Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - iandoug

Pages: [1] 2 3 4 ... 16
Keyboards and Other Interfaces / KLE renders
« on: Yesterday at 02:40:11 PM »
BTW has been updated, should be able to handle full width of your matrix layouts now. Also other changes.

I'll need to modify my program that generates KLE from KLA to make matrix full width. At present I skip a bunch of keys in the middle.

Sample render attached.

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: Yesterday at 02:20:53 PM »
Study says more like left/right hand = 0.8887, or about 47/53. Compare finger usage of BEAKL9 is 38.7/43 = 0.9.

Index is slightly stronger and agile than middle. So we give both a high share of keys pressed. But index is also shorter, so needs more effort to reach distant keys. So distant penalty for index is higher.

Curious how extra line appeared as I quoted. :-)

We've actually got a lot of layouts that are in the 45/55 percentage balance range.

Re index and reach, it's one of the reasons why I think key to "inside" of index should be part of Home Block/whatever ... it is easy to reach . I think H from J is easier than M from J, on ANSI QWERTY. Going to M requires an awkward hand shift.

I saw some research the other night that concluded that the left hand index is stronger for gripping than right hand index. (IIRC... was something like that.) Found that rather odd.
Really struggling to find research that measures relative finger strength for pushing buttons ... they're all focused on "pinch" and different hand grips etc.
Would also like relative flexibility, and endurance... I basically search these things on my phone when I go to bed... tires me out :-)

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: Yesterday at 03:38:57 AM »
Left is weaker and slower. Wants to stay in place. So more penalty for distance. Can't handle too many letters. Finger usage should be less, and higher penalty for same finger.

so that puts us back with vowels and punc on left .... been there, done that :-)

or do you want to overload right with etanoisrh ? :-)... put zxkqj etc on left?

Feels like you are saying hand balance should not be 50/50 ...

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: March 20, 2018, 03:52:40 AM »
Uploaded BEAKL 11 to 15 on KLA test and 3, under BEAKL section when selecting layout.

Thanks. Still need to run Den3 with previous collection, will add these and BEAKL MU to next batch.

Staring at .klc and PKL ... will see if can get at least ANSI layouts to work. Will need to find samples of other form factors.

Cheers, Ian

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: March 20, 2018, 02:12:39 AM »
Hi there. I want to try and learn Beakl. Do you have any quick pointers on how to use it in Windows?

Windows uses some sort of keymap file, yes? Mmm .klc files?

I should generate those for the layouts... will see what I can do. Also xkb files....

Cheers, Ian

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: March 20, 2018, 02:08:28 AM »
The downside of concave wells is not as effective for less-fingers typing (e.g. 6-finger style), versus flat keyboards. Thus concave keyboards fitted for 10-finger typing limit freedom of motions. ( and limits even creative optimizations of bigrams.

Yeah, on the one hand I see the design rationale for concave well, on the other I know they ain't gonna work for my non-standard hands.
Also I'm sure there are issues between big hands and smaller hands.

You got KLA json for latest version please?

Thanks, Ian

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: March 18, 2018, 07:40:08 AM »
Browsing around the book.


Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: March 18, 2018, 06:27:41 AM »
Attached page from Barnes' book.

Browsing around the book.

Has a guideline (these things are aimed at efficiency on production line etc, but some things still relevant somewhat to keyboard analysis):

"Smooth continuous motions of the hands are preferable to zigzag motions or straight-line motions involving sudden and sharp changes in direction" ...

which kinda sums up the problem with typing ... we're continuously changing direction.

So how do we arrange keys to be in circular motion rather than zig-zag....

Suppose this ties in with inward and outward rolls to a degree.

Modelling typing can get very tricky :-) ... suppose that's why Patrick took a simplified model.

Keyboards and Other Interfaces / Re: Finger strengths
« on: March 18, 2018, 04:57:04 AM »
4. time to fatigue. Perhaps some fingers can work longer without tiring than others?

A few months back I was thinking about this in relation to same-finger usage... and wondering about modelling it.

A finger (well, the muscles in the hand and forearm that work it) will eventually get tired after repetitive action.

And will recover after some rest.

Question is, is this actually a factor in real-world typing? Probably most people don't type continuously these days, unlike secretaries/data-capture people in times gone by.

We could pick some number (x) and say that for every keystroke typed continuously by same finger, finger loses 'x %' energy. And for every keystroke where finger is resting, it gains 'y %' energy.
x may or may not be equal to y. I suspect y < x.

My thinking was along the lines of say we use pinky twice in a row (ie loses 2x %), then it rests for 1 key (recovers y%), then gets used again twice in a row, etc.
Do this for all fingers, probably with different x and y for each finger.

These numbers then directly impact "effort" calculations... tired fingers need more effort to do same job.

Yes I know this is probably overkill :-)

Just putting it out there in case it gets someone else thinking...

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: March 17, 2018, 04:59:55 AM »
Finger typing and arm typing certainly employ different dynamics. Yet there could be similarities for the sake of measurement. So we must reason which metrics are important for either typing style, and how these metrics are same or different.

Attached page from Barnes' book. I found it very interesting that operator A2 was able to move the object 8 and 16 inches in effectively the same time... and even faster for the longer distance.
Comparing those numbers to his 24 inch time, and to both operators' "loaded" time, I'm guessing two things:

1. difference in body type, A2 was more slightly built and used forearm+wrist for 8 and 16 inches, but had to use shoulder as well for 24 inches. A1 was heavier-built, less flexibility, and had to use shoulder for 16 and 24 inches.

