Keyboard Layouts Fight!
Xah Lee
〈Dvorak, Matron, de-ergo, NEO, Colemak, Programer Dvorak, Keyboard
Layouts Fight!〉 http://xahlee.org/kbd/dvorak_and_all_keyboard_layouts.html
plain text version follows
--------------------------------------------------
Dvorak, Matron, de-ergo, NEO, Colemak, Programer Dvorak, Keyboard
Layouts Fight!
Xah Lee, 2010-08-30
Noticed that today there are a lot of keyboard layouts. There's your
qwerty, and there's Dvorak. That's good enough, right? No. Apparantly,
a lot people are making a lot layouts. Some are specialized on a
particular language (such as German), some aim for easier transition
from qwerty, some are designed for programers.
Here's a list of them.
QWERTY
Th good old QWERTY, always around when u need it or don't need it.
qwert yuiop []\
asdfg hjkl; '
zxcvb nm,./
Dvorak
My favorite, see: Dvorak Keyboard Layout.
',.py fgcrl /=\
aoeui dhtns -
;qjkx bmwvz
Following is the numbers row of the original dvorak design, but is not
widely used.
75319 02468
Maltron
The Maltron keyboard is actually a keyboard hardware. Its design shape
is similar to Kinesis Contoured Keyboard, featuring split key clusters
for each hand, bowl shaped surface, straight aligned keys, major key
clusters for the thumbs (Enter, Delete, Space, ...). The company
created its own layout the Maltron layout; design goal is similar to
Dvorak. But their keyboard hardware also support qwerty and dvorak.
maltron.com.
qpycb vmuzl
anisf dthor
.,jg, ;wk-x
e
If you are wondering about that odd “e”, that's one of the key for the
left thumb cluster. Other keys for the thumbs are Space, Enter,
backward delete, forward delete, Tab, arrow keys, home, end, page up/
down.
Maltron keyboard with Dvorak layout.
One problem Maltron never appealed to me is because the keyboard is
quite ugly. Also am not sure why they have to invent the Maltron
layout since dvorak is already there, perhaps due to patent issues of
the time.
Time mag actually ran a article on it in 1981. 〈Science: The Case of
QWERTY vs. Maltron〉 (1981-01-26) at time.com
Programmer Dvorak
Programmer Dvorak is just like Dvorak, except that it inverted the
number row with symbols and re-arranged the symbols.
http://www.kaufmann.no/roland/dvorak/
~%7531 90246 8`
$&[{}( =*)+] !#
',.py fgcrl /=\
aoeui dhtns -
;qjkx bmwvz
I have tried the inverted number row, for a few weeks in mid 2000s. I
don't particular find them useful. Because, in programing, the numbers
are needed almost as frequently as those symbols. But now you need to
press Shift for the numbers. If you need to type several numbers such
as date “2010-08-30”, this layout will be painful. On the other hand,
you never need to type those symbols as a sequence. (especially bad if
you don't have a number pad. I do have number pad, but i use them as
extra function keys.)
My personal solution for programing is to have modifier keys ((Win =
Super), (Menu = Hyper)), together with the home row keys, to produce
the most frequently used symbols: (), [], {}, "", =, +. The matching
pairs such as paren are inserted in pairs always, and cursor placed in
between. (You can get my emacs init files at: Xah Lee's Emacs
Customization Files.)
Colemak
Designed for easy transition from QWERTY. colemak.com. The colemak
site appeared in 2005.
qwfpg jluy; []\
arstd hneio '
zxcvb km,./
de-ergo and NEO
The de-ergo layout is optimized for Germany language. forschung.goebel-
consult.de de-ergo. A supposedly improved version is the NEO layout.
It started around 2005. pebbles.schattenlauf.de neo layout. The
following shows the NEO layout.
qvlcw khgfj
uiaeo snrtd y
öüäpz bm,.x
Xah ∑ http://xahlee.org/ ☄
Xah Lee
75319 02468
but it is not in the ANSI standard version, and the layout that comes
with Windows and Mac OS X does not have that arrangement. However, i
heard that (some?) linux's dvorak layout use this arrangement?
Xah
> number row with symbols and re-arranged the symbols.http://www.kaufmann.no/roland/dvorak/
Scott L. Burson
> might be interesting.
>
> 〈Dvorak, Matron, de-ergo, NEO, Colemak, Programer Dvorak, Keyboard
> Layouts Fight!〉 http://xahlee.org/kbd/dvorak_and_all_keyboard_layouts.html
>
> plain text version follows
>
> --------------------------------------------------
> Dvorak, Matron, de-ergo, NEO, Colemak, Programer Dvorak, Keyboard
> Layouts Fight!
>
> Xah Lee, 2010-08-30
>
> Noticed that today there are a lot of keyboard layouts. There's your
> qwerty, and there's Dvorak. That's good enough, right? No. Apparantly,
> a lot people are making a lot layouts. Some are specialized on a
> particular language (such as German), some aim for easier transition
> from qwerty, some are designed for programers.
Here are two more for you. First, my own YAKL (Yet Another Keyboard
Layout):
q w d f j y u k l p ( ) \
a s e t g h n i o r
z x c v ; b m , . /
I designed this in 1990 or thereabouts and have used it on all my
machines ever since.
David Piepgrass's Asset Keyboard has an almost identical home row:
q w j f g y p u l ; [ ] \
a s e t d h n i o r
z x c v b k m , . /
Asset and YAKL were inspired by the same observation, which is I think
why they turned out so similar: for someone who already knows Qwerty,
Dvorak seems unnecessarily hard to learn. The only two keys that are in
the same location on Dvorak as on Qwerty are A and M. The benefit of an
improved layout is that the high-frequency keys are more accessible; the
locations of the low-frequency keys are not important. Given that, I
and Piepgrass thought, why not a layout that retains the Qwerty
locations for low-frequency keys, and just does the minimal rearranging
needed to get the high-frequency keys onto the home row? Surely this
would be easier to learn than Dvorak.
