Hacker News Reader

"Unless you get everyone to switch to it, you will be disabled whenever you switch to someone else's computer, a kiosk, etc. Why bother?"

I take it you have not tried any alternate layouts. Why do you as someone who has not tried it feel inclined to lecture about the consequences to those who have? Which of us do you think has the clearer picture about the consequences?

Theorize about the potential downsides all you like. In practice, I find it has been a net gain for me personally, and no amount of mere conjecture on your part will change that one bit.

Agreed. I've been using dvorak for 11 years and when I read "Unless you get everyone to switch to it, you will be disabled whenever you switch to someone else's computer, a kiosk, etc." I knew that this person wasn't speaking from personal experience but from imagining what it would be like if he/she switched.

I've typed Dvorak for a couple years, and have been quite happy with it. I'm curious about Colemak, though. Dvorak is pretty established as the alternative English-language keyboard layout - It's not hard to set up on Windows, Unix, or (I'm guessing) OS X. Is Colemak? While I'm convinced that Dvorak is significantly better than Qwerty, if you're the stubborn sort that would use a different keyboard layout for English (spoiler alert: I am), I suspect there's diminishing returns after that. Still, I'm really curious about Colemak (what was the rationale?) - I just don't know anybody who uses it.

My strategy of "heavily customized Emacs / Dvorak on my computers, standard Qwerty / vi otherwise" has been a good compromise over the years, incidentally.

Also, I work with Swedish hackers who remind me that Qwerty is only "standard" so far across the ocean. These things only make sense in a local context.

I use Colemak. It's supposed to be easier to learn than Dvorak, and it leaves some keyboard shortcuts intact (e.g. ctrl+zxcv). There is a colemak.vim file that completely customizes vim for colemak. Warning: it changes nearly every single key binding.

The only downside to colemak is that d is next to t. This sometimes makes it difficult to type Dutch, because some words end with "dt". A typo "td" is easy to make.

Funny that they're Swedish - Swedish is one of the few non-English languages for which there is a version of Dvorak - it's called Svorak.

Yeah, I think there's a Danish Dvorak as well. Pretty useless for Japanese, Russian, or Arabic though. :)

How often do you work on someone else's computer? I almost never do. Kiosks (I presume you mean touchscreens like at airports, which are about the only place I use them) are similarly rare, and need little to no substantial text entry.

But for those times that I do need to use other people's computers (most usually, when at an internet cafe while on vacation, when I don't have my laptop with me, and I don't have internet connectivity with my phone, all increasingly rare times), I have a little Windows app I wrote that I can download and run from the web which uses a low-level keyboard hook to translate Qwerty to Dvorak. Installing a keyboard hook[1] on Windows doesn't require any privileges, and runs fine in user mode until the application is shut down.

[1] http://msdn.microsoft.com/en-us/library/ms644985%28v=VS.85%2...

I switched to Colemak, and I am also forced to use qwerty on a lot of keyboards. Colemak is very similar to qwerty, so its not that big a switch. Mostly the r/s location is the biggest issue.

Besides that, it's worth mentioning that there is an executable that you can use for Windows machines to easily switch to Colemak without installing it on the box. I carry it around on a thumbdrive, place it on a network share, and keep it in my Dropbox. I can always get to it. So 80-90% of the time I can use Colemak, and I am reasonably fast at qwerty when I can't.

The reason I switched (like the OP, if you read it again) was to reduce RSI pain/fatique. It's worked wonders for me in that respect. I'm not that much faster, so speed's not a major incentive to switch, IMO.

> it's worth mentioning that there is an executable that you can use for Windows machines to easily switch to Colemak without installing it on the box. I carry it around on a thumbdrive, place it on a network share, and keep it in my Dropbox. I can always get to it. So 80-90% of the time I can use Colemak, and I am reasonably fast at qwerty.

Can you link to a copy of this executable? Is it just an AutoHotkey script?

Yeah, it's a portable keyboard layout in autohotkey. Here's the link (you can get to it from colemak.com):

http://sourceforge.net/projects/pkl/files/Standalone%20Layou...

Similarly for me.

Also, I didn't think Colemak would actually let me type faster, but I did a speed test and scored 102wpm with 0 errors. I used to type 80.

It may not directly be Colemak. I may just type more because Colemak is that much easier to type. Because I type more, I learn to type faster.