2. adding weight involved the shoulder and thus slowed the times down.

But given that a typical keyboard alpha area width is less than 16 inches, in theory there is hardly any time difference between moving 2 keys or 4 keys, for some people at least. Think the issue is not so much the "moving" as stopping in the right place.

Keyboards and Other Interfaces / Re: Finger strengths
« on: March 17, 2018, 12:52:43 AM »
On the other hand, and the reason why I want to touch type, I keep doing the bobbing head thing from kbd to screen .. not good for neck and shoulders,not good at all.

I think we need to introduce some new terminology... as I understand it, "Touch Type" has two components:

1. each key is assigned to a particular finger. Thou Shalt Use No Other Finger.

2. the keyboard map is integrated into your head-hand continuum. So you should not need to look at the keyboard because you "know" where the key is.

Now I'm reasonably good at (2), even when I hit the wrong key, I hit backspace and try again (sometimes 2 or 3 times) but rarely need to look at the keyboard, I just know I was a key or so off to the left or right, and try again...)

But there is no way I comply with (1). So am I "touch typing" or not?  (*)

At this point I don't really have better terms. Perhaps (2) could be "blind typing" or somesuch, and (1) could be something referencing "correct fingers".

Cheers, Ian

(*) The other day my therapist asked if I was "touch typing" (ie yet/again) and how my speed was (after the whole smash-my-left-arm incident). I didn't know how to answer because I don't consider my typing to be "touch typing"....

Keyboards and Other Interfaces / Re: Finger strengths
« on: March 16, 2018, 02:20:10 PM »

marked degree. Barnes (/) reviews evidence

Available at 173MB PDF.

It's a textbook on Time and Motion studies... 3rd edition. Scanned so text not searchable :-((

Found attached, possibly text changed between 1st and 3rd editions.

FWIW, the other day I was thinking back to my days playing arcade video games, in particular Asteroids Deluxe, which had five buttons. You steered with your left hand (2 buttons)(ob reminder: yes, you can use the mouse with your left hand...), left thumb button for shields, and right hand controlled thrust (forefinger) and fire (middle finger).  The way to fire off a rapid volley was to freeze your right hand and arm then move the whole thing up and down rapidly, rather than trying to just move your middle finger.

Keyboards and Other Interfaces / Re: Finger strengths
« on: March 16, 2018, 01:20:13 PM »
Been looking for some research/book/whatever that actually looks at how strong each finger is.

Came across this in attached PDF, freely available on web.

Ballistic movements are rapid motions,
usually repetitive, in which active muscular
contractions begin the movement, giving mo-
mentum to the member, but cease or diminish
their activity throughout the latter part of the
motion. It is unlikely that, of themselves, the
fingers utilize this type of motion to any
marked degree. Barnes (/) reviews evidence
that in repetitive work finger motions are
more fatiguing, less accurate, and slower than
are motions of the forearm. Consequently, in
repetitive finger activities in which there is a
ballistic element, such as piano-playing, typing,
and operating a telegraph key, wrist and elbow
motions predominate while the fingers merely
position themselves to strike the proper key.

Which I suppose ties in with previous observations that people who type with 2 or 4 or 6 fingers can be just as fast as touch typists... because in this case they're relying on the strong muscles of the arms to do the heavy lifting rather than stretching and contracting their fingers to get to all the required keys...

All of which has major implications for the modelling we are doing in KLA.....

Nevertheless, regarding 'strength of fingers' .... was thinking that there are maybe three or four factors to consider...

1. actual strength, in particular, for pressing, while keeping rest of fingers reasonable still, as well as whether pressing almost vertically or at assorted different angles.

2. flexibility of each finger to move in different directions

3. speed of finger in doing such movements... we can not assume all fingers are equally fast, can we?

4. time to fatigue. Perhaps some fingers can work longer without tiring than others?

Also related to the "average finger strength" table I posted above... I think these were measured by measuring "pinching" force, which would mean that the thumb was pressing with the "pad" / finger-print part, rather than the edge as is common in typing. Which could invalidate my calculations, unless we can determine that the thumb is more or less equally strong pressing with edge as with pad.

I have actually modified my programs that used relative strength of fingers in evaluating layouts based on bigrams, trigrams, quadgrams and common words typeable on the home keys, just want to double check them again before uploading the results. On the one hand the results look better than what I got with previous rather arbitrary linear assignment of finger strengths ... now, layouts that can type all the required letters on home keys score higher, whereas previously layouts missing some letters came first. I hope this is a consequence of the weighting and not a logic bug. :)
As expected the split-case layouts come out on top, and the QWERTY/alphabeticals at the bottom.

The new results affect the Best Layout Overall results as well.

Cheers, Ian

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: March 14, 2018, 07:11:03 AM »
I was using Alice (1st in the last hehe) on Den v2, followed the link on your site's Tools page.
Is the 1st entry in the list, "Den's fork", den3?  is Den1

Den 3 is here:

Will update my site to be more accurate.
"My fork" is mirror of Den1 with some cosmetic changes.

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: March 13, 2018, 11:07:55 PM »
BEAKL5 seems to do better than 9 in both Ian's and Den's kla?

either I made a mistake entering 9, I am using the wrong kla versions .. or am just plain confused haha

I'm just mirroring Den1 ... presume you were comparing this to Den3.

