• You are not logged in.
  • Index
  • General
  • What makes Colemak better than the alternatives?

What makes Colemak better than the alternatives?

  • Started by luke
  • 32 Replies:
  • Reputation: 214
  • From: Viken, Norway
  • Registered: 13-Dec-2006
  • Posts: 5,361

Yep, I was stuck at around 55-60 WPM for a long time. Then I started the TypingMaster course which isn't optimized for Colemak at all but rather nicely laid out which made it pleasurable, and in a short time I went to 65 WPM; I attribute this mostly to better flow and a lower error rate. My record at higames is now 70 WPM although I cannot readily reproduce it.  :)  This is much better than my 50 or so QWERTY WPM and 55 or so Dvorak WPM of the past, but of course it's impossible to judge how much of that stems from the layout and how much stems simply from my work with it. Typing progress is somewhat similar to what I've experienced learning the piano: Long stretches of little progress and then a leap forward as something new "clicks into place".

Jammycakes, I think you're right... in part. Whether you are a good judge of a layout in a month would be very individual I think. A typing master going from 150 WPM and having tried both QWERTY and, say, Dvorak, could probably assess any layout in a week or two, whereas a non-accomplished typist well could need several months to form a really balanced opinion.

Last edited by DreymaR (23-Dec-2008 19:29:50)

*** Learn Colemak in 2–5 steps with Tarmak! ***
*** Check out my Big Bag of Keyboard Tricks for Win/Linux/TMK... ***

Offline
  • 0
  • Reputation: 0
  • Registered: 14-Nov-2008
  • Posts: 11

When I switched to dvorak about 2 years ago, I had read a bunch of fluff about getting up to speed in a matter of weeks etc...  It took me over a year to reach my qwerty speed.  During that year, my understanding of the dvorak layout changed.  Once you reach a certain level, you begin to understand the limitations/strength of a given layout.  For me, that understanding came when I had completely forgotten qwerty and reached about 65-70wpm on dvorak.  The fact is, you can't judge a layout too early on because you're constantly fighting your memory muscle. 

As for the stationary XCV on the new layout...  I have no need of making a layout easier to learn by keeping some things the same.  I have completely retrained my mind once already, I can do it again.  In fact, having them be in the qwerty positions is causing me to relearn again because my mind has been branded with the dvorak seal for the past 2 years.  They just happen to work well in those positions for this layout.  Although I did end up moving V elsewhere - I recently shuffled Q, K, V, -, "   I'm quite pleased with the core of the layout so far.

As for the digraphs PL, OL, LO, PH, HO...  For me, PL is actually very comfortable and fast to type.  The most annoying one I think is OL because the index finger is a bit short, so (at least for me) it forces you to shift your whole hand up.  At first I thought HO was going to be a problem, but I can now type it very fast and it doesn't bother me at all.  I'm definitely enjoying the layout so far - the flow is really good and I feel like I'm exerting very littly effort when typing.  However, I won't know the layout's real limitations until I reach that magic threshold - at which point I hope it still continues to sing. 

Here's my current modification:

Q B U L P K M Y F V Z = |
R I A O H D T E N S ?
  : X C _ J G W < > "

For anyone curious about the dash/underscore just below O, I'm a software developer (C/C++), and I type a whole lot of '->' sequences to dereference my pointers.  I also use VI a lot, so having the colon nearby is very useful.  L and H are backwards for VI, but I can live with that since I mostly use B & W for left/rigt navigation.

Offline
  • 0
  • Reputation: 0
  • Registered: 17-Dec-2008
  • Posts: 59

Okay, I can roll with this.

I've used QWERTY for 16 years.  Sequences like "SWEATERDRESSES" and "FASTED FADS" and "RECEDED" fell smooth and natural to me (no joke).  A few months ago (before tinkering with multiple layouts), I could average 105 WPM on hi-games and 120 WPM on keybr's random sentences, no warm-up, no problem.  While not world class, it'd be an understatement to say I just "got by."  I could have been perfectly content typing QWERTY for the rest of my life, dismissing those who complained of awkwardness or pain as lazy, inexperienced, poor learners (if they stuck with QWERTY anyway), rebels without a clue (if they switched), too old, etc.  (And I did for most of that 16 years.)

Clearly, since I have *16* years of experience with QWERTY--far more than anyone could possibly have with Colemak (and some of them can have opinions about that layout!)--that means I'm right!  QWERTY must be smooth and comfortable to type on, because it is for me and I have so much experience to back me up!  If you have a comparable length of experience and disagree, that just means you practiced wrong, or are rationalizing, or for some other reason aren't as good as me.  :P

Or MAYBE, maybe it's that experience actually HIDES flaws from us!  Maybe we get so used to those flaws and working around them that we become blind to them!  Maybe our first impression is MORE accurate than our 365th impression because it's pure--untainted by habit (and other, less flattering mechanisms).  Stuff like "you did not spend sufficient time to allow your motor skills and muscles to get accustomed" reads exactly to me like "just get used to the awkwardness and you won't notice it anymore!"  Yeah, no kidding, I did exactly that with QWERTY.  Do you really think that means the awkwardness doesn't exist?

Ever been writing the same paper for days and days, see it as grammatically flawless, and then hand it to someone to proofread and they find something like a repeated word in in the first sentence?  (Ahem.)  Same phenomenon.  I'm sure the C and Java coders among us have stared at the same piece of buggy code for hours, seeing nothing but flawless code, yet when we hand the code to someone else, it turns out the bug was a comma instead of a semicolon, or missing braces on an if-statement (which was "properly indented"), or something of that nature.  Same phenomenon.  Ever see one of those parents in the supermarket with an unbearable child, yet the parent seems completely oblivious to it?  Same thing.  Ever hear a crappy song on the radio that you found yourself humming after the 50th time you heard it?  Same.  Was there ever some food or beverage you hated at first, but were forced (or forced yourself) to eat or drink, and you eventually developed a taste for it?  Same.  Been stuck some place with a horrible smell and eventually be unable to smell it?  Same.  Guess why companies spend millions of dollars to keep telling you day after day about their products.  I mean, you heard them the first time, right?

Familiarity breeds acceptance.

Just something to think about.

EDIT:  I acknowledge that further experience can REFINE one's opinions (e.g., adding new insights, exceptions to rules, etc.).  What I do not accept is that one's first impression OF FLAWS is always wrong because those flaws tend to "disappear" with experience.  As a rule of thumb, flaws ALWAYS tend to "disappear" with experience, because we get used to them and learn to ignore them (or, in some cases, even admire them).

Last edited by Phynnboi (24-Dec-2008 07:49:38)
Offline
  • 0
  • Reputation: 214
  • From: Viken, Norway
  • Registered: 13-Dec-2006
  • Posts: 5,361

Good points there.

Indeed, too much familiarity can mask flaws and prevent a fresh approach. I often see this in software: Some functions can be very unintuitive and user-unfriendly but the old users don't mind anymore. In that case, it takes a new user to spot the flaw.

However, it also takes a new user who knows what to look for. As I mentioned above, a typing expert might be able to assess layouts very speedily in comparison with others. With your 100+ WPM you're definitely in the upper segment of typists. Maybe your experience cannot be compared to, say, mine - which took a number of months to mature (then again, I had to unlearn Dvorak which is a harder transition to Colemak; then again again, I had experience with learning another layout fully before switching...). It's not easy to answer summarily.

But you have given the impression that you did something too hastily and/or improperly - both with your guitar learning and your keyboard learning. It's not only how long you stick with it, but the approach.

And of course, your massive experience with QWERTY will also mean that you need to work past a LOT of muscle memory before you can say that you've 'learnt' a new layout. Tricky.

*** Learn Colemak in 2–5 steps with Tarmak! ***
*** Check out my Big Bag of Keyboard Tricks for Win/Linux/TMK... ***

Offline
  • 0
  • Reputation: 0
  • Registered: 17-Mar-2008
  • Posts: 192

I hadn't really thought of it that way, but you make a good point. I know practice makes perfect, I never thought practice makes blind... However, while agreeing, I would like to temper what you are saying a little.

Some things that feel uncomfortable may not actually be flaws. For someone new to touch typing everything will feel uncomfortable. Eventually people get used to it and it's now comfortable, but that doesn't mean that every movement is flawed. If a layout contains technique that you are not used to, such as a heavier pinky load, that may or may not be an actual flaw. Note: May or may not be; I believe Shai for instance thinks Dvorak loads the pinkies too much. Other people have no problems with it.

It's a bit like your taste example; If you acquire a taste and now like it, does the food taste bad except you are now oblivious to it? I would say not; You got used to it and now the food tastes good. Other situations will be more analogous to the screaming baby example - they truly are flaws that you have now tuned out.

It follows that when you pick up a new layout, you will naturally be using the techniques you are used to from your qwerty typing. That means you are implicitly comparing the layout to "difference to qwerty technique". Flaws in the layout that are the same as for qwerty will not stick out as badly as the ones that aren't present in qwerty.

As I've said many times, it's still perfectly legit to quickly evaluate which layout suits you best. So what if it happens to be the one that allows you to employ your qwerty technique instead of starting over? But if you want to opine on flaws, you will be doing so from a precariously subjective position unless you give each layout more time.

One last question. If qwerty is perfectly comfortable, why bother switching? You probably won't beat the speed of qwerty, and you can't beat perfectly comfortable in the comfort department either. After all, the proof is in the pudding - perhaps for you qwerty is a good layout.

Last edited by tomlu (24-Dec-2008 11:21:36)
Offline
  • 0
  • Reputation: 0
  • Registered: 14-Nov-2008
  • Posts: 11

@Phynnboi:  I agree with you that practice can (and will) mask flaws.  I was merely stating that some flaws are not readily discoverable until you reach a certain speed.

Offline
  • 0
  • Reputation: 1
  • Registered: 22-Aug-2008
  • Posts: 36

Phynnboi already said he thinks Colemak is better than QWERTY. So if the "Colemak hype" makes more people switch to a better layout that's a good thing. And that was Shai's goal after all.

Phynnboi, placing not-so-common letters on the home row under the pinkies, as you do in your "Colemak competitor", increases same-finger digraphs (you have oi, ge, ce, ec). So you too don't mind same-finger digraphs that much? Because if you want to reduce them people really need to train their pinkies and use them even though it may feel awkward. Shai thought same-finger digraphs slowed him down the most.

bsdhacker, maybe you don't mind, but placing the apostrophe there makes n't and 's not very comfortable to type. I think the semicolon needs to be on the left (as it is in your layout), because for programmers the semicolon is usually followed by Enter.

Offline
  • 0
  • Reputation: 0
  • Registered: 17-Dec-2008
  • Posts: 59

I'm strapped for time, so I'll address other points later, but I wanted to respond to vVv's question since it's something I've been thinking about, lately.

Yes, less common characters on the pinkies does increase same-finger.  There seems to be an unavoidable trade-off there--either you can have low pinky use or low same-finger, but not both.  (Makes sense, I guess--less pinky use means cramming more common characters into a smaller area.)  I've very recently had a change of heart on the issue, and have retooled my algorithm to match.

Here's an interesting potential competitor that popped out of my program:

qygb m   j pdu;
oesn r   l thai
zxcv f   k w,./

It has the regrettable property of the middle fingers being underused and pinkies being overused (IMO), as well as some aesthetic defects (such as the positions of E and R).  However, when I test drove it, it actually felt pretty good.  It has a bit of an "airy" feel to it, probably due to all the pinky/index pairs.

I'll post if anything better pops out.

Offline
  • 0
  • Index
  • General
  • What makes Colemak better than the alternatives?