> Interesting, but about as useless as Dvorak. Unless you get everyone to switch to it, you will be disabled whenever you switch to someone else's computer, a kiosk, etc.

Not true. I learned Dvorak a few years ago, and I can still sit at a Qwerty keyboard and type reasonably fast. It's really not that difficult to go back and forth.

Yeah, the brain has a built in ability to identify context and fit the expression in those parameters. It uses the pre-frontal cortex. This is the same 'functionality' that allows bilingual people to identify which language is being spoken and respond in that one.

People don't trust themselves. You can learn to switch back and forth the same way you learn to talk in different style to your grandparents than your friends, and so on.

What I do is touch type with dvorak (max about 90 wpm) and look at my hands on those rare instances when I'm using qwerty (probably about 20-30 wpm).

If I'm already over 90 on qwerty, is it worth switching to something else? Just do it to ward off CTS?

I really enjoy dvorak and I'm glad I switched but I can't say that the difference is enough to justify the pain of switching. It took me a long time to get used to dvorak. The first few weeks were very unproductive and it took me many months before it felt natural and I was faster than before. But I find it really comfortable since about 80% of the letters I type are on the home row. The benefits of dvorak are less when you're programming because there's a lot more punctuation and dvorak is mainly optimized for English words.

I'm not sure of the relationship between dvorak and repetitive stress injuries. Your fingers travel a lot less with dvorak, so they are doing less work. But I'm not sure if it makes any real difference.

> Unless you get everyone to switch to it, you will be disabled whenever you switch to someone else's computer, a kiosk, etc. Why bother?

I already feel disabled when not using [my favorite brand] keyboard, regardless the layout.

Hi, Chetan here. First off, thank you for your opinions.

About Colemak, I think a bunch of the replies to your posts hit the nail on the head. It's been a really rewarding switch, and I can still type QWERTY fine when I have to.

As for biphasic sleep, if you do a bit of research on it, you'll find that it's even more natural than a monophasic sleep cycle (think about why the siesta is so popular in other countries). It's just not as convenient for the average American, since everyone doesn't follow it. That being said, if you have some flexibility with your schedule and can implement it successfully, it can turn out to be better for your health and time than a monophasic sleep cycle.

And you're right about the 30/30 cycle being hard to use in a professional setting. That's why I would use it for personal projects, or when I'm working at home. It's just another tool for your productivity toolbox; you don't have to use it in every situation for it to be useful.

Right. The argument is that we force ourselves into monophasic sleep, which is unnatural and inefficient.

"Why We Nap" (ISBN 0817634622) is a great resource about polyphasic sleep. Most of us are in the habit of monophasic sleep, even if we have crazy or limited sleep schedules. The thing is, if your body goes through more than one sleep cycle in a period of sleep, then there is a period in the middle where you are basically not getting anything from the sleep. Biphasic sleeping is a way to reset the cycle more quickly, so instead of 2 hours of low value sleep, you just wake up for a half hour or so.

Personally, I've found that another important thing is simply sleeping consistently. That means either certain times of the day or for certain amounts of time. For example, always sleeping in three hours means your body knows what to expect and makes the most out of the time. Biphasic is just the easiest (and perhaps most natural) way to do it.

I found this page convincing that polyphasic sleep won't be optimal for performance in practice, despite its obvious appeal

http://www.supermemo.com/articles/polyphasic.htm

""" Healthy body clock runs a 24 hour cycle. This cycle will make you sleepy during the subjective night (which can be midday too). This is why you won't be able to wake up from your nap in your subjective night without an alarm clock. Alarm clocks are unhealthy. They prevent sleep from fulfilling its function. The choice is yours: either (1) sleep polyphasically or (2) sleep naturally and let your brain develop its full intellectual potential. """

Here is a rebuttal to that article by a prolific polyphasic sleeper:

http://www.puredoxyk.com/index.php/2006/11/01/an-attack-on-p...

Thanks for the pointer. I found a follow up from the same author I posted:

http://www.supermemo.com/articles/polyphasic2010.htm

he addresses some of the critiques of his original article.

I switch between US-English and Traditional Chinese input all the time, sometimes mid-sentence. Surely going from QWERTY to Colemak can't be that bad.

http://en.wikipedia.org/wiki/File:Keyboard_layout_Chinese_Tr...