Omarchy Is Revolutionary for Linux: and It's Not for Me

· 8 min read

I tried Omarchy. I am not staying. And I still think it is the most important thing to happen to the Linux desktop in years.

All three of those statements are true at the same time, and this post is me holding them without letting one eat the others.

Let me get the correction in before anyone sends it: this is not a review from a Hyprland native. I am a KDE guy with twenty years of floating windows and a mouse in my hand. I ran Omarchy long enough to understand what it is and why it matters, not long enough to rewire my brain around it. So take this for what it is; a field note from someone the product was never really for, who still thinks the product is revolutionary.

What Omarchy actually is

Omarchy is David Heinemeier Hansson’s, DHH, Rails, Basecamp, opinionated take on the Linux desktop. The word he uses is omakase: chef’s choice. You trust the chef, you get the full course, and you can send anything back.

The stack, concretely:

  • Arch Linux underneath, rolling release.
  • Hyprland as the Wayland tiling compositor. Keyboard-driven, tiled by default, no floating-window nostalgia.
  • Quickshell as the shell toolkit. Since the 4.0 “Quattro” release in August, the bar, launcher, notifications, lock screen, and on-screen displays are one coherent Quickshell process instead of a federation of Waybar plus a launcher plus a notification daemon plus three other things.
  • One omarchy surface for everything: a single searchable menu (Super+Space) for apps and system commands, a theme and wallpaper switcher with a 24-color palette, plugin management, updates, keybindings, monitors; all in one place.
  • A full developer baseline out of the box: Neovim, the Foot terminal, tmux, Docker, a curated set of CLI tools, and, very 2026, nine coding agents to pick a system default from (Claude Code, Codex, Gemini, Copilot, and others), launched from a keypress instead of living as terminal afterthoughts.

The install is the part people don’t believe until they watch it. Write the ISO to a stick, answer about five questions, and you get a finished, themed, working desktop in under two minutes on most machines. Current ISO is 4.0.x, under 6 GB, with dual-boot beside Windows finally supported.

Why this is revolutionary, without exaggeration

Here is the sentence I keep coming back to:

The Linux desktop spent twenty years assembling parts. Omarchy is the first project in a long time with the taste and the stubbornness to ship a product.

Distros give you choices. Desktop environments give you settings panels with four hundred checkboxes. Tiling window manager setups give you a GitHub README that says “just copy my dotfiles, bro”; and then you spend a weekend debugging why the bar doesn’t render on your second monitor.

Omarchy deletes an entire class of pain: the week-long yak-shave between “I installed Arch” and “I have a desktop I can work in.” Every default is chosen. The theme touches everything at once because there is one theme system, not five config formats pretending to agree. The keybindings are documented, consistent, and the same on every Omarchy machine; which means a YouTube tutorial from six months ago still applies to your box today.

That last point is underrated. When every Hyprland setup is a snowflake, nobody can help you except the person whose dotfiles you stole. When a hundred thousand people run the same opinionated baseline, “how do I do X in Omarchy” has one answer. That is a product property, not a dotfiles property. It is the same reason Ubuntu won in 2008: not because it was technically superior to Debian, but because it deleted decisions.

And only a project with a single opinionated owner ships this. Every component Quattro replaced, Waybar, the walker launcher, Mako, SwayOSD, has its own community that would have voted no to being absorbed into one shell process. Consensus would never have produced it. Taste did.

If you are tired of GNOME telling you how to work, KDE giving you seventeen ways to configure the same panel, or compositors that assume you enjoy editing three TOML files before lunch; Omarchy with Hyprland is a genuinely interesting idea. Tiling, keyboard-first, everything a keypress away, zero mouse-chasing across a 34-inch ultrawide. For a lot of developers, that workflow deletes real, daily, measurable friction. I have watched people use it at speed and it looks like playing piano while the rest of us hunt-and-peck.

Why it is not for me

Because I like windows. Floating ones. That I move with a mouse.

That sentence sounds flippant and I mean it completely seriously. My brain does layout spatially: terminal here, browser there, reference window pinned on top, half-overlapping the way I left it. I run every tool change through one filter: what measurable operational pain does this remove, and is it worth the migration debt? Tiling relocates my problem instead of deleting it. It asks me to memorize forty keybindings to get back a layout I already had with a drag and a click.

This is not a complaint about Hyprland. Hyprland is excellent at what it is. It is a complaint about the fit between Hyprland and me. I spent days fighting the instinct to grab a window border, and every one of those fights was the tool telling me my muscle memory was wrong. Maybe it is. I do not care enough to retrain it. A good answer to a question I am not asking is still not an answer I need.

There is also the Arch part, honestly stated. Omarchy inherits Arch’s rolling model; latest everything, breakage occasionally included. I respect that trade deeply; it is why Arch users always have the newest Hyprland and the newest Mesa on day one. But I am the guy who runs Debian on his servers because he wants boring. A rolling desktop plus a keyboard-driven paradigm shift plus a new shell toolkit is three migrations stacked on top of each other. The knowledge migration alone, every failure mode I know cold on Fedora KDE, reset to zero overnight, is a price I only pay when the new tool deletes a 3 a.m. pain. Omarchy deletes a pain I never had: I never suffered setting up a tiling desktop, because I never wanted one.

So no hard feelings. Just no more Super+Enter for me.

What Omarchy did for me anyway

Here is the twist, and the reason I wrote this post instead of just quietly going back to Plasma.

Omarchy reminded me that the Linux desktop can be fun again. Curated. Cohesive. Themed down to the lock screen on first boot. I had forgotten what that felt like after years of sensible, gray, enterprise-approved defaults; and watching DHH’s thing boot into something beautiful in ninety seconds made me jealous in the productive way. Not “I want his desktop.” More like “why doesn’t my desktop feel this considered?”

That jealousy sent me somewhere I did not expect: back to Wayfire and Quickshell.

Wayfire is the compositor I actually wanted all along. Wayland-native, floating by default, mouse-first; but with the Compiz-era soul: wobbly-adjacent animations, the cube, Expo, window rules that feel like playing instead of configuring. It is the anti-tiling compositor for people who liked tiling’s modernity but never wanted its discipline. Pair it with Quickshell, the same toolkit Omarchy 4 built its shell on, and you get the part of Omarchy I genuinely coveted (one scriptable, themeable shell process you own completely) without the part that fought me (a window manager that won’t let me overlap things like a barbarian).

So that is where I am now: plain Arch, Wayfire, Quickshell, ricing the perfect desktop by hand. No ISO doing it for me, no omakase menu, just me and config files at midnight like it is 2009 again. It is slower. It breaks more. It is also mine in a way no opinionated distro can be, and for a floating-window mouse user who enjoys the craft, that is the whole point.

The irony is not lost on me. The most opinionated distro on earth turned me into a ricer. DHH would probably call that a success; “use this as a paved path,” he keeps saying, and I took the paving stones and built a different road. Fine. The inspiