I don't know about emacs but for vim I feel that'd be quite counterproductive as the key bindings are mnemonically based (p for paste etc). If you really want to use the QWERTY positions for all commands then I have a feeling you might be better off just using QWERTY for that editor then.

Short of that, it'd be a major hassle to perform all the necessary remaps (I don't think you could cheat and remap the keys completely because then you'd be back to QWERTY when entering text as well I think?). One thing you could do in vim would be to remap all the keys in all modes except i(nsert) – but I'm still uncertain as to whether that'd work for g and z commands?

What I do instead is Extend mappings (see my sig topic). Then I don't have to remap anything in vim and have the navigation keys with me regardless of application.

Thank you guys for the help.  Going to attempt various solutions this week.

Lalop: May I ask why you think it's a terrible idea?  I feel we probably just won't see eye to eye on the subject, but I'm interested in why one might think it's not a great idea (beyond it being a slightly time-consuming endeavor).

It forces two different paradigms into the same key.  Is this the qwerty key, or the colemak key?  If the shortcuts alone are bound to be qwerty-like, that could change almost immediately.  It's probably even worse in emacs, where the macros get a lot more complicated.  C-o C-r k whatever.  Unless you irreducibly think of all of them as a motion (not likely), it will become confusing to have to change which layout is which on the fly.

@DreymaR:  Right, the idea of breaking something is half the reason I posted the question.  It would seem venturing to mess with bindings much at all runs this risk.

@both of you:  I'm not sure I understand your guys' belief that I would be stumped on other peoples' machines or that I would need to switch layouts on the fly.  The whole point would be so that I would more easily be able to use other peoples' machines while never having to change the layout at all.

Using the shortcuts quickly becomes a matter of muscle memory.  Were I to make the changes I desire, then my muscle memory would serve me whether I was on my own client typing in Colemak, or someone else's client typing in QWERTY.  There would be no on-the-fly switching of layouts.  The only problem I can see is when first learning a keystroke command (or on the rare occasion where I would remember the letter associated with a command but not its motion).  In this case, I would have to simply learn the key position by looking at my QWERTY keyboard and then practicing the motion, still using the Colemak layout.

I understand it may seem unusual to think of the commands simply as motions, but that's how I generally work.  I mean, in the terminal you have CTRL commands for A, E, K, U, C, D, Y, W, and I'd assume some others as well.  Of the 8, I know only Kill and Yank thanks to an emacs tutorial, and I don't even think of their functions in terms of their names.  **Edit** Also, don't Shai's remappings and Evil kind of work to destroy some mnemonics?

Anyway, I once again very much appreciate the resources you guys have supplied.  I'm sorry that I'm being a little argumentative, I just don't understand a couple of your comments.

Learning by going through the motions, sounds harder that referencing the character shortcut in your head.  But perhaps that's just me.

I toyed with the idea of keeping the Qwerty shortcuts with Vim, but as soon as you are confident with a layout, you can just type the shortcuts out, and in time they'll be confined to muscle memory anyway.  I feel it would break my brain any other way.  The hard part is just remembering all the shortcuts in the first place.

Different strokes for different folks!