I’m an average typist by nature in the first place, around 80wpm according to typeracer.com, now about the same with Minimak and didn’t get any faster with QGMLWB either. I like challenging myself and wanted to see what it would be like to try a new layout. I honestly didn’t know if my fingers would be capable after 20+ years of QWERTY!

I’d say the primary benefit rather than speed would just be reduced finger motion compared to QWERTY. I loved the feel of Carpal-x fully-optimized; it felt like hardly moving. Minimak is somewhat similar to that feel, although probably not as aesthetically pleasing because it doesn’t optimize for all the things Carpal-x takes into account. We’ll see in time as it’s still new even to me by comparison. I’d imagine Colemak feels similar, going by your description.

The whole experience just emphasized to me how hard it is to learn a new layout, and how important it is to take that into account in layout design. It was also hard to maintain QWERTY (although Colemak surely fairs better). While I came to believe in the principles behind Colemak, I thought there was still a way to take it further.

Since I only came up with the layout a month ago, the jury is definitely still out, but I hope it can be helpful to someone else as well.

This is the most epic comment I’ve ever received since starting to write for MUO. Have looked at your site, and the layout looks really interesting.

What typing speeds did you get with it? Do you have a rough estimate of how many users are using this layout?

Unfortunately, I found that learning a new layout was more painful than reviving my suppressed QGMLWB skills. However, having a basis to compare all three layouts taught me something important.

I realized that the greatest difficulties I had with learning new keys were those that moved off of the same finger, and those that moved in relation to the other keys around them.

I figured out that it I could move a much greater number of keys if a) they all moved together and b) they stayed with the same fingers. What I found is that this approach valued the flow of my typing and the finger impulses I had. I only had to modify the aim, not the entire finger or hand, and this resulted in less thinking and greater accuracy.

The simplest example of this would be to switch the entire home row with the upper row of the keyboard. I’m using this just as an example, although according to the stats, this is actually an improvement for home-row typing percentage.

If you make that switch, you’ve changed over 2/3 of the keys on the keyboard. However, you’ll quickly find that this layout is *easier* to learn than any other. After an initial period of confusion, your brain snaps into the switch and gets it that it’s just a row transposition, which is a simpler transform than swapping individual keys. It doesn’t have to figure out individual key transpositions after that.

After making that realization, I decided to use that as the guiding principle for a new layout, one that achieves the best possible home-row frequency while making transpositions in chunks, and one that honors finger impulses where it doesn’t. I wrote up my results as a website: http://minimak.org/. After using it for a couple days, I was able to achieve more progress than I had with either QGMLWB or Colemak in the same amount of time. I’m interested to hear other people’s experiences, if you’re willing to try.

It shares a tiny bit with Colemak in that it moves substantially fewer keys than most layouts, but it’s more like the Asset layout in its goal. It resembles neither when you look at it though. Take a look and see what you think.

What I find to be as important as Colemak itself is another nice feature of the Portable Keyboard Layout program: The Extend mode! Instead of remapping the CapsLock to Backspace you make it a modifier to get not just Backspace but arrows and all sorts of useful keys right at your home position. Very useful for both productivity and comfort.

I also use an ergonomic mod called Angle/Wide that leads to straight wrists which is important, and some finger load rebalancing. More for the interested that one, but I recommend it at least.

Also, have you tried speech recognition software? I was impressed with Dragon when I tried it. Fairly expensive, but it worked amazingly well.

I think that Colemak may help, since less distance travelled means less work for the fingers. I just need the time and dedication to start using it.