Once I had actually gone to the trouble of learning YAKL, however, I
concluded that I was mistaken. Learning to touch-type mostly isn't
about learning the locations of individual letters; that doesn't take
long. It's mostly about learning common digraphs, trigraphs, etc., up
to entire common words. Since YAKL intentionally rearranges many of the
most common letters to get them on the home row, the vast majority of
digraphs and practically all longer sequences are different vis-a-vis
Qwerty. These all have to be relearned anyway. Very little learning
effort is saved.
For this reason, even though I still use YAKL (having already gone to
the trouble of learning it), I don't recommend it to others looking for
a better layout. Just go with Dvorak. It has the advantage of being
menu-selectable on practically all OSes these days; setting up a
completely custom layout is quite a bit harder.
-- Scott
Thad Floryan
> [...]
> Here are two more for you. First, my own YAKL (Yet Another Keyboard
> Layout):
>
> q w d f j y u k l p ( ) \
> a s e t g h n i o r
> z x c v ; b m , . /
That's NOT a layout. Where are the numeral keys and the
Control key, Return, etc.?
Xah Lee
Thanks! interesting finding.
I searched and found David Piepgrass's Asset Keyboard page at:
http://millikeys.sourceforge.net/asset/
have updated my page with thanks to you.
I like your comment about how these new designs are not really worth
it.
part of the equation is standization and practical convenience. Dvorak
isn't available in all popular operating systems until about 10 years
ago. (i think Microsoft is actually the first one to include it out of
the box, in mid 1990s? I know that Mac didn't include a dvorak layout
until OS 9 or OS 8 in about 1998 or so. (i switched to dvorak in 1993
or 1994, and basically created my own layout using ResEdit))
today, dvorak is somewhat well known, and available in basically all
operating systems out of the box. Creating more layout seems a step
backward. Because it worsens the main problem: universal support.
Many have tried to improve dvorak, however, that is basically not
possible unless the language is not english.
there are several sites i've came across in the past years that has
javascript or java code to accept a input text and come out result
comparing efficiency of different layouts. Basically, depending on
what in the class of input text, some might come out marginally better
than dvorak, but worse as soon as the type of input text is changed.
(e.g. shakespear, common mag journals, religious scripture, chat
text.)
Xah ∑ http://xahlee.org/ ☄
Xah Lee
when not written out, it's assumed to be the same as qwerty. :)
Xah
Xah Lee
>
> 〈Dvorak, Matron, de-ergo, NEO, Colemak, Programer Dvorak, Keyboard
> Layouts Fight!〉http://xahlee.org/kbd/dvorak_and_all_keyboard_layouts.html
added much material. Added text follows.
-----------------------------------------------------------
Which is The Most Efficient?
carPalx (at http://mkweb.bcgsc.ca/carpalx/) is a site dedicated to
studying efficiency of keyboard layouts, by the method of using
computation and statistics to arrive at a best model of efficiency
parametrs, then evaluate the different layouts. The site is created by
a Martin Krzywinski, a Bioinformatics scientist.
According to him, the Colmak is better than Dvorak, but 3 other
computer generated layouts is better than Colmak. See: 〈Full
Optimization - Q*MLW* Layouts 〉 http://mkweb.bcgsc.ca/carpalx/?full_optimization.
------------------------------
No Consideration of Hardware Key Layout Problems
Also note, layout studies online always assume the common PC
keyboard's hardware layout. The PC keyboard hardware key layout has
several major problems with respect to ergonomics. For example, the
Enter, Tab, are important keys but requires the pinky to press. The
right hand has a lot more keys, all for the pinky. The key alignment
are un-naturally jagged. All these, are fixed by both the Maltron
keyboard and Kinesis Contoured Keyboard, and to some degree by
Microsoft's Natural Ergonomic Keyboard 4000. (See: Keyboard Hardware
Design Flaws)
------------------------------
No Consideration of Number Arrangement or Symbols
Another major problem is the very inefficient arrangement of the
number keys, as well as programer's need to type some of the symbols.
None of the layout studies after Dvorak seems to consider this.
(Kinesis arguably fixed the problem by using a left thumb modifier
that turns the right hand keys into a number pad.)
------------------------------
Is Improvement Over Dvorak Necessary?
The science of Ergonomics assumes that the activity is prolonged. For
example, if you just type less than 30 min a day, like vast majority
of people, qwerty is perfectly fine.
If you are a professional coder or writer, who's hands are actively
typing perhaps 3 hours a day, then Dvorak will be significantly
important to you.
If you often have coding or writing sessions that makes you work 70-
hour weeks, or worse, if you are a professional data entry clerk and
you type continuously for 8 hours a day, then a serious ergonomic
keyboard such as Kinesis or Maltron is critical.
When you consider designing the most perfect layout, it's important to
consider practical values of such layout. If you created a most
efficient layout, but the layout is for is standard PC keyboard, you
are likely to get RSI if you need to type continuously for several
hours a day.
(from my recent experience (see: Left Wrist Motion Pain; vi Esc key
Syndrome.), if you just type the Enter key and Tab key all day, just
these two keys and nothing else, on a standard PC keyboard, with hands
at touch-typing positions, say do it once every 30 second, for 8
hours, you'll get RSI FAST.)
------------------------------
What About a Reasonable Efficient Standard Layout for All Languages?
Another common problem is for international users, of non-English
languages. For example, German, Spanish, French, and even Chinese and
Japanese can benefit because their input methods commonly rely on
latin alphabet.
In these languages, usually there are few extra characters that needs
to be typed. There are many standardized layouts for them (e.g.
QWERTZ, AZERTY), but often they still requires you to type the special
chars by a combination of key press using AltGr modifier, and these
layout usually do not consider any ergonomics in the sense of dvorak.
(see: Idiocy Of Keyboard Layouts.)
It's much better to find ways to create a universal layout that are
largely efficient, fixes the hardware layout problem, fixes the number
arrangement problem, and can be used for all languages. I think this
is quite doable.
Xah ∑ http://xahlee.org/ ☄
Tassilo Horn
Hi Xah,
> In these languages, usually there are few extra characters that needs
> to be typed. There are many standardized layouts for them (e.g.
> QWERTZ, AZERTY), but often they still requires you to type the special
> chars by a combination of key press using AltGr modifier, and these
> layout usually do not consider any ergonomics in the sense of dvorak.
> (see: Idiocy Of Keyboard Layouts.)
Well, at least with standard QWERTZ and German Dvorak Type II (which I
use), the keys for äöü can be typed without modifier. With QWERTZ, even
ß is not modified, although you don't need it too often in German.
The problem is that although the German Dvorak Type II is quite
convenient when writing plain text, it's not that good for programming,
cause {[]}| and others are all modified. It's still better than QWERTZ,
where even / needs a modifier...
I would like to have some programming mode, which assigns more important
symbols to the äöü keys. You don't need those when programming anyway.
> It's much better to find ways to create a universal layout that are
> largely efficient, fixes the hardware layout problem, fixes the number
> arrangement problem, and can be used for all languages. I think this
> is quite doable.
I'm far less positive. I can't even think about a reasonable compromise
for only latin languages...
Bye,
Tassilo
Xah Lee
On Aug 31, 4:04 am, Tassilo Horn <tass...@member.fsf.org> wrote:
> Xah Lee <xah...@gmail.com> writes:
>
> Hi Xah,
>
> > In these languages, usually there are few extra characters that needs
> > to be typed. There are many standardized layouts for them (e.g.
> > QWERTZ, AZERTY), but often they still requires you to type the special
> > chars by a combination of key press using AltGr modifier, and these
> > layout usually do not consider any ergonomics in the sense of dvorak.
> > (see: Idiocy Of Keyboard Layouts.)
>
> Well, at least with standard QWERTZ and German Dvorak Type II (which I
> use), the keys for äöü can be typed without modifier. With QWERTZ, even
> ß is not modified, although you don't need it too often in German.
you are right. I glossed over that.
haven't done much study on this but did spend a couple of hours to
look at various international layout before...
〈Idiocy Of Keyboard Layouts〉 http://xahlee.org/kbd/keyboard_layouts.html
my impression is that i see lots of them rely on Alt Gr... and i don't
really find some of the qwerty variation of swapping 1 or 2 pairs of
keys makes any ergonomic sense to me.
it'd be nice to throughly study the various layouts for different
langs, to get a sense if the special char needed are a direct key
press or not, by what percentage, etc.
(e.g. first step is probably to get to know what special chars are
needed in that lang, and how frequent for each, then look at the
layout(s) of that lang, see if the char needs AltGr, their
position, ... a evaluation of efficiency can come by devicing a
formula that maps char frequency to a value of key press ease)
one thing i wondered about the different layout for different lang is
that how they came by? is it much designed by a single person just by
chance and somehow become the standard? or was it more thru formal
study and evaluation... (my random guess is the former)
e.g. here's the Wikipedia page showing different keyboard layouts
http://en.wikipedia.org/wiki/Keyboard_layout
> The problem is that although the German Dvorak Type II is quite
> convenient when writing plain text, it's not that good for programming,
> cause {[]}| and others are all modified. It's still better than QWERTZ,
> where even / needs a modifier...
>
> I would like to have some programming mode, which assigns more important
> symbols to the äöü keys. You don't need those when programming anyway.
>
> > It's much better to find ways to create a universal layout that are
> > largely efficient, fixes the hardware layout problem, fixes the number
> > arrangement problem, and can be used for all languages. I think this
> > is quite doable.
>
> I'm far less positive. I can't even think about a reasonable compromise
> for only latin languages...
humm... i think starting with Maltron or Kinesis, such a keyboard and
layout can come...
one thing i always wondered about Kinesis is that why no other
keyboard manufacture make those bowl shaped key surface and with thumb
key clusters. I vaguely recall it was due to a patent. Can anyone
confirm?
Xah
Stefan Nobis
> I would like to have some programming mode, which assigns more
> important symbols to the äöü keys. You don't need those when
> programming anyway.
You might want to have take a look at
--
Stefan.
Alan Mackenzie
> Dvorak isn't available in all popular operating systems until about 10
> years ago. (i think Microsoft is actually the first one to include it
> out of the box, in mid 1990s?
Xah, that's not English. Only a very careless native English speaker
could have written those two sentences. The correct verb tenses in those
two sentences are "Dvorak hadn't been available .... until 10 years ago"
and "Microsoft was the first one". Humour me, just for once, and tell me
what your native language is, please.
> I know that Mac didn't include a dvorak layout until OS 9 or OS 8 in
> about 1998 or so. (i switched to dvorak in 1993 or 1994, and basically
> created my own layout using ResEdit))
There seems to be an assumption through this interesting topic that
Dvorak is superior to Qwerty and friends. Is there any solid evidence
for this or is it all anecdotal?
> Xah ? http://xahlee.org/
--
Alan Mackenzie (Nuremberg, Germany).
Xah Lee
> In comp.emacs Xah Lee <xah...@gmail.com> wrote:
>
> > Dvorak isn't available in all popular operating systems until about 10
> > years ago. (i think Microsoft is actually the first one to include it
> > out of the box, in mid 1990s?
>
> Xah, that's not English. Only a very careless native English speaker
> could have written those two sentences. The correct verb tenses in those
> two sentences are "Dvorak hadn't been available .... until 10 years ago"
> and "Microsoft was the first one".
on the surface i think you are right, but am not sure Alan if looked
in depth.
(e.g. if you ask a university professor of literature; or professor of
linguistics; or, those people who write on Language Log @
http://languagelog.ldc.upenn.edu/nll/ (and those who are itchy to
digress on this, start a new thread pls! i will enjoy it, and will we
all!))
> Humour me, just for once, and tell me
> what your native language is, please.
am chinese by blood, alan. Chinese is what i grew up with in first 14
years. (not to mention body language)
can't believe you really need to ask that twice, of all these years.
y'know, google 30 seconds will find u the answer, right from the
dragon's mouth. My basic bio with clear and explicit info on this is
online for like a decade.
> > I know that Mac didn't include a dvorak layout until OS 9 or OS 8 in
> > about 1998 or so. (i switched to dvorak in 1993 or 1994, and basically
> > created my own layout using ResEdit))
>
> There seems to be an assumption through this interesting topic that
> Dvorak is superior to Qwerty and friends. Is there any solid evidence
> for this or is it all anecdotal?
god! you are not trolling right?
i tell u what. i read all sort of info on dvorak online since the web
started (and before that, it was books) Beginning in about late 1990s,
starting with slashdot.org, questions like basic doubt of dvroak has
been drivelled all over. But that seems to be over in the early 2000s.
But you still asking that??
sometimes, when discussing with some of the emacs developers, i
really, really find things so curious. You, lennart, and eli...
reading the emacs dev, i also see the same phenomenon, where A is
absolutely sure of a fact, while B doesn't even understand what A is
talking about, and vice versa, and took long convo back and forth to
possibly clear the communication but not necessary agree (and i think
after a few weeks each party really have forgotten whatever the
other's point of view on a issue at all, even if it is a plain fact)
So, i thought perhaps some of this has to do with the nature of
texture communication, and the fact that most of us hardly know each
other other than online posts of very narrow scope... but i think part
of it has to do with not really being serious. In a professional
environment (real professional environment, like having to see each
other's face weekly in a company, where the goal is more precise and
familiar to all...), persistent mis-communication doesn't happen to
this degree.
but still, i'm quite surprised often of this...
you asking me, showing some doubt, about dvorak's efficiency over
qwerty. I dunno what to say, but have you try to read what's been
posted? I think if you really think about the question and ask
yourself, i think you'll know the answer and convinced (not to mention
actually took the time to learn dvorak or do research on this) I mean,
literally, this question has been discussed just about in every online
forum that's remotely related to computers or keyboard or input
devices. slashdot mentions dvorak every few years... hacker news,
delicious, reddit... hundreds and hundreds of discussions, and the new
comer stackoverflow... but even, i think the question doubting the
basic tenets of dvorak has been over since early 2000s. I think the
topic of discussion has moved to more exotic ones, such as what's most
efficient layout, and dvorak with emacs or vim, or other new layout's
questions... etc.
ok, am getting ranty, if not already. Let me answer your q directly.
> Dvorak is superior to Qwerty and friends. Is there any solid evidence
> for this or is it all anecdotal?
this really depends on what you mean. If taken your question as is
without some special assumption of what you are really asking, then
yes of course, dvorak is superior to qwerty, in easier to type, more
comfortable to type, reduce RSI, reduce error, and for those who never
learned touch typing, its even easier to learn than qwerty, and
there's massive amounts of solid evidence.
evidence? well the first is the research done by Dvorak. Also
published as a book. Then else, in the thread i mentioned the caplx
site, which does studies and comfirm this, and there are quite a few
other sites with java or javascript that compute the various
distances, di-graphs, hand balance, etc on dvorak vs qwerty.
going deeper, perhaps you are one of those skeptics who doubt that all
this measure of key distances don't mean shit. Yeah, such voice has
been heard in online drivels, but it doesn't mean such doubt is not
valid. How to investigate that? well, you can look into stat research
of rsi over people, or more logical analysis of why key distances etc
shouldn't be valid, or spend few weeks to read all blogs of those who
actually learned dvorak (like, this is a method social research), or
read all blocks of those coders who developed rsi and thought about
going into dvorak ...
like, there's Wikipedia, which is my first stop in research just about
anything. So, maybe spend 20 min to read it? note that it usally links
to 10 or more other references or sites. Not saying it's all valid
just because lots of links , but it can be a starting pointing of
research to judge on this issue yourself??
how serious are you about this? i mean, if really, than start learning
touch type on dvorak to ultimately judge for yourself??
Nothing wrong with asking, nice question Alan. :D
Xah ∑ http://xahlee.org/ ☄
Xah Lee
http://hi-games.net/typing-test/
quite fun. My first try is 82 wpm. :)
those on high score list of top 30 starts at 126, and top is 164. :D
Xah
Marc Mientki
Josef Wolf
> might be interesting.
>
> 〈Dvorak, Matron, de-ergo, NEO, Colemak, Programer Dvorak, Keyboard
> Layouts Fight!〉 http://xahlee.org/kbd/dvorak_and_all_keyboard_layouts.html
I once tried to learn dvorak, and actually got so far that I could touch
type (NB: with qwertz, I type quite fast, but can't touch type). I found
dvorak much more efficient and ergonomic. But then, I came back to emacs
and found that I was no longer able to find all the key combinations. Even
C-x C-s was hard to get used to. So I resigned on dvorak and came back to
emacs.
So now to my question: whats the best way to get used again to emacs with
a dvorak (or better: NEO) layout? Anybody out there who managed this?
Xah Lee
i type dvorak since 1994 and started to use emacs in 1997, so i
actually never had qwerty+emacs experience.
i don't find any problem with dvorak+emacs.
since i switched to a split ergonomic keyboard in 2006, i do find it a
problem for Ctrl+x. So i swapped it with Ctrl+t. That solves it.
(dvorak x is qwerty's b; the dvorak t is qwerty's k)
also, you can try a ErgoEmacs keybinding.
〈ErgoEmacs Keybinding〉 http://xahlee.org/emacs/ergonomic_emacs_keybinding.html
〈Emacs, Dvorak, C-x〉 http://xahlee.org/emacs/emacs_dvorak_C-x.html
〈A Review of The Microsoft Natural Ergonomic Keyboard 4000〉
http://xahlee.org/emacs/ms_keyboard/ms_natural_keyboard_4000.html
If you search the web for the dvorak+emacs you'll find many accounts
of dvorak+emacs users don't do anything special. (and i've read same
for vi. I use vi occationally, and don't find any problem at all. the
cursor keys, hjkl is no longer on the home row, but is still not bad.)
〈Emergency vi (vi tutorial)〉 http://xahlee.org/emacs/emergency_vi.html
on Mac, there's a dvorak layout that support qwerty when modifier is
down. Some people find that great. I've always went full dvorak so
never cared for it. Though, in last 3 years of intensive keyboarding
studies, i find that having z x c v scattered is quite inconvenient.
Xah ∑ http://xahlee.org/ ☄
Xah Lee
that's fantastic. I got extremely excited. Downloaded, and tried
immediately. Though, i'll have to say it's quite disappointing. The
hitting return for each word is pretty bad. I think instead of
following 〈Typing Of The Dead〉, it'd be much better if just do normal
sentences
if following 〈Typing Of The Dead〉, i think one thing essential is to
have the words drop down.
though, kudos to Alex Schroeder for creating it.
here's some more thoughts
〈A Typing Game in Emacs〉 http://xahlee.org/emacs/emacs_game.html
Xah
Tassilo Horn
Hi!
>> I would like to have some programming mode, which assigns more
>> important symbols to the äöü keys. You don't need those when
>> programming anyway.
>
> You might want to have take a look at
>
> http://www.neo-layout.org/
Well, but to get to Layer 3 (the programming layer), you have to press
CapsLock, and that's an additional Ctrl key for me. I think, currently
I'm going best with Dvorak and sticky keys.
Bye,
Tassilo
Xah Lee
> 〈Dvorak, Matron, de-ergo, NEO, Colemak, Programer Dvorak, Keyboard
> Layouts Fight!〉http://xahlee.org/kbd/
dvorak_and_all_keyboard_layouts.html
more update.
--------------------------------------------------
discovered 2 competing layouts for french. (Thanks for Hugues)
Dvorak-fr and Bépo (French)
There appears to be 2 layouts for the French language, both designed
for efficiency. Both designs follows Dvorak layout principles. Both
have vows on the left hand home row.
One is called “Dvorak-fr” at algo.be dvorak-fr. This layout appeared
in 2002.
* 12345 67890 +%
_ =/-è\ ^(`)" []
:’ég. hvcmk z¨
oaueb fstnd w~
à ;q,iy xrlpj
Dvorak-fr has 2 other variations. “Dvorak-fr-e” for entering all euro
lang characters. “Dvorak-fr-k” for the Kinesis Contoured Keyboard.
The other is Bépo, at bepo.fr.
# 12345 67890 °`
$ "«»() @+-/* =%
bépoè !vdlj zw
auie; ctsrn mç
ê àyx:k ?qghf
Both Dvorak-fr-e and Bépo are designed to enter most or all accented
characters for other european languages.
It's interesting that both inverts the number row. Their designer
seems to be programers. However, they do not use the original Dvorak
layout for the number keys arrangement.
Xah ∑ http://xahlee.org/ ☄
Xah Lee
hi Scott,
some after thought. I didn't add your layout there because i try to
avoid private layouts that are not published. If you do have written
about it with intent to let other use... please let me know!
Xah
Elena
> When you consider designing the most perfect layout, it's important to
> consider practical values of such layout. If you created a most
> efficient layout, but the layout is for is standard PC keyboard, you
> are likely to get RSI if you need to type continuously for several
> hours a day.
Typing on a non-standard keyboard kills your ability to type on a
standard one, and standard keyboards are here to stay (think about
laptops). So a design for a standard keyboard is most useful.
All keyboard layouts I've seen, except Arensito, fail to relocate
often used control keys. Moreover, it seems to me that many users do
switch layout, but fail to switch keyboard and keep typing on a cheap
one, with lesser gain. Typist on typewriters did type a lot, yet it
seems they weren't catching any RSI.
Elena
> studying efficiency of keyboard layouts, by the method of using
> computation and statistics to arrive at a best model of efficiency
> parametrs, then evaluate the different layouts. The site is created by
> a Martin Krzywinski, a Bioinformatics scientist.
>
> According to him, the Colmak is better than Dvorak, but 3 other
> computer generated layouts is better than Colmak. See: 〈Full
> Optimization - Q*MLW* Layouts 〉http://mkweb.bcgsc.ca/carpalx/?full_optimization.
Carpalx has been shown to be a bad profiler, missing same finger
frequencies among other things (check the Colemak forum to learn
more).
Thad Floryan
Bingo!
I've been typing since I was 6 years old using an IBM electric
typewriter; I'm now retired (and drawing SS) so that's over 60
years of typing.
The key (no pun) is a proper ergonomic sitting position at the
keyboard: upright in a comfortable chair with arms horizontal and
at the level of the keyboard. Over the decades since the early
1960s I've typed millions of lines of code and never even had so
much as a hint of RSI, and I'd often be at the keyboard 14+ hours
a day. Though "retired", I still am at the keyboard for long hours
because I still develop and program.
Most of the keyboards I've used had an explicit CONTROL key to the
left of the "A" using terminals such as TTY ASR 33, Datapoint 3300
(a CRT clone of the ASR 33), Dec VT100, Datamedia DT80 (a clone of
the VT100), etc.
You can see the TTY ASR33 keyboard here:
<http://upload.wikimedia.org/wikipedia/commons/0/0b/ASR-33_2.jpg>
which is identical to the one I had at Tymshare and at home in the
1960s and early 1970s.
You can see the VT100 keyboard layout here:
<http://vt100.net/docs/vt510-rm/chapter8>
The Datamedia DT80 can be seen here (scroll down):
<http://www.classiccmp.org/dunfield/tty/index.htm>
What I use now (attached to a multi-gang IOGEAR GCS138 Miniview
Ultra KVM switch) is a Unicomp SpaceSaver which can be seen here:
<http://pckeyboards.stores.yahoo.net/en104bl.html>
with the [Caps Lock] mapped to be a Control key on *ALL* my
Linux, UNIX and Windows systems. The best keymapper for the
Windows systems is Ctrl2Cap which can be seen/downloaded here:
<http://technet.microsoft.com/en-us/sysinternals/bb897578.aspx>
<http://download.sysinternals.com/Files/Ctrl2Cap.zip>
which works perfectly on Win2K, WinXP, Vista and Win7 since I
have emacs on those systems, too. Here's my work area:
<http://thadlabs.com/PIX/Thad_desk.jpg>
Earlier comments about alleged evils of [Ctrl] being to the left
of [A] are bogus in my experience. Attempting to use emacs on
a PC keyboard *without* mapping [Ctrl] to [Caps Lock] *will*
cause strain and, likely, RSI since the pinky finger has to be
stretched and bent down in an extremely awkward and speed-
limiting position.