Results depend a lot on your input text.... what were you using?

Cheers, Ian

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: March 13, 2018, 11:06:07 PM »
is the beakl9 kla json file available somewhere ?

You can find others here  (use the "Search" to filter the list)

Cheers, Ian

Keyboards and Other Interfaces / finger strengths
« on: March 11, 2018, 07:27:03 AM »
Revised numbers, from the original paper.

Interesting that middle finger comes out half as weak as thumb... maybe giving someone the finger is only half as good as a thumbs up... :-)

Guess I need to redo some of my non-KLA tests/calculations using these numbers instead of my previous "simplified" ratios of 1 - 5, thumb = 5, pinky = 1.

Cheers, Ian

Keyboards and Other Interfaces / Finger strengths
« on: March 11, 2018, 06:47:41 AM »
Been looking for some research/book/whatever that actually looks at how strong each finger is.

Finally found attached table which is from attached PDF, available to freely download on web.

Can't find anything that measures finger strength pushing a key, but even that is tricky to measure if you want the isolated finger action while rest of hand is held relatively still.

I guess these numbers are heavily based on "pinch strength" which I suppose is a reasonable proxy for "pushing a key" ... the relative strengths between fingers should be similar for the two tests.

So now .... do we use these ratios to recalibrate the scoring algorithms?....

Something like


Discuss :-)

Keyboards and Other Interfaces / Give me independence or give me death
« on: March 10, 2018, 01:31:31 PM »
Looks like our fingers are not really capable of moving independently of the others.   (can export to PDF as well, if you like)

I have no opinion yet as to whether this is relevant to our modelling efforts or not.
Yes, the math(s) is a bit above Pythagoras.

Keyboards and Other Interfaces / Assorted thoughts
« on: March 10, 2018, 10:21:43 AM »
Hi all

Various things running around in my head, thought I should write them down for future reference... mainly to do with KLA and the modelling process.

In no particular order...

1. At the moment AFAICT we calculate the effort or work required to type as a function of distance moved and finger used. If I remember correctly we have
  a. Patrick: linear distance between keys, calculated by Pythagoras
  b. Den1 : as for Patrick, plus 4mm down and 4mm up for each key press
  c. Den3: Distance between keys doubles x-distance, also 4mm up and down, and different weightings for the fingers, where pinky is punished. Can't remember if distance is sqrroot( (2x)^2 + y^2) or sqrroot( 2(x^2) + y^2).

Den3 calcs have bothered me for a long time, and in recent days think I have come to realise why. I understand it wants to punish horizontal movement, but I really think that it is actually easier for Index finger to move one key to left or right, than to go up one row. Going down may be easiest of all.
The situation with the pinky may be the same. Regret due to personal bad pinkies I am not qualified to offer opinion on this, but both index and pinky are blessed with sideways movement facility, and I'm not sure if it's fair to punish them for that.

Contrawise, the two middle fingers are not designed to move sideways when curled over the keyboard, and thus I agree sideways movement, as in ANSI layout, should be punished.

2. Fingers have a limited range of movement. For some keys, you need to move your hands as well. This 'extra effort' is not included in calculations.

3. Related to (2), how accurate are our finger maps? Do people really use pinky to hit backspace or \| or ~ on QWERTY? Or do they make like me and use ring finger? If so, then we need to redo the finger maps to match reality and not some ivory tower touch typing ideal scenario. All of which is going to invalidate all the current scores.

4. Since people have different size hands (eg ) and we can't pick some magical 'average' (well maybe we can)), I was wondering if we could "allow" for the issue in (2) by changing the distance calculation somewhat, by loading the distance.
eg instead of say sqrroot( x^2 + y^2) we just use  x^2 + y^2 so that larger distances (implying having to move hand as well) are punished. The downside of this is that vertical motion is not adjusted accordingly... maybe we need to do 2(z^2) instead of just 2z.

5. At the moment (as in (1)) we assume the effort required to press a key is always the same. However I contend that pressing basically straight down on the home row requires less effort than in a situation that I use a lot, which is ctrl-insert on ANSI. In this scenario my hand is flat, thumb on Ctrl and middle finger extended straight out to Insert, and then press down. The muscle motions (and stresses on the joints) are very different to pressing K on QWERTY home row. I don't think the current "distance and which finger" models take this sort of action into account.

Okay enough for now :)

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: March 06, 2018, 10:58:40 AM »
For the MS Natural, I guess the one I have, the MS Sculpt Ergo is kind of the replacement for it.

Everything after version one was a step downhill, in my humble opinion... I've owned most of them (not the Sculpt, but have examined it in store). Still have 2 model 4000s, one in use.

I'm still using mine (MS Natural original). Smashing my wrist got in the way of many projects, but I have put the new keyboard on the dining room table, along with the tools and soldering iron. Still busy adding the wires to the Teensy, then need to connect Teensy to keyboard and see if any thing actually works... then it's the whole Program The Teensy step that's looming like a monster over me....

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: March 06, 2018, 10:41:34 AM »
I have been looking again at the reviews at and thought maybe the SmartYao could be an interesting compromise.
(sold as Koolertron on Amazon)
Physical keys (Cherry Mx or Gateron), split kbd, not staggered, some thumb keys, programmable and not as expensive as other better known alternatives.

Odd.. the pic on Xah's site has non-staggered keys, but the one on Amazon is staggered ...

Was going to suggest you look at MS Natural original but can't find it online now (only 2nd hand), I thought they were still being manufactured. Will be cheaper than more modern designs.

Else build your own ... do it slowly and spread the costs over months. It will work out more expensive, but then you get exactly what you want.. :-)

Keyboards and Other Interfaces / Re: not good
« on: February 22, 2018, 01:15:12 AM »
smashed my left wrist. back on ansi keyboard and mouse on right hand and mostly one handed typing for now. At least it has mech switches (gateron red I think)

We are the Borg. You will be assimilated. Resistance is futile.

Had another quick op to remove the wire. Surgeon was not very encouraging about regaining full use of left wrist, physiotherapist was more hopeful. Will have to see how it goes. Still on ANSI keyboard and mouse on right hand, can type a little bit with left hand.

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: February 02, 2018, 09:49:10 PM »
On a final note, I found it very intriguing that BEAKL 9 seemed to score comparatively poorly against other popular layouts in distance using the Keyboard Layout Analyzer. At first it put me off, but then I considered that if the fingers were moving up and down more then they might actually be less prone to fatigue (and indeed the same finger scores for BEAKL 9 were very good, especially for pinky), so what are your thoughts on distance as a metric of layout quality?

Which version of KLA? Patrick's original?

Cheers, Ian

Arts, Literature, and Crafts / Re: [lang] Flownetic: The Phonetic Script
« on: February 01, 2018, 10:53:42 PM »
is left column supposed to be empty as this stage?

Keyboards and Other Interfaces / different approach to one-hand keyboard
« on: January 22, 2018, 11:56:57 AM »
was doing some research on one-handedness, found stats:

But wait! there's more!

Their 'solution' is to use kiddie-size keyboards:

which seem seriously overpriced, compared to, for example

or from China for half that....

I shall refrain from further comment :-)

Keyboards and Other Interfaces / Re: Thoughts on Gigabyte K83 keyboard
« on: January 22, 2018, 11:44:42 AM »

3. Spacing... the numpad is too close to the nav keys, my fingers can't fit in the gap, and I end up typing numbers  (given (1) above) when I don't mean to. For this issue alone, I'd suggest you look at other options.

after staring at a few other keyboards lying around, realised the problem is actually a combination of narrow gap PLUS their island-style design.... other boards have a raised surface which is high enough to rest your fingers on without triggering keys.

Keyboards and Other Interfaces / Thoughts on Gigabyte K83 keyboard
« on: January 21, 2018, 03:28:42 PM »
When I got involved with layout design I thought it would be good to have a spare keyboard to test new layouts on, so I bought a low-end Microsoft keyboard.

Then discovered two problems with swapping keycaps around:

1. Profile... different row, different shape... (in truth I should have seen that one coming...)

2. F and J have different stem alignment... don't fit nicely in other places.  Or other keys in their spots. Guess this is an assembly aid.

So I was annoyed and splashed out on a low-end mech keyboard instead... Gigabyte K83, with Cherry Red switches (wanted to see how reds were, since I had bought browns for the keyboard I'm building, and not going to use blues....)


1. Reds are quite sensitive. See (3) below.

2. Switches bottom out quite hard.... maybe this is a function of me banging my MS Natural for 20 years and I need to develop a lighter tough. I have now installed black rubber ring bumpers (from Massdrop) on Ins and Del, since using them for copy-paste was hurting my middle finger a lot. The bumpers seem to help quite a bit.

3. Spacing... the numpad is too close to the nav keys, my fingers can't fit in the gap, and I end up typing numbers  (given (1) above) when I don't mean to. For this issue alone, I'd suggest you look at other options.

Cheers, Ian

Keyboards and Other Interfaces / Re: I only use one hand...
« on: January 19, 2018, 02:39:06 PM »
Try the CLP model.

Yeah, came to same conclusion, but with a twist.. no upper case slots needed.

Just another toggle on thumb for CAPS.

So that helps reduce the key count considerably.

Get a mouse with many buttons. Assign them meta and editing actions that replaces any ctrl- stuff. copy, paste, select all, save, back, etc.

Am considering it. Problem is my usual online stores take a week or two to deliver (no next-day Amazon-style here AFAIK)
This cast is supposed to come off on 30th and be replaced with modern design and at that point they'll probably want me using my fingers as much as possible (to ensure tendons etc don't get too attached to the steel plates). So may end up NEEDING fancy mouse for only few days. Am going back to my trackball :-)

Speaking of mouse, using one on my right hand again is damaging indicated spot on attached, Years ago when I used mouse I put blanket on table then covered it with sheet, and that was my desk/mousepad. Suppose that protected my hand. Sheet gave nice solid colour for mouse laser to measure on, blanket provided padding.

Keyboards and Other Interfaces / I only use one hand...
« on: January 18, 2018, 12:34:14 PM »
So yeah, Ansi 104 sucks even more for one hand than for two.

As such, both one-handed Dvorak layouts are pointless wastes of time... the form factor is completely wrong.

For right-hand use, the lack of alt-tab on right is particularly annoying, as is the constant switching between mouse and keyboard. At least the right-hand copy/paste operations are more comfortable than ctrl-xcv would be, but ctrl-a is awkward.

Maybe need better mouse with more buttons that I can assign to such things.

As far as keyboard goes, it feels like something like my mobile layouts (eg Z8 here are what is needed but I have some issues with my own layouts that annoy me. (in particular, the M, and distance between Y and OU).

So started with a mockup on KLE, still lots to do. Am thinking of main block with letters and 4 puncs, then shift layer, then two more layers, one numeric, one punc. Not sure about navigation yet.

Will advise any progress or bright ideas.

Oh yeah, having ctrl-Q and ctrl-W next to each other is NOT GOOD.

Must maybe pick my surgeon and physiotherapist's brains about market for one-handed layouts... such users probably less pedantic than most about qwerty and maybe medical aid does not freak about price....

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: January 17, 2018, 01:44:39 PM »
Sorry to hear that.

Are you now gonna devise the perfect one hand layout? For that, I suggest sticky modifiers--so you don't have to hold down shifts, which would demand dexterity.

I did wonder if the universe was pointing me in that direction.

More worryingly so because a few days ago I had been thinking about one-handed layouts for some reason ... but more along the lines of "I'm glad I don't need to work like that" .... and now here I am doing that...

I'll see if the universe throws any bright ideas my way over the next few weeks....

Cheers, Ian

Keyboards and Other Interfaces / not good
« on: January 16, 2018, 07:26:15 AM »
smashed my left wrist. back on ansi keyboard and mouse on right hand and mostly one handed typing for now. At least it has mech switches (gateron red I think)

Keyboards and Other Interfaces / Keyboard for mouths
« on: January 13, 2018, 07:35:41 AM »
Stumbled across this:

Tried playing around with it but requirement of having vowels and liquids on home row works at odds with putting common letters on home row. T, N and D can easily be put in better (but not optimal) spots.
Also have H and E on same finger is never going to work.

Came to the conclusion that it may be comfortable for the mind, but we type with fingers not brainwaves, so comfortable for fingers is more important.

Keyboards and Other Interfaces / GhettoDox
« on: January 05, 2018, 04:12:26 AM »
For when you can't afford to buy an ErgoDox.

Keyboards and Other Interfaces / Re: Same finger bigrams
« on: January 04, 2018, 02:57:19 PM »

Colemak 82.48
MTGap 93.05
Capewell 128.75
Klausler 181.35
Dvorak 196.04
Dangvu 204.9
Workman 212.57
QGMLWY 350.73
Norman 503.58
QWERTY 525.73

Voice in my head was still complaining... something about finger usage.... so I did a quick experiment, and multiplied each letter pair on same finger by a weighting per finger (simplified model: thumb = 1, pinky = 5).

Overall resulting order was sorta the same, but some differences, here's the revised list:

Colemak 210.78
MTGap 264.3
Capewell 377.79
Klausler 440.98
Dangvu 482.96
Dvorak 523.69
Workman 567.26
QGMLWY 802.54
QWERTY 1426.34
Norman 1446.09

Which is not bad, and except for QGMLWY, compares order-wise with the same-finger list (here in reverse order) on

    Norman: 4.7%
    QWERTY: 4.6%
    Minimak: 3.1%
    Asset: 2.6%
    Workman: 2.4%
    Dvorak: 2.2%
    Klausler: 1.6%
    Capewell: 1.4%
    Arensito: 1.3%
    Colemak: 1.3%
    QGMLWY: 1.3%

Norman is known to have bad same-finger metric.
I assume QGMLWY is out of order for me because of the finger weighting.
It's score was calculated thus: (first digit is KLA finger number)

1 => 'qdz': 0.42
2 => 'gsx': 20.0084
3 => 'mtc': 94.0185
4 => 'lwnrvj': 111.011
5 => ' ': 0
6 => '': 0
7 => 'yfiakp': 447.6328
8 => 'ue': 31.4118
9 => 'bo': 98.0376
10 => 'h': 0
11 => '': 0

Cheers, Ian

Keyboards and Other Interfaces / Corpus update
« on: January 03, 2018, 12:09:57 PM »

FWIW an update to the corpus used in tests.

At the moment, for comparison purposes, I am only using the files in EnglishTexts and Programming.

The WordLists, MixedLanguages, and DigitsEtc are useful for fine-tuning particular aspects of a layout, or just cross-comparisons to things non-English or rare English. But don't think it's fair to compare layouts using them as inputs.... my focus is on English and Programming. I believe if you're going to enter hundreds of digits, you should use a numpad....

These versions SHOULD be free of non-ANSI/ASCII characters, ie only stuff found on normal ANSI 104 keyboard. Think one of programming files still has two(?) arrows (or something) hiding somewhere.

The home-alone.txt is the list of words from the various home-key, home-block, easy-block and one-handed tests... the sample long words from the various layouts, deduped. (home == home key/block, alone == one-handed). Yes I know my sense of humour is weird.

Cheers, Ian

Keyboards and Other Interfaces / Re: Same finger bigrams
« on: January 03, 2018, 01:22:28 AM »
Added another metric of my own invention, which possibly has some similarities to methodologies used by the scientists in one of those studies comparing QWERTY with Dvorak etc.

The Famous Brands stack up like this: (all standard ANSI variants)

Colemak 46.68
Klausler 60.06
Dangvu 64.53
MTGap 70.13
Workman 81.62
Capewell 94.94
QGMLWY 109.93
Dvorak 129.02
Norman 162.00
QWERTY 181.65

After a restless night of tossing and turning, I finally gave in to the voice at the back of my head and changed the calculation a bit.

Consider qu and et. The chances of u following q, and t following e, are:

qu: 98.7
et: 2.66

So if a layout had qu on same finger, I would have added 98.7 to the score, and only 2.66 if et were on same finger... but clearly having et on same finger is much larger problem than qu, since et are two most common letters.

So now I multiply the above numbers by the frequency of the first letter in English, so we have

qu: 98.7 * 0.1 == 9.87
et: 2.66 * 12.1 == 32.186

which is a much better reflection of reality.

So the Famous Brands now stack up as follows:

Colemak 82.48
MTGap 93.05
Capewell 128.75
Klausler 181.35
Dvorak 196.04
Dangvu 204.9
Workman 212.57
QGMLWY 350.73
Norman 503.58
QWERTY 525.73

The relative percentages are going to have a much wider range, since the scores now range from 8.01 to 1,757.11.

Keyboards and Other Interfaces / Same finger bigrams
« on: January 02, 2018, 02:28:52 PM »

Added another metric of my own invention, which possibly has some similarities to methodologies used by the scientists in one of those studies comparing QWERTY with Dvorak etc.

It's based on the probability of one given letter following another ( ) (took me ages to manually type all those probabilities).

Anyway, we see which letters are on a given finger, then look up the probability in both directions for all letters on same finger (ie so we add "er" and "re", for example), and add them all up, and rank the results.

I ignored Capital letters, and also when both letters were on same key (as in CPL layouts), since the idea is to see how much a given finger has to fly around.

No distances were harmed in this production. That's under ToDo, with Row Jumps, Rolls, etc... when I figure out some way to score it.

Anyway the top end looks like this:

1   seelpy-1-1.en.ergolinear   15.48
1   seelpy-1.en.ergolinear   15.48
3   seelpy-1-4.en.ergolinear   15.61
4   beakrak-1-2.en.ergodox   16.06
5   hieaqmtsrn.en.ansi   16.39
6   proxkb.en.ergodox   17.19
6   proxkb-thumbshift.en.ergodox   17.19
8   maltron-us-90-mod-Andreas.en.ergodox   17.51
9   hieamtsrn.en.ansi   17.61
10   dvormax.en.ansi   17.77
11   opuiany.en.ergodox   19.14
12   balance-twelve.en.ansi   20.60
13   seelpy-1-8.en.ergolinear   21.03
14   beakl-clp-0.en.matrix   21.27
15   widely-alternating-A.en.ansi   21.33
16   beakl-modified.en.ergodox   21.88
17   mtgap-mod-joey2216.en.ansi   21.97
18   x1-ou.en.ergolinear   22.71
18   m2-tweak.en.ansi   22.71
20   maltron-us.en.ergolinear   22.86
20   maltron-us-90.en.ergodox   22.86

and the bottom end like this:

354   a-joy.en.ansi   238.65
354   a-joy-rehomed.en.ansi   238.65
356   qwerty-mod-noah.en.ansi   239.73
357   blick.en.ansi   241.01
358   neo2.en.ansi   242.76
359   ina-dv.en.ansi   247.47
360   ward.en.ansi   247.70
361   hoke.en.ansi   255.81
362   reverse-qwerty.en.ansi   258.62
363   rhythmic.en.ansi   269.31
364   tyler.en.ansi   270.50
365   qwerty-q-layout.en.ansi   272.72
366   cyberswarm.en.ansi   274.09
367   dvorak-onehand-right.en.ansi   275.74
368   wong-poiea.en.ansi   293.10
369   trotman.en.ansi   295.47
370   culemak-mod-Ian-hp.en.ansi   299.42
371   as-in-red-hot.en.ansi   308.38
372   mcgunnigle-peoples.en.ansi   343.95
373   tnwmlc.en.ansi   366.12
374   dvorak-onehand-left.en.ansi   384.13

For a change the bottom end is not QWERTY and Alphabetics-dominant.

The Famous Brands stack up like this: (all standard ANSI variants)

Colemak 46.68
Klausler 60.06
Dangvu 64.53
MTGap 70.13
Workman 81.62
Capewell 94.94
QGMLWY 109.93
Dvorak 129.02
Norman 162.00
QWERTY 181.65

Gory details here:

The details of how the score was arrived at are on each layout's page.

The graph showing the spread over all the layouts is actually the smoothest progression of the lot. Which kinda gives me some sort of confidence in the metric... :-) (FWIW)

Cheers, Ian

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: December 31, 2017, 11:36:54 PM »
You really put in a lot effort for this.

You're okay with Den3 at the current state?

I accept your reasoning behind it (ie punishing pinky usage). I may not agree with the exact weightings but have not looked at it hard enough to debate it.

Think it will be more useful to first run all the layouts through all the tests and see how they stack up... if for example, QWERTY suddenly comes out ahead of Dvorak, then we will need to talk... :-)

First need to get it running locally and then see where I need to put in extra "write to console" stuff if necessary, to get the data and error conditions I need. Should only be in one or two places I think.

I think I will change the weightings in the "average" calculations, at the moment each score is added once, except for Patrick English, and Den 1 English, which get added twice each, in an attempt to balance the numerous word-based scores.

