Retested today. Up to 55 WPM, 100% accuracy, 2.8% errors. Feel like I'm progressing quite quickly now that I'm not changing any keys.

Yep, I did mean 2013 :o

The transitional layouts worked really well for me because of the following:

* High bang for buck in the initial stages.
* Low risk - still a worthwhile endeavor if I get stuck at a stage for a year.
* Allowed me to concentrate only on a few keys at a time, and mostly master them before moving on.
* Low productivity loss
* The only keys that are moved twice are infrequently used. By moving on to the later stages after a shortish time, their new position had barely been set in my mind.

I'm actually a bit sad now that there aren't any magical new stages... Oh well, I guess I will have to play with some macros now :P

Thanks again. It really is quite genius... The hardest step was the Colemak step as I had made a lot of changes already and the RS keys were a bitch!

No XKB for the Mac. I may try something similar using other utilities though. I have already been trying to get the Extend mod working on work (we're all PCs at work). Unfortunately my caps is already bound to ctrl in the registry. I'm also not sure that I want to give control up. I also don't feel like rebooting. I have a Kinesis Advantage which already puts a lot of other keys in very acceptable positions. That said, I like the idea of using the number keys as function keys because the Kinesis function keys suck...

@Lalop: Thanks for the Reddit plug! :)

Quick update: 22/05/2013 66-69 WPM 3-4.3% error rate

Yep, really happy with it. Those speeds are over 10 minute sessions in Master Key, so it's not just burst speeds. I'm working on slowing things down and improving accuracy from here.

Thanks again for Tarmak. I don't think I could've switched without it!

I started my Tarmak project on Friday the 13th (nice omen!), June, 2014, with a plan to step up each Friday after work.  I made the final transition on Monday, 14 July, as I took a long weekend mini-vacation that weekend.  It's been just over a week and I've timed out in Amphetype twice at over 50 wpm / 97% accuracy, though my average is still only in the 40's.  Each week I was able to fairly easily climb back from the instant "hit" to reliably over 40 wpm / 97% accuracy before the next step.  I've hit my goal already for July, and I'm optimistic about my odds of hitting my "final" goal of 70 wpm by Halloween.

I typed on Qwerty for over 30 years.  DreymaR's Tarmak method finally gave me the gradual methodology I needed to move to a better keyboard layout.  Although Dvorak may be better established, and other new-comers may be technically (marginally) superior, Colemak represents to me the sweet spot.  Better than Dvorak and only (IMO) insignificantly improved upon by others, Colemak keeps my geek CTRL-keys where they belong (all the ones I regularly use, anyway), it's better supported than ever before (even rivaling Dvorak), and (for me the final piece of the puzzle) the Tarmak transition method puts it truly within reach of people who actually have to keep getting work done while working on the transition.

Thank you again, Shai Coleman, DreymaR, ghaz and lalop for all the contributions along the way that made this possible!

That's good progress ghaz. What was your wpm when you started? I'm just getting started with this and I hope to get to 100 wpm soon.

Threads in this forum aren't generally hyperactive. The "dead" term usually doesn't quite work here. And why would you ask whether GHen has left?

Ghaz happily types Colemak still, as far as I know. He's only moderately active in this forum (which would count as pretty inactive in some other forums, I guess).

He's a contributor, having made a Russian Colemak (Rulemak) and having done development on MessagEase layouts. Great guy! :-)

Indeed, I'm confused. I was thinking of Ghen, being a big hasty and preoccupied. Sorry, Ghaz!