Improving Colemak

The layout comparison reported in the popular alternatives demonstrated the virtues of Colemak. The Colemak layout has the lowest total effort out of all the popular alternatives, owing to an excellent combination of low penalty and stroke efforts. If you've been following the case studies, and specifically the full optimization the component optimization sections, you noticed that generating a layout that improves on each effort component of Colemak isn't very easy &mdash component weights need to be tweaked. In this section, I report three layouts that that improve every component of Colemak's effort.

Colemak has the lowest total effort of all the popular alternatives described here. Although its base effort component isn't the lowest, it ranks highly in both penalty and stroke components. The full optimization layout (QGMLWB) has a lower total effort (1.668 vs 1.842) than Colemak and it achieves this by forcibly lowering the penalty at the cost of an increased base effort component.

Figure 1. The Colemak layout.

While the QGMLWB distinguishes itself by having the lowest effort, unseating the previous champion Colemak by 9.4%, it doesn't improve on Colemak across all characteristics. In order to find a layout that does, the simulation parameters needed to be adjusted from a base:penalty:stroke ratio of 1:1:1 to a ratio of 10:1:10. In other words, in order to find a lower base effort than Colemak, some of the gains in penalty effort need to be sacrificed.

Improving Colemak - ZXCV moved

If we allow for the relocation of ZXCV, then the PBFMWJ is the layout that most greatly improves over Colemak. This effort lowers Colemak's total effort by 5.6% and individually improves every component. The largest improvement is in the penalty component (11.5%), with a smaller base (3.5%) improvement and negligible stroke path improvement (0.5%).

Figure 2. The best improvement over Colemak with relocated ZXCV.

Colemak improvement
model	keyboard	total effortrel%	effort contributionsrel%
			base	penalties	path
mod_01	colemak	1.842	0.34418.7	0.76341.4 R0.158 F0.487	0.73539.9
	colemak-imp-01	1.738 (-5.6)	0.33219.1 (-3.5)	0.67538.8 (-11.5) R0.148 (-6.3) F0.462 (-5.1)	0.73142.1 (-0.5)

Improving Colemak - ZXCV stationary

If we require that ZXCV are held stationary, as in Colemak, then the best improved layout is GYLMWP, which lowers the effort by 4.7%. The profile of improvement is different than seen in the layout reported above. Here, a very large row-based penalty improvement is seen (12%) and a significant stroke path improvement (2.7%). The base effort and finger-based penalty improvements are minor, though, at 0.9% and 1.0%, respectively.

Figure 3. The best improvement over Colemak with stationary ZXCV.

Colemak improvement
model	keyboard	total effortrel%	effort contributionsrel%
			base	penalties	path
mod_01	colemak	1.842	0.34418.7	0.76341.4 R0.158 F0.487	0.73539.9
	colemak-imp-02	1.756 (-4.7)	0.34119.4 (-0.9)	0.70039.9 (-8.3) R0.139 (-12.0) F0.482 (-1.0)	0.71540.7 (-2.7)

Improving Colemak - Detailed Statistics

Detailed statistics of the Colemak improvements are shown below. Recall that the first improvement migrates ZXCV, whereas the second keeps these keys as stationary.

Home row use is the same as Colemak, with decreased use of the bottom row (recall that the bottom row is penalized more than the top row by the row-based penalty). Hand asymmetry is very low, at 0.01 and -0.02 for these two improved layouts. This is lower than Colemak, which is unbalanced by 6% in favour of the right hand. While Colemak has a higher hand asymmetry, it has a marginally better finger run length (93% of key strokes pairs use different fingers, as opposed to 92% and 91% for the two proposed improvements).

carpalx effort optimization	keyboard name	statistics					effort
		rowh	rowb	hand asym	finger freq	cumulative run distribution	mod_01
colemak improvement 1	colemak-imp-01 PBFMWJLUY;[]\ SNATDHOEIR' ZVGCQXK,./	0.74	0.07	0.01	0.14 0.18 0.28 0.40	1 2 3 4 5 6 7 8 9 rhl 0.50 0.80 0.91 0.96 0.98 0.99 1 1 1 rhr 0.51 0.81 0.92 0.97 0.99 1 1 1 1 rh 0.50 0.80 0.92 0.97 0.99 0.99 1 1 1 rrt 0.76 0.95 0.99 1 1 1 1 1 1 rrh 0.26 0.46 0.60 0.71 0.79 0.85 0.89 0.92 0.94 rrb 0.95 1 1 1 1 1 1 1 1 rr 0.55 0.72 0.81 0.86 0.90 0.93 0.95 0.96 0.97 rf 0.92 0.99 1 1 1 1 1 1 1 rj 0.83 0.96 0.99 1 1 1 1 1 1	1.738 0.33219.1% 0.67538.8% 0.73142.1%
colemak improvement 2	colemak-imp-02 GYLMWPFUB;[]\ RSNTDHAEOI' ZXCVQJK,./	0.74	0.05	-0.02	0.15 0.18 0.28 0.39	1 2 3 4 5 6 7 8 9 rhl 0.57 0.88 0.98 1 1 1 1 1 1 rhr 0.57 0.87 0.95 0.99 1 1 1 1 1 rh 0.57 0.88 0.96 0.99 1 1 1 1 1 rrt 0.75 0.95 0.99 1 1 1 1 1 1 rrh 0.26 0.46 0.60 0.71 0.79 0.85 0.89 0.92 0.94 rrb 0.94 1 1 1 1 1 1 1 1 rr 0.53 0.72 0.80 0.86 0.90 0.93 0.95 0.96 0.97 rf 0.91 1 1 1 1 1 1 1 1 rj 0.84 0.97 0.99 1 1 1 1 1 1	1.756 0.34119.4% 0.70039.9% 0.71540.7%
none	Colemak QWFPGJLUY;[]\ ARSTDHNEIO' ZXCVBKM,./	0.74	0.09	-0.06	0.16 0.17 0.26 0.41	1 2 3 4 5 6 7 8 9 rhl 0.55 0.85 0.94 0.98 0.99 1 1 1 1 rhr 0.49 0.77 0.90 0.96 0.98 0.99 1 1 1 rh 0.52 0.81 0.92 0.97 0.99 0.99 1 1 1 rrt 0.79 0.97 0.99 1 1 1 1 1 1 rrh 0.26 0.46 0.60 0.71 0.79 0.85 0.89 0.92 0.94 rrb 0.95 1 1 1 1 1 1 1 1 rr 0.57 0.73 0.81 0.86 0.90 0.93 0.95 0.96 0.97 rf 0.93 1 1 1 1 1 1 1 1 rj 0.83 0.95 0.99 1 1 1 1 1 1	1.842 0.34418.7% 0.76341.4% 0.73539.9%