Think I will change that to Patrick English x 2, Den 1 English x 3, and Den 3 English x 1, because Den 3 already includes some word-based metrics in its score.

My gut feel is still that Den 1 is "most accurate", in being largely agnostic about any particular design philosophy (where Den 3 takes the deliberate position against pinkies. There may be some people (eg Colemak with both A and O on pinkies) who would challenge this.)
But my gut has been known to be wrong :-)

Cheers, Ian


If you find any bugs/issues, please let me know.... probably at least one or two things that I forgot to update.


Still Den's beakl-clp-0.en.matrix.
Congrats :-)

I've rolled the results of the many hours of work over last few months to live... also updated the per-layout pages with extra data.

Added a few more home-key tests of my own invention, dunno how valid they are. And averaged all the tests in two different ways, leading to above result.

Started working on the "author" data, renamed to Inventor. Done mine and Den's and MTGap's (except the mods) so far. Still lots to do.

Next year will get Den3 scoring to work and retest all the layouts with that ....

Happy new year and all that (if it is your new year now ...)

Cheers, Ian

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: December 30, 2017, 12:58:21 PM »
Reverting back to BEAKL 9 and immediately see my speed improve.

Amazes me how you wrap your head around all these layouts so fast :-)

Keyboards and Other Interfaces / ZX 1
« on: December 29, 2017, 03:38:36 PM »

Was capturing which tries to improve QWERTY as much as possible by moving as few keys as possible, and started playing around...

Over the last few days I managed to rewrite the page on my site that lets you compare how all the layouts do for the different tests, but this time have combined all previous 4 pages (Patrick + Den 1 scoring for ANSI/Ergo) onto one, and includes over 360 layouts. Still fine-tuning it a bit before going live.

Anyway, I was annoyed because previously my S2 layout was the best "plain ANSI", where the keys were just re-arranged, no fancy tricks or anything (on Den 1 scoring).

Now the results came out as

1   light-anderson-mod-Ian-2.en.ansi
2   vukeys.en.ansi
3   dangvu.en.ansi
4   oneproduct.en.ansi
5   light-anderson-mod-Ian.en.ansi
6   liigol-mod-Ian-2.en.ansi
7   embizone.en.ansi
8   mtgap-tea.en.ansi
9   vukeys-swapped-mod-Ian-2.en.ansi
10   mtgap-iaet.en.ansi
11   goettl-brugh-julstrom.en.ansi
12   box-x.en.ansi
13   colman.en.ansi
14   juli19.en.ansi
15   pietergen-1.en.ansi

My S2 is not in that list because I swapped some keys on the top row, the layouts above don't do that. But it would have been about 4th on the list. Which was annoying because I was somewhat proud of it....
Light-Anderson is a layout I found in a paper, and applied my mind to it a bit (twice).

VuKeys and DangVu are closely related and in KLA.

OneProduct and Embizone I found on Geekhack I think. Colman is a Colemak variant.

Anyway, so I was playing around... and eventually arrived at a new layout which on average JUST makes it to number one, with an effort score of 137.30 (vs the winner above's 137.3923)

Screenshot and json attached.

Distance figure is good, samefinger not so good, because of right index and left middle. Really don't like the right index, with PH and GH and YO ... but could not find better solution.

Left hand will not like to type uncle or unclear or nuclear ...

Den won't like pinky usage. But pinky usage is in line with the other top layouts I was testing against. And hand balance, while terrible, was better than the others.

But the Good Features:

1. maintains ZXCV
2. maintains comma/period/slash/quote
3. maintains top row
4. Home Row seems to be unique (not found amongst all the other layouts, even mirrors).

Vowels ended up on the right for a change...

Called ZX 1 because of ZXCV... yes I know it sounds like a Sony cellphone :-)

If you swap Q and J scores are slightly worse but typing QU will be more comfortable.

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: December 28, 2017, 09:11:21 AM »
'Twas the night before Christmas
And all through the house
Not a creature was stirring
Not even a mouse....

Deathly quiet here lately :-)

Anyway stumbled across this:

Which I thought was interesting, and of course had to have a go at tweaking it ...

I think that sort of design might actually get traction in the market... it's a lot more "normal" that the ergo layouts...

Suspect the designer uses Emacs and Lisp, which led to some of the key placements, which are not optimal for English.

Cheers, Ian

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: December 14, 2017, 03:31:42 PM »
Code: [Select]

qhoux ,drcb
yiea. lstnp
j/'kz wmgfv

I was thinking the other day... one of the (many?) criticisms against new layouts is whether they maintain ZXCV at bottom left, for the purposes of short-cuts. And possibly A as well.

But in truth, it is not so much the LETTER that is important, but the closeness of those keys to the left Control key.

So let's say a swanky new layout puts PKQX there instead. Surely we can just tell the OS/keyboard driver to treat Control-P as Undo, Control-K as Cut, etc?

Or do people really need to see the XCV to know what they're going to produce? (FWIW I use the right-hand shortcuts mostly - ctrl-insert, shift-insert, etc) and my fingers do that without me thinking too much... no need to look.)

(Yes I know Den says cut/copy/paste should have dedicated keys, but ANSI/ISO does not always oblige... :-) )

Keyboards and Other Interfaces / Ranking
« on: December 14, 2017, 01:54:25 PM »
How do I rank thee? Let me count the ways...

Am fiddling around with averaging rankings from various tests. But have a philosophical problem....

Let's say 5 people take a test with max score of 10, and score as follows:
A and B: 10
C and D: 9
E: 8

Now to rank them, do we say:
A and B: 1st
C and D: 2nd
E: 3rd

A and B: 1st
C and D: 3rd
E: 5th

Or relative ranking:
A and B: 1/5  ... or 1/3 
C and D: 2/5   or 2/3 ... or  3/5
E: 3/5   or 3/3  or 5/5 ?

I can make arguments in favour of all variants, but trying to pick fairest one to use...

Keyboards and Other Interfaces / Re: Balanced Keyboard Layout
« on: December 14, 2017, 01:47:05 PM »
Didn't like the row jumping in BEAKL 9 (mainly the L in the bottom middle, "cl" "fl"). So messed around again to get something that feels better.

Mmmm... is "hi" not a problem? or "ls"? I remember reading somewhere about someone complaining about one of the well-known layouts (Dvorak, I suppose), putting l and s on same finger ( AND on the pinky...). Becomes a problem in Linux command line.

Today finished testing with Patrick's scoring (Den1 already done, Den3 next), Beakl14 will have to wait for the next batch of test next year... :-)

FWIW Patrick's scoring thinks Essie 2 layout is best. No idea why. From my point of view it was a 'dev' layout, not final.

Cheers, Ian

Keyboards and Other Interfaces / Klausler
« on: December 12, 2017, 11:38:14 PM »
Found a link to Klausler's evolved layout process, which was dead, eventually tracked it down on wayback machine.

I suppose the crux is this, which goes to the heart of the Colemak/Dvorak debate (rolls vs alternation):

The next step was to actually try using the layout.  I spent a couple days with it, and learned that my layout evaluation function was just too smart for its own good.  Too many words required complicated patterns using the fingers of the right hand. The word bottom convinced me that Dvorak was on to something when he designed a keyboard that maximized alternation between the hands.

(The insight is that hand alternation increases parallelism.  When the fingers of one hand are hitting keys, the fingers on the other are getting into position atop the next keys.  This should have been obvious, but it wasn't until I started the third experiment and saw some empirical timing data that I realized how much faster things are with high rates of hand alternation.)

So I updated -- simplified, really -- my evaluation function.  Now I charge points when too many keys are hit in succession by fingers of the same hand, with some credits for hitting adjacent keys.  Specifically, the new simplified rules are:

    Every position has an assigned cost that's looked up from a table.  The OEU/HTN positions of Dvorak have no cost.  A and S are 1 unit; I and D are two; <>P and GCR are three; Y and F are four; ", QJK, MWV, and L are five; : and Z are six; and X and B are seven.
    Using the same finger twice in succession on distinct letters costs 10 units.
    When two keys in a row are struck with the same hand, it costs two units if they're on different rows or on the bottom row, and one unit if they're not adjacent.
    If three or more keys are hit in succession by the same hand, one unit is charged for each key after the second.

Now I need to see how much I agree with that, and how to incorporate it into some sort of "comfort" metric for a layout.

Keyboards and Other Interfaces / Average rankings...
« on: December 12, 2017, 02:09:12 PM »
Am still busy running tests with Patrick's scoring (about half way through the English stuff), then need to run with Den3 / Klatest scoring.

In the mean time, I've done an experiment. I took the results from finger and word tests, ranked the layouts in order from best to worst in each test, then averaged the ranks, which produces the attached spreadsheet.

Results may change slightly as I finish up with Pat's scores, and may change considerably when I add in Den3 scores.


1. congrats to Den for coming first with beakl-clp-0.en.matrix.
2. the Schizo layouts did considerably better until I added the "efficiency" ranking, I must check that those calculations are correct.

What's in the mix:

1. Patrick's scores, highest to lowest
2. Den1 scores, lowest to highest
3. HomeKeys words, highest to lowest
4. EasyBlock words, highest to lowest
5. HomeBlock words, highest to lowest
6. OneHanded words, lowest to highest
7. Efficiency, highest to lowest, using Den1 distance measurements. Patrick's distance measurements are way off for AltGr style layouts.
Efficiency is calculated by assuming you will need to move 1 key width for every letter typed (plus up and down). So we work out that distance, then compare to ACTUAL distance, to get an efficiency score.

Then take a simple average of  each layout's rank (NOT score) for these 7, and rank that.

As per usual, the tnwmlc.en.ansi, alpha and QWERTY layouts are at the bottom, so as a methodology, it can't be all bad.

You should be able to find all the layouts here:

Keyboards and Other Interfaces / Flying fingers
« on: December 12, 2017, 01:42:48 PM »
Was poking around trying to find a way of measuring "ease of use" of a layout ... the Seelpy/Essie variants get top scores, but I have my doubts about how usable they are in real life, so trying to find a way to measure that.

Anyway found this PDF

Which includes this chart (attached) which is very interesting. Basically it says you can type nearly as fast with 1 or 2 fingers as with 10, depending on how well your keyboard/OS allows key rollover, as well as, I suppose, depending on how fast your hands/fingers can fly around to exploit the rollover capabilities.

Which means (possibly....) that the whole KLA model (and anything else based on touch-typing-you-will-use-index-finger-for-this-and-return-to-base-afterwards) is only one way of doing it, and a different model based on 2-finger typing, with no return-to-base is another....

Pages: [1] 2 3 4 ... 16