I've created something similar to Omarchy and I've been using it and improving it in the last 10 years

[account1009]

Only some parts of what I've done are public.
I've never published the whole idea because I think that is even more opinionated than your project and some radical parts about improved Dvorak keyboard are not for everyone.

I've discovered today about Omarchy and reading on your manual I think that you share really similar goals to what I've done!

The ideas of what I have created and the goals are:

• switch or open the apps I use with super keyboard (left hand) + app associated letter (right hand)
• every app has to have the same copy and paste shortcuts (yes, including terminal)
• every app has to have the same shortcuts to open, close tabs, go to right tab, go to left tab
• every app has to have the same shortcuts involving arrows, alt, control and delete such as go to beginning of line, end of line, previous word, next word, delete line, delete word
• kinetic scrolling has to work in every app with the same calibration
• the super key is for global navigation shortcuts and control key is for in app shortcuts
• only left hand is for super and control while the right hand is for shortcuts (superfast from Dvorak keyboard basic ideas)
• super key + scrollpad has to globally zoom moving up and down

I've been using a MacBook Air for years and reaching these goals was relatively easy with Hammerspoon scripting to keep every apps open. On MacOS I was using only opensource using app compatible with Linux and was preparing for the big jump to Linux. I was only missing a great hardware to run Linux.

At the beginning of this year with Asahi project mature I've moved all my ideas to Linux on the great Mac hardware. Let say that I didn't like the inconsistency in border radius that is forced on MacOS apps even when they are using the whole screen.

To switch from app to apps I've a customized version of Gnome + Run or raise (never published configs)

I've patched Deperto to have fluid global zoom even on hires linux screen (fix is upstream): https://extensions.gnome.org/extension/9256/deperto-zoom-by-scroll/

I've added custom keyboard shortcuts to Nautilus (never published code) to reach consistency in the shortcuts related to tabs, open, rename. it is possible with python scripting but using not documented stuff. Let me know if you are interested.

The super key was not working on Firefox Linux for shortcuts but now it is fixed: https://bugzilla.mozilla.org/show_bug.cgi?id=2012218
I've proposed a simple patch in Firefox to make move to right and left tab customizable: https://bugzilla.mozilla.org/show_bug.cgi?id=2012210 but that idea still needs a cascade of refactoring and bureaucracy to go officially upsteam

I've created a WebKitGTK based browser (never published) to also have consistency with control,alt+arrows to navigate text. But maintaining a browser is too much and I still have to find a way to contribute mainline on Firefox for this

I also use Ghostty as you do and I've customized it and zsh to have consistency with control+c, interruptions, etc. I think that we have done similar stuff reading your doc

The differences: I'm against Signal because they are anti-linux (expecially mobile linux) and I'm contributing to Deltachat instead.
I use Zed instead of Neovim and I think that solving this is important for the scroll consistency: zed-industries/zed#14473

Maybe I've forgotten something...
but it looks like we share similar goal @dhh Do you agree?
Do you have in mind if we can collaborate in the intersection of the two projects? (I think for example about making apps shortcuts customizzable or kinetic scrolling that seems good for both of us)

Intrusive question: now that you have huge amount of money that has been donated for the project do you think to use a part to make external projects more Linux frendly?

I mean: Chromium for example have added kinetic scroll on day one for their ChromeOS but have been lagging the feature for more than 10 years for Linux. Chromium is postponing out of experimental Pipewire integration with no motivated reasons.
It looks that they don't care at all about making Linux cool

As you can noticed by my linked Firefox issues it looks that the Firefox main developing team does not even have a Linux native machine to test features

I try to reformulate: do you have the idea of throw some of the money to fix upstream (not with forks) these issues that are now not a priority for the projects?
Or to convince Signal to be a little less anti-Linux? (they are even against arm linux and mobile linux but that's another battle)

I think that we have similar goals (maybe 80% of the goals) but my focus is in avoiding forks and in collaborating with upstream projects.
Do you share this mainline first focus?

If you have any other ideas to collaborate let me know!

Enrico