Xah Lee
often used control keys. Moreover, it seems to me that many users do
switch layout, but fail to switch keyboard and keep typing on a cheap
one, with lesser gain. Typist on typewriters did type a lot, yet it
thanks a lot for the various photos and links. It is very useful to
me, and helpful in my research in history of keyboard and layouts.
when seeing the last photo, i remember we chatted before. ☺
> Earlier comments about alleged evils of [Ctrl] being to the left
> of [A] are bogus in my experience. Attempting to use emacs on
> a PC keyboard *without* mapping [Ctrl] to [Caps Lock] *will*
> cause strain and, likely, RSI since the pinky finger has to be
> stretched and bent down in an extremely awkward and speed-
> limiting position.
i think the opinion is mostly from me. Yes we disagree.
do you have any scientific evidence for this?
the main problem i have with the opinion that making left Ctrl is good
is that:
• it's simply that way because old keyboard are like that. People get
used to it.
• such opinion almost ALWAYS came from a few vocal older generation of
programers. And always without hint of any evidence that can be
verified in some scientific way.
• when you look in online discussions, you do not see much support of
this. In particular, i've seen many who said they tried mapping Caps
Lock to Ctrl, didn't find it helpful, thus either abandoned emacs or
find other solutions.
i've written a summary account why i think swapping caps lock with
ctrl is not good in general.
• 〈Why You Should Not Swap Caps Lock With Control〉
http://xahlee.org/emacs/swap_CapsLock_Ctrl.html
also, every programer has a lot opinions about keyboarding. There
needs to be scientific qualification of their experiences to judge
their opinions. In particular, in my recent experience of close
encounter with RSI, one thing i realized is that, just exactly how
much you actually type a day makes a lot difference in what kind of
advice or method of keyboarding. See:
• 〈Emacs, RSI, My Experiences〉
http://xahlee.org/emacs/emacs_rsi_my_experiences.html
Xah
Elena
> Bingo!
>
> I've been typing since I was 6 years old using an IBM electric
> typewriter; I'm now retired (and drawing SS) so that's over 60
> years of typing.
>
> The key (no pun) is a proper ergonomic sitting position at the
> keyboard: upright in a comfortable chair with arms horizontal and
> at the level of the keyboard. Over the decades since the early
> 1960s I've typed millions of lines of code and never even had so
> much as a hint of RSI, and I'd often be at the keyboard 14+ hours
> a day. Though "retired", I still am at the keyboard for long hours
> because I still develop and program.
Twice bingo! ;-)
Indeed, it seems that some modern typists blame the layout when they
should blame their typing technique (or lack thereof). I'm amazed at
learning how many typists are very fussy about layouts and/or keyboard
models, whilst ignoring basic ergonomic expedients (straight wrists,
floating hands, etc.). As for myself, I've learned all that by
watching videos about old-time typists at work.
Elena
> the main problem i have with the opinion that making left Ctrl is good
> is that:
>
> • it's simply that way because old keyboard are like that. People get
> used to it.
>
> • such opinion almost ALWAYS came from a few vocal older generation of
> programers. And always without hint of any evidence that can be
> verified in some scientific way.
I'm not from an older generation of typists, yet I've found the Ctrl/
Caps swap to be more comfortable than other solutions. IMHO, I think
left Ctrl can't have a better position on the left side of a standard
keyboard.
As for the right Ctrl, I press it with my thumb. To make it easier to
reach, I've swapped right Ctrl and right Alt and moved the right side
of the layout one column to the right (which has the added bonus of
making the Enter key more accessible on European layouts). I've tried
applying the same expedient on the left side, but the Ctrl/Caps swap
worked better.
Xah Lee
humm. I see. I was just about to ask whether you tried the Ctrl/Alt
swap.
in ErgoEmacs keybinding, it's basically like a Ctrl and Alt swap, and
i found it to be much more useful.
... discussing this, i think it is also important to mention what
keyboard one's using. e.g. keyboard on a laptop or a full sized
keyboard. And whether the keyboard has modifiers on both sides, in
positions symmetric to each hand's index finger.
I think the capslock/ctrl swap makes much better sense on a laptop, or
keyboard that lacks right side of modifiers.
for me, i use palm or edge of fist to press left Ctrl. I think Caps
Lock and left Ctrl swap is pretty good too provided that one still
keep using the right Ctrl in standard combo key practices in touch
typing. (that is, always use both side of modifiers, and use the one
on the other side of the letter key.)
right now i set Caps Lock for Tab.
Xah
Claudius Hubig
>> Earlier comments about alleged evils of [Ctrl] being to the left
>> of [A] are bogus in my experience. Attempting to use emacs on
>> a PC keyboard *without* mapping [Ctrl] to [Caps Lock] *will*
>> cause strain and, likely, RSI since the pinky finger has to be
>> stretched and bent down in an extremely awkward and speed-
>> limiting position.
> i think the opinion is mostly from me. Yes we disagree.
I “solved” this problem altogether in my opinion by
a) Avoiding touch-typing: This has two main effects:
i) Slower typing: Not really a problem – I usually don’t
think as fast as I type (maybe I should improve on thinking…)
ii) more flexibility: I can move my hands wherever I want
them to and use the “stronger” fingers to hit most keys – the
pinky finger is used for return (sometimes).
b) A small keyboard, such as in the Thinkpad X300: It’s no problem to
press Ctrl (which is directly left of the windows key) with your left
thumb, the same goes for AltGr and your right thumb. Both of these
are arguably stronger than your pinky finger (I do observe some
strange uncomfortableness at the edge of my thumb’s nail, since I
mostly use the right side of my thumb to press the space key…).
c) Keeping the original Thinkpad-layout, i. e. “Fn” at the far left
instead of Ctrl.
Best regards,
Claudius
--
BOFH excuse #175:
OS swapped to disk
Elena
> Maltronkeyboardwith Dvoraklayout.
>
> One problem Maltron never appealed to me is because thekeyboardis
> quite ugly.
Did you know about the Maltron executive?
http://www.maltron.com/maltron-kbd-flat.html
To me it looks quite slick.
Beware that AFAIK Maltrons use the Cherry "black" switches, which are
not loved by keyboard geeks.
> Also am not sure why they have to invent the Maltronlayoutsince dvorak is already there, perhaps due to patent issues of
> the time.
