The effort to learn different layouts is certainly not the same! Like user wibble once said, the commotion you cause to your "keyboard gestalt feeling" by changing anything might be the same pretty much regardless of what you change. But like it was pointed out back then, how much you change and how drastically greatly impacts how fast you recover!

When I learnt the bottom-left-row ZXCVB shift it only took me a couple of days before full typing comfort was regained because it only changed the position of 6 keys and only one rarely used one changed fingering. Based on that experience, I guess you could learn Minimak in a matter of days too. For all the good it will do you...

Good point about Minimax not improving balance (hand, finger, what-have-you) either.

BTW: It's funny that Arensito still has such a good position in people's sentiments. That layout was developed for typing on a Kinesis in Norwegian and puts æ and å on important keys! Also, it changes everything around and looks like a mess. How many users has it got these days, I wonder?

(To answer my own question in a way: If Google hits are an indicator of anything, there's ~1,000 for 'Arensito layout', ~2,500 for 'CarpalX layout', ~79,000 for 'Colemak layout' and ~1,480,000 for 'Dvorak layout'. Heh. I had to add the 'layout' term to the searches to distinguish the Dvorak hits from the composer. Adding 'keyboard layout' led to different results again.)

Basically I'm rehashing what has been said already, but here goes.

While I like the concept of minimal switching, switching is still switching, and frankly the changes aren't enough to make it worthwhile. As DreymaR noted, I've stated before that fingering can fix a lot of the distance issues in Qwerty without bothering with layouts. The E position isn't actually a horrible place to rest your L.Mid, so it is effectively home row already. Likewise, on the left hand IO and even P (for left shifters) aren't horrible home positions either. On this 'Minimak', the same finger ratio is unchanged from Qwerty; the T is unforgivably still not a home position; the new U placement, while it isn't bad (decades of hitting the Qwerty N is perhaps coloring my opinion), is still particularly unkind when typing YOU; there are no non-Windows solutions and the layout doesn't even bother to incorporate the relatively simple and painless Wide-mod.

On a side note, I'm not sure the Qwerty L.Hand bias is as much of a problem as we make it out to be, even for right handers. Granted I've never particularly noticed as I've been more-or-less touch typing since I was... like eight? Anyways, considering the amount of R.Hand mousing most of us do, I expect it balances out somewhat. I've heard complaints by Dvorakers that they had to switch to left hand mousing for similar reasons.

AND once again, a quick round of what I like to call "Defending arensito from DreymaR". I strongly suspect arensito's popularity is due to it being one of the few layouts to advocate alternate home positions (awkward beginnings aside, it does actually fix the L.Hand angle), use of the thumbs and significant use of a more accessible AltGr layer. It is basically like a demonstration of what efficient use of an ANSI keyboard should look like, and I suspect that the majority of its hit count is earned from academics referencing it in their dissertations. Personally the only gripes I have about it are the placement of the numbers in the AltGr layer, and the lack of ports to non-Linux operating systems.

If we are going to discuss layouts that are still oddly popular, how about Capewell? That one never made a lot of sense to me, though I do appreciate the placement of the semi-colon.

I'm the author of the original Minimak post.

Sorry for getting booted by the spam filter.  Was just excited to share my work and the board only lets you know that it's taking action *after* it lets you make the posts.  If I'd been told while making them, I wouldn't have put my announcement on more than one post.

I also had difficulty registering (the visual captcha didn't work on my browser), so I decided to leave well enough alone.

I just wanted to note that I've reworked the layout several times over, and the final result is posted to my website (http://www.minimak.org/).  After reading a lot of what is posted in these forums, I don't think you'll necessarily be all that impressed, as the target user is a different audience than you'll find here.  But I'll be interested in any feedback in any case.  The intended audience are people who aren't willing to go to the effort of learning a new full layout such as Colemak or Dvorak.  Minimak fosuses exclusively on improvements that can be made to QWERTY while sacrificing a minimum to the learning curve.  In every incarnation of the layout, only 3 keys switch fingers.  For metrics I look at finger distance and same-finger repetition.  Trying to squeeze other optimizations in with so few key changes is an exercise in futility, although I did spend a lot of time with considerations like digraphs and same-hand row jumps.  The final layout exhibits better behavior than QWERTY on those fronts as well, but only as a side-effect of the main design philosophy.  So no attempt to quantify them is made in the results.

Also, the graphic that was linked to by your posts here is gone or modified, so sorry if any of your prior posts which make reference to it seem curiously off.