Edit (Sun Jan 6 07:30:20 CST 2013): I am currently using Colemak and satisfied by it. What I thought were big defects of Colemak turned out to be mere small trivial imperfections that are negligible and meaningless, specially when you compare them to the actual problems other layouts have. I think probably a mathematician could be able to mathematically prove that Colemak is the optimal, or very close to the optimal, solution to improving the typing experience while having some niceties like mantaining zxcv in place, among others.
(Of course all this applies only to English, it might be that Colemak is actually very adecuate for typing other languages like Norwegian or other languages)
The Colemak layout "is designed for efficient and ergonomic touch typing in English.
Learning Colemak is a one-time investment that will allow you to enjoy faster and pain-free typing for the rest of your life"
It is also quick to learn it, but....
Is it also quick and easy to become profficient and productive in Colemak?
If the following hypotheses turn out to be true, then they might mean it is hard to become profficient and productive in Colemak, not impossible maybe, but quite expensive in time and effort:
1. Humans would have a tendency to rest their fingers on the keys of the home row when the most frecuent letters in english are in that row, as in the Colemak layout.
If this is true then it would mean that when users use Colemak they unconsciously rest their fingers on the home row and then they'd need to stretch to reach for the two central columns, and if frecuently used letters are also in those positions then that could cause strain to the hands on the long term.
Users could float their hands pianist-like but that would require a conscious effort.
(It would be more natural for pianists when playing, and qwerty users when typing, to float their hands because, pianists are half expecting a musical scale to appear anytime and are prepared for it, and qwerty users would naturally float their hands because qwerty doesn't have the most popular letters on the home row and so hands have more movement and users would unconsciously and naturally would float their hands.)
2. If two keys that are 1 or 2 positions away one of the other horizontally on the old layout (which the user already masters) appear in new positions on the new layout but are now in inverse order along the horizontal axis and on the same row (example S is to the left of R in qwerty and it is now to the right of R in Colemak) then that would cause the user to tend to confuse the two keys. (until the motor memory of the old layout is overcome)
What do you think? Are they true, false, or are actual measurements with people necessar