Maybe. However, it's more likely that having extra keys available and
aid of computers prompted them to take the challenge. Indeed, the
Maltron layout proves to be superior to Dvorak:
http://www.maltron.com/recognitia/script.html
> Programmer Dvorak
>
> Programmer Dvorak is just like Dvorak, except that it inverted the
> number row with symbols and re-arranged the symbols.http://www.kaufmann.no/roland/dvorak/
>
> ~%7531 90246 8`
> $&[{}( =*)+] !#
>
> ',.py fgcrl /=\
> aoeui dhtns -
> ;qjkx bmwvz
You have swapped ' and ;.
Xah Lee
> On Aug 30, 12:11 pm, Xah Lee <xah...@gmail.com> wrote:
>
> > Maltronkeyboardwith Dvoraklayout.
>
> > One problem Maltron never appealed to me is because thekeyboardis
> > quite ugly.
>
> Did you know about the Maltron executive?
>
> http://www.maltron.com/maltron-kbd-flat.html
>
> To me it looks quite slick.
yeah. Though, i always thought that's kinda gimmicky. I think people
complain about Maltron's ugliness, so they came up with that. But,
creating such a flat, modern sleek looking keyboard, am guessing it
also threw out much of its ergonomic advantage.
i can't imagine feeling comfortable placing my wrists on those huge
glass panels, even though the wrist is not supposed to rest on
something when typing...
> Beware that AFAIK Maltrons use the Cherry "black" switches, which are
> not loved by keyboard geeks.
>
> > Also am not sure why they have to invent the Maltronlayoutsince dvorak is already there, perhaps due to patent issues of
> > the time.
>
> Maybe. However, it's more likely that having extra keys available and
> aid of computers prompted them to take the challenge. Indeed, the
> Maltron layout proves to be superior to Dvorak:
>
> http://www.maltron.com/recognitia/script.html
am not sure about that, especially that being the company's brochure.
> > Programmer Dvorak
>
> > Programmer Dvorak is just like Dvorak, except that it inverted the
> > number row with symbols and re-arranged the symbols.http://www.kaufmann.no/roland/dvorak/
>
> > ~%7531 90246 8`
> > $&[{}( =*)+] !#
>
> > ',.py fgcrl /=\
> > aoeui dhtns -
> > ;qjkx bmwvz
>
> You have swapped ' and ;.
oh, thanks a lot for the correction. Corrected and put credit to you.
interesting. Didn't realize that he swapped the ' and ;.
am not sure why he swapped it.
I think the qwerty's z is a easier key than qwerty's q. For
programing, i think ; happens more frequently than '.
So, his swapping is illogical to my assumptions.
(i remember reading that from dvorak's studies, the top row is better
than the bottom row, but i've never seen the actual figures that
indicate each key's ease that he might have created. In anycase, on
the pinky, i am not sure the top row is still better than bottom, and
if so, perhaps only marginally so. For me personally, the qwerty z is
certainly easier than q.)
... in general, i think almost each these layouts are created by one
invidual programer, and when looked in detail, some of their choices
don't stand to scientific scrutiny.
though, i'll have to say from what i've read, colemak is pretty good.
Xah
Elena
> yeah. Though, i always thought that's kinda gimmicky. I think people
> complain about Maltron's ugliness, so they came up with that. But,
> also threw out much of its ergonomic advantage.
>
> i can't imagine feeling comfortable placing my wrists on those huge
> glass panels, even though the wrist is not supposed to rest on
> something when typing...
Neither would I. Indeed, the Maltron Executive looks as a compromise,
however I think that as far as flat keyboard layouts go, it's the best
available.
On Sep 10, 11:27 am, Xah Lee <xah...@gmail.com> wrote:
> am not sure about that, especially that being the company's brochure.
Yes, of course there is a little marketing behind it. For instance,
some people have noticed how the picture showing bent wrists is
exaggerated. I've noticed how Dvorak's hand load is skewed: if you
hit the space bar with your left thumb - as I do - the hand load is
much more balanced. However, you can't counter the hard numbers:
Maltron has less same finger frequency, less hurdles and so on. Of
course, had been Dvorak able to move some keys under the thumbs, my
bet is that his layout would be radically different. Moreover, Shai
Coleman - the designer of Colemak - once acknowledged Maltron to be
the best alternative layout. He later penalized it because it isn't
available for standard keyboards.
> oh, thanks a lot for the correction. Corrected and put credit to you.
>
Thanks for the acknowledgement.
> interesting. Didn't realize that he swapped the ' and ;.
>
> am not sure why he swapped it.
>
> I think the qwerty's z is a easier key than qwerty's q. For
> programing, i think ; happens more frequently than '.
>
> So, his swapping is illogical to my assumptions.
Maybe here we are entering the realm of subjective issues: I find
position of Q easier to reach than Z. OTOH, I find position of <
(European's 105th key, bottom row) easier than Q. On the left side, I
find the / (bottom row) easier than P.
Or maybe he just took the thumb rule of top row being better at face
value.
> ... in general, i think almost each these layouts are created by one
> invidual programer, and when looked in detail, some of their choices
> don't stand to scientific scrutiny.
Agreed. Even worse, sometime they are designed with wrong assumptions
or wrong typing technique in mind, and lack peer reviewing.
Xah Lee
has a visual web app for layout comparison... since the month i wrote
the article, also discovered about 7 more layouts...
• 〈Dvorak, Maltron, Colemak, NEO, Bépo, Turkish-F, Keyboard Layouts
Fight!〉
http://xahlee.org/kbd/dvorak_and_all_keyboard_layouts.html
here's a summary of some findings i think i haven't posted before, but
parts may be repeat.
--------------------------------------------------
Misc Other Layouts
There are also a Dvorak layout for single left hand, and one for
single right hand. And there are also various ergonomic-oriented
layouts (inspired from the Dvorak layout) for several other european
languages. e.g. Turkish-F, Spanish, Italian, Greek, Swedish,
Norwegian, Finnish, and more. Except the Turkish-F, most seem to be
designed by a single programer. For some languages, there are
competing layouts. For links to these layouts, see: Dvorak Simplified
Keyboard.
There are also a lot personal, non-published layouts out there. Here's
one created by Claudius Hubig, at chubig.net.
Which Layout is the Most Efficient?
carPalx Study
carPalx (at http://mkweb.bcgsc.ca/carpalx/) is a site dedicated to
studying efficiency of keyboard layouts. This is the most thorough
study and quite technical. You can spend several days studying the
site.
The site uses computation and statistics to find a best model of
efficiency measurements, then evaluate the different layouts. The site
is created by a Martin Krzywinski, a Bioinformatics scientist.
According to him, the best layouts from best to worst are: QFMLWY,
QGMLWB, QGMLWY, Colemak, Dvorak, Capewell, Arensito, QWERTY. See:
〈Full Optimization - Q*MLW* Layouts 〉 http://mkweb.bcgsc.ca/carpalx/?full_optimization.
carPalx site does not seem to evaluate the Maltron layout, because the
e key is on Maltron's thumb cluster, not on a standard PC keyboard.
Also, carPalx's program does not seem to search layout spaces where
the 「, . /」 keys on qwerty are re-arranged.
PAT Or JK's Keyboard Layout Anayzer
Patrick Gillespie wrote a web app at http://patorjk.com/keyboard-layout-analyzer/.
The app is very simple and visual oriented. It takes a input text, and
computes simple distances and finger frequency among 5 layouts:
QWERTY, Dvorak, Colemak, Capewell, Arensito. Plus, it also generates a
new layout that is most optimal for the input text you used.
Though, be warned that the site only compare simple measures such as
distances, row postions, and figer ease. It does not take
consideration of ease of di-graph, hand alternation, etc.
Keyboard Evolve
Michael Capewell, author of the Capewell layout, wrote a software
Keyboard Evolve, written in C++, that uses evolutionary algorithms to
arrive at a optimal layout. michaelcapewell.com Keyboard Evolve.
Does Software Key Layout Solve Typing Problems?
No Consideration of Hardware Key Layout Problems
Keyboard layout studies on the web always assume the common PC
keyboard hardware's key layout. Apparantly, for practical reasons.
It's easy to write computer programs to compute the various
frequencies and distances of the PC keyboard's keys. It is also easy
to create a keyboard layout. But it is difficult to invent a different
physical key layout that fixes some serious PC keyboard problems.
Thus, all online layout studies simply focus on shuffling of the keys
of the PC keyboard.
(See: Computer Keyboards Gallery.)
The PC keyboard physical-keys layout has several major flaws with
respect to ergonomics. For example, the Enter, Tab, Backspace, are
most frequently used keys but requires the pinky to press. In
particular, the Tab, Enter, right Shift, Backspace, keys requires the
pinky to strech over 1 or more keys.
You might think that these keys are not used that often. But most
keyboarding activities are not straight input of letter characters,
especially in computer programing. Most of the time you are editing.
Your use editing keys such as arrow keys, Backspace, Enter, Tab, more
often than some of the letter keys. From statistical study of emacs
commands, the percentage of editing related key presses is 48% of all
key presses!
(See: Emacs's Command Frequency)
Another problem of PC keyboard's physical-keys layout is that the
right hand has 5 more keys letter/symbol keys than the left, and all
of them are pressed by stretched pinky. Also, the key alignment are un-
naturally jagged.
See: Keyboard Hardware Design Flaws.
From my recent experience, if you just type the Enter key and Tab keys
all day, just these two keys and nothing else, on a standard PC
keyboard, with hands at touch-typing positions, say do it once every 5
seconds, for 8 hours, you'll get RSI fast. (See: Left Wrist side-to-
side Motion Pain; vi Esc key Syndrome.)
No Consideration of Number Arrangement or Symbols
Another major problem is the very inefficient arrangement of the
number keys, as well as programer's need to type some of the symbols.
Most alternative layouts still stick to “1 2 3 4 ...” arrangement.
( Kinesis keyboard arguably fixed the problem by using a left thumb
modifier that turns the right hand keys into a number pad.)
Is Improvement Over Dvorak Necessary?
Just How Much Do You Type?
The science of ergonomics assumes that the activity is prolonged. For
example, if you just type less than 1 hour a day, like vast majority
of people, qwerty is perfectly fine.
If you are a professional coder or writer, who's hands are actively
typing perhaps 3 hours a day, then Dvorak will be significantly
important to you.
If you often have coding or writing sessions that makes you work 70-
hour weeks, or worse, if you are a professional data entry clerk and
you type continuously for 8 hours a day, then a serious ergonomic
keyboard such as Kinesis or Maltron is critical.
When you consider designing the perfect layout to improve Dvorak for
English language, it's important to consider practical values of such
layout. If you created a most efficient layout, but it's for standard
PC keyboard, you are likely to get RSI if you need to type
continuously for several hours a day.
Suppose you are using a Dvorak variant that improves efficiency by
let's say 5%. How many more hours of continuous typing per day will
the 5% improvement have a impact?
What About Keybinding Efficiency?
vi users often develop RSI due to Esc key Syndrome (or, Left Wrist
side-to-side Motion Pain). Emacs users may develope the Emacs Pinky
Syndrome. Both do not have much to do with the keyboard layout. So, if
your heavy typing is in vi or emacs, you might consider the efficiency
of your keybinding layout. See:
Why Emacs's Keyboard Shortcuts Are Painful
ErgoEmacs Keybinding
What About a Reasonable Efficient Standard Layout for All Languages?
Another common problem is for international users, of non-English
languages. For example, German, Spanish, French, and even Chinese and
Japanese can benefit because their input methods commonly rely on
Latin alphabet. (See: Chinese Pinyin Letter Frequency and Dvorak
Layout.)
In these languages, usually there are few extra characters that needs
to be typed. There are many standardized layouts for them (e.g.
QWERTZ, AZERTY), but often they still requires you to type the special
chars by a combination of key press using AltGr modifier, and these
layout usually do not consider any ergonomics in the sense of dvorak.
(see: Idiocy Of Keyboard Layouts.)
It's much better to find ways to create a universal layout that are
largely efficient, fixes the hardware layout problem, fixes the number
arrangement problem, and can be used for all languages. I think this
is quite doable. Neo, Bépo, and i think br-Nativo already claim to be
general for inputing all eruo langs, except themain alphabet keys are
arranged for particular language.
Xah ∑ xahlee.org ☄