given that laptop keyboards usually don’t permit easy rearranging of keys, but I stuck it out and found... that my typing speed didn’t really change.

I stuck it out though and still use Dvorak (to the chagrin of anyone who takes over my keyboard) for one reason only: Like Inkoate, I found that typing on QWERTY stressed my hands out more than Dvorak did (that top-row ERT!).

One thing I can’t do is speak to the relative virtues of Colemak vs. Dvorak though, and I’ve never bothered to try it out. I wonder if anyone has tried both and can speak to that? I know I came across Colemak at the time I was switching, but the OS support wasn’t there (except on Linux) back in 2006, so I went Dvorak instead.
posted by The Situationist Room with Guy Debord at 11:30 AM on February 5, 2020


I wonder how different a keyboard optimized for transcription might be from one optimized for for writing or coding?
posted by sfred at 12:12 PM on February 5, 2020


I'm always interested in things like the optimal typing system or the most ergonomic way to sit but then I immediately go back to lying on the chaise lounge with my laptop propped up awkwardly with my elbows out (so that it's close enough to see the screen). Luckily I don't type a ton so I can continue to get away with it.
posted by Margalo Epps at 12:24 PM on February 5, 2020


I wonder how different a keyboard optimized for transcription might be from one optimized for for writing or coding?

My gut call is very; programmers would use much more punctuation--depending on the language--semicolons, colons, brackets (square and curly), equals, plus/minus, asterisks, less- and greater-than, pipe (for shell commanders), etc. Plus word/letter frequency would probably be a little different (looking at it as a dialect (?) of, mainly, English).

IOW, as someone mentioned about, there's probably not a "best" keyboard; maybe a decent enough general keyboard, but depending on the purpose, I could see a more optimized layout based on the specific purpose.
posted by MikeKD at 1:14 PM on February 5, 2020 [1 favorite]


I can tell you what a keyboard that's optimized against coding looks like: a Macbook with a touch bar. I say this as someone who has to deal with it on a daily basis (thankfully with my external Kinesis keyboard most of the time).

I love a lot about my dev system, but whoever decided function keys (and even worse, the escape key) could be relegated to non-tactile fake touchscreen buttons that can move around depending on the app clearly hates people who actually write code for a living and type without looking at their hands.
posted by tocts at 1:23 PM on February 5, 2020 [5 favorites]


I'm not even super mad about the loss of the esc and function keys, but I hate that if I do not hit the - button square on, I'm suddenly trying to take a screenshot. The touchbar is awful. /derail
posted by snwod at 1:26 PM on February 5, 2020


I also wonder about the age thing. How many of the younger generation had some phone or computer through childhood and learned to sorta-type *before / if even* took a serious mid/high school semester long every day of the school week typing class.

That's my dad in the early 80's going "computers are going to be a big deal" and I should take a typing class even though it sorta seemed like a girly thing for future secretaries or such. (There were only a couple of boys in that class, might have well have been cosmetology...) I guess I sorta think the speed or efficiency or lack of RSI or whatnot is much more tied to practice and learning it well and proper in the first place rather than a slow and incremental self-taught half good half bad way.

Anyway, I ended up going to the state Future Business Leaders of America competitions and placed 4th. I can type pretty fast if I put my ind into it or even better if I don't put m mind to it. Or I can just think and type while watching TV anot evel looking at the words pop up on the screen because my fingers are just connected to my brain in some weird way.

Not sure how the teaching typing programs and such stack up to the old flip top typing textbooks that taught a bazillion secretaries to just bash out memos or take live transcriptions. I doubt there's much to be had in the different keyboard layouts trying to eek out an extra 10WPM.

Interesting post.
posted by zengargoyle at 2:53 PM on February 5, 2020


That's my dad in the early 80's going "computers are going to be a big deal" and I should take a typing class even though it sorta seemed like a girly thing for future secretaries or such. (There were only a couple of boys in that class, might have well have been cosmetology...)

I guess something to be said about male privilege — I didn't even think about it until now, but my "keyboarding" class (renamed from typing on account of the Apple IIgs we were using) was 98% girls (seventh grade) and the 2% of it the boys were just computer nerds who wanted to play with computers for the hour.

Also, just wanted to add that character layouts aside: The quotation mark being on the 2, instead of to the right of the L key, as on the C64 and other 8-bit computers of its day, is far superior to where it landed on PC keyboards since.
posted by General Malaise at 5:18 PM on February 5, 2020


> I doubt there's much to be had in the different keyboard layouts trying to eek out an extra 10WPM.

The way I think about it, I have a long career ahead of me in typing. And when I'm done with my day job of typing on keyboards, my hobbies include pressing buttons on controllers. Since Siri dictation isn't going to be banging out bash scripts and Python any time soon, and the neural links facebook is developing behind close doors are presumably going towards predicting ad clicks not language. So like, eeking out a little bit of speed isn't gonna make me more productive, but not getting carpal tunnel or arthritis sounds like a great long term goal.

And like, if all you are interested in speed, DVORAK is going to lose to stenography all day erry day.
posted by pwnguin at 6:23 PM on February 5, 2020 [1 favorite]


I learned Colemak a couple of years ago. Like many of us here, I type for a living. I found myself frequently suffering from two problems: right wrist pain, and extreme tension in my upper back. (The wrist pain isn't RSI-related; I shattered my wrist about 15 years ago and it's never been quite right since, often acts up if I over-work it, or if the phase of the moon is wrong.)

Seeking some solutions, I decided to give an alternate keyboard layout a try. I'd attempted to pick up Dvorak at one point in the past, but found the relocation of common short-cuts and punctuation to be frustrating. Casting about for alternatives I discovered Colemak, and put some effort towards learning that.

I couldn't go cold turkey, because my productivity would have suffered too much, but I switched back and forth between QWERTY and Colemak over a period of 2 or 3 months or so and by the end of that was able to work a full day on Colemak and not feel like I'd lost time to it.

Since then I've noticed that, while I'm not typing notably faster, I'm definitely not any slower using Colemak, and more importantly I don't find it nearly as tiring. I can go for much longer sessions before my wrist begins to hurt me.

At the same time, I began experimenting with different physical keyboard formats to see if I could do something about the back tension. After playing with ortholinears and small-format boards I gave split keyboards a try... and found them magical. I now use one of these (except not RGB) at home and at work, along with their tilt/tent kit and wrist rest.

With the keyboard halves about 8 to 10 inches apart, I find my typing posture to be much more neutral and natural, with no scrunching of my back, it's more like I'm sitting in a comfy arm chair. I can also move the board halves closer together or farther apart, or rotate them in and out, to shift my position up a bit and ease fatigue over the course of a day. It has made a phenomenal change in the amount of tension I carry in my back, and I've found myself much less in need of routine visits to a massage therapist to work all the kinks out.

The physical change has definitely had more of an effect than the layout switch, but I think the two combined have really reduced the overall discomfort of doing what I do all day. I'd definitely encourage anyone who was interested to try. You don't need to go cold-turkey if you're willing to stretch your transition period out a bit, so your productivity doesn't need to suffer too much.

(And, definitely, if you can, do your layouts in keyboard, not in the OS. I have one layer on my Kinesis for Colemak, and one for QWERTY, and I can switch with a button press if I need to hunt-and-peck a tricky password or similar without needing to worry about OS configuration.)
posted by jammer at 7:32 PM on February 5, 2020


Second, that name itself is horrible. D'oh-vore-ack. Homer and Cathy aside, it kind of sounds like a corny fantasy/scifi version of "divorce" or maybe just a reptilian alien who goes by name D'vorak.

Can we not make fun of actual people’s names, please, even in jest? Please see the site guidelines, which say “If you're an American or an extremely-online native English speaker, allow for the possibility that other people are coming from a very different cultural or linguistic context. Be kind.“ Names are specifically mentioned on the microaggressions page.
posted by fedward at 5:27 AM on February 6, 2020 [3 favorites]


 The quotation mark being on the 2

That's a UK thing: I got so much wrong when I moved from UK→Canada. You probably wouldn't like what the UK keyboard does with #, though. The Canadian French keyboard has it too, but it's sweary-inducing if you're not used to its quirks.

Can I just say it gives me mild joy to see people in North America ranting on the Raspberry Pi forum that the (UK maintained) Raspbian system software defaults to a UK keyboard and A4 paper? Welcome to the rest of the world, folks! Your @feelings@ are noted, or at least will be as soon as we PC LOAD LETTER
posted by scruss at 7:18 AM on February 6, 2020 [1 favorite]


I guess something to be said about male privilege — I didn't even think about it until now, but my "keyboarding" class (renamed from typing on account of the Apple IIgs we were using) was 98% girls (seventh grade) and the 2% of it the boys were just computer nerds who wanted to play with computers for the hour.

I think I may have mentioned this here before, but my mom insisted that I take a typing class before I got out of high school. I no longer remember what the consequences were going to be if I didn't, but this meant that my senior year I had to figure out how to fit a typing class into my schedule, which had a few classes only offered in one time slot per day. The worst example of this was that there were seven National Merit Scholars (7!) who didn't take calculus our senior year because we wanted to be in chorus and they were both only offered fourth period. My school offered two different sorts of typing class, one called "Typing" that ran a full year, and one called "Personal Typing" that was only one semester. "Typing" seemed to be more or less a vocational training type of thing, and I think it got into dictation and maybe some other office skills. For all I know it may have been 95% girls. "Personal Typing," on the other hand, had a much higher number of boys in it because it was filled with sophomores who needed something in their schedules the semester they weren't taking driver's education.

"Personal Typing" was the typing class that fit into my schedule. Since most of the classroom was filled with kids who really didn't want to be there, the teacher loved me.
posted by fedward at 8:25 AM on February 6, 2020 [1 favorite]


The finest iteration was for something called the Quadex system, a high-end 1980s phototypesetter with a keyboard that had an incomparable touch. Just enough springy give and bounce, with a satisfying click when a keystroke registered.

Have you asked about that on keyboard forums? I'd bet something like that would have used commodity parts, so you might be able to get a new mechanical keyboard with a really similar feel, if not exactly the same type of switches.
posted by fedward at 8:53 AM on February 6, 2020 [1 favorite]


« Older Images of America   |   Carbon dating counterfeit whisky Newer »


This thread has been archived and is closed to new comments