Xah Lee, 2008-07
On the issue of keymaping, modifier keys, ergonomics... part of the problem is the keyboard hardware itself.
The standard keyboard out there used by some 99% of computers worldwide, namely the PC keyboard; is derived from the design of type writers.
The design of the typewriter itself, is largely concerned about getting the machine to actually work. Like most inventions, in the beginning the concern is just to get it to work. The concept of keyboarding ergonomics didn't come about or become popular after few decades of keyboard use.
Let's consider some examples:
• The Backspace/Delete key, the Enter/Return key, are among the most useful special keys. However, they are placed in the most inconvenient spots.
• The vertical key column positions are jagged. i.e. The columns 1QAS, 2WSX, 3EDC, etc, are slanted. Worse is that the jagging is not a regular, not symmetric for both hands, but in a ad hoc slant from top left to bottom right. So, for example, going from key D to E, your middle finger moves upwards and in the direction of your pinky. Going from K to I, your right middle finger also moves upward but in the direction of your thumb.
• The number of keys for the Left and right are not symmetric.
` 12345 67890 -=
QWERT YUIOP []\
ASDFG HJKL; '
ZXCVB NM,./
Notice in the above pict, how the right side has lots of extra keys, operated by your stretched pinky.
So, all of the above combined, makes today's conventional keyboard really a bad piece of hardware.
One keyboard that is well known and loved, is the Kinesis (keyboard), which fixed all of the above problems.
For keyboard gallery with photos and more detailed commentary, see: Computer Keyboards Gallery.
Now, when we consider the placement of modifier keys, or consider keymapping, or consider optimal keybindings for emacs, the given PC keyboard hardware puts a lot constraint.
* * *
There are many variety of input devices in the past 2 decades, some intended to replace the keyboard. (few of them based on “chording”. See Chording keyboard ) Some of them are good in principle but maybe bad in practice. Perhaps some of them are truely good. None of them actually caught on. (except the Kenesis in some degree) The main problem is getting people to change (if there's such a need at all). The PC keyboard, although its fairly bad when considered ergonomically, but it works ok. It is widespread, and most people who has to use a keyboard, don't need to type that much. Vast majority of computer users today, use computer to read check/write email, instant chat, read news, watch video, play music, play games. Typing does not happen that much. Even for programers, perhaps a majority don't need to do continuous, intensive, typing in prolonged hours. They don't want to change, partly because they don't need to. As a fact, a lot professional programers who code in their day job, do not even touch type.
The problem is similar for acceptance of the Dvorak layout. Major change is hard to get adapted. Major change is costy.
Small change or gradual change do happen. For example, most keyboard sold today has several special buttons that launch apps or control music players. The split ergonomic keyboards has also been widely adapted. Mouse has become universal since mid 1995, and Mouse wheel too, since Microsoft introduced/popularized it in late 1990s.
The Ctrl and Alt keys, on both sides of keyboard, symmetrically distanced to the thumbs when hands rests in a touch-typing position, are on all Microsoft's ergonomic keyboards, and almost all PC keyboards today (only notable exception is Apple's keyboards).
On the key layout side, the Dvorak is now available on all major operating systems, and there are gradually more programers using it.
Related essays: