[HN Gopher] ShannonMax: A Library to Optimize Emacs Keybindings ...
___________________________________________________________________
ShannonMax: A Library to Optimize Emacs Keybindings with
Information Theory
Author : sammy0910
Score : 63 points
Date : 2026-02-19 11:15 UTC (11 hours ago)
(HTM) web link (github.com)
(TXT) w3m dump (github.com)
| lorenzohess wrote:
| This looks great. Would there be an easy way to generalize this
| program to tiling window managers? Maybe initially I can use this
| by modifying the WM to forward all its keybindings to a dummy
| Emacs instance. For WMs is the entropy theory also applicable?
| sammy0910 wrote:
| it should be -- as long as you have like the right logging set
| up, I think the theory would also be applicable.
|
| currently the calculations in this library are done with a
| clojure jar, so if you're interested, you might have an easier
| time calling that directly
| oritron wrote:
| Some people use Emacs /as a tiling window manager/ :)
| https://github.com/emacs-exwm/exwm
| Jeff_Brown wrote:
| Something I love about emacs is the ability to tab complete the
| name of a command. I do know a lot of keyboard shortcuts, but I
| use way, way more commands than I know the shortcut for. Need to
| rename a buffer? M-x ren-buf TAB should do it. Etc.
| setopt wrote:
| Me to, but to be fair, I think this is no longer unique to
| Emacs. See for example the "command palette" in VSCode; it
| isn't "tab completion" per se but similar to e.g. M-x with
| Vertico.
| goodmythical wrote:
| I was thinking I was crazy...I use command completion in lots
| of different applications...
| BeetleB wrote:
| Probably he's referring to "fuzzy find"?
|
| Yes, VSCode has something similar, I believe. But Emacs had
| it before VSCode existed ;-)
| tra3 wrote:
| Are there any similar tools for the OS as a whole?
|
| I'm trying to switch to Corne keyboards and the key maps are
| critical.
| chills wrote:
| What do you mean, "the OS as a whole"? This is the OS!
| Pay08 wrote:
| This is theoretically pretty extensible, but you need an OS-
| wide keylogger. On the Linux side, I'm not sure if Wayland
| allows for that at all.
| aghilmort wrote:
| this is great / will try!
| snikeris wrote:
| This is cool.
|
| While we're discussing optimizing emacs keybindings...I've found
| it key to have my bindings set up such that my thumbs operate the
| control modifier key.
| kleiba wrote:
| I'm fine with the standard CAPS_LOCK is CTRL setup...
| Pay08 wrote:
| I could never get used to that. I should probably try forcing
| the issue to see if I can rewire my muscle memory, but I'm
| afraid that it'll be a problem in places where I don't want
| caps lock rebound to ctrl.
| setopt wrote:
| I got a pretty bad case of RSI with that setup, since it
| encourages one-handed chording (e.g. pressing C-x C-s by
| holding down your pinkie on Caps Lock while twisting your
| wrist to tap X then S using other fingers on the same hand).
| It's far more ergonomic to do two-handed chording, where you
| press one key at a time with each hand to the extent
| possible. For me, that meant using Karabiner Element (Mac)
| and Keyd (Linux) to map Return to another Ctrl key when held
| down (in addition to the Caps as Ctrl mapping). Then I can
| simply hold down Return with my right hand and tap X then S
| with whatever fingers feel natural on the left hand, without
| twisting my wrist at all.
| BeetleB wrote:
| Indeed. I had RSI issues very early in my career, and the
| standard advice by ergonomists was "Use both hands when
| doing any multi-key sequence". If you're doing Ctrl-C, use
| the _right_ Ctrl button, and so on.
___________________________________________________________________
(page generated 2026-02-19 23:01 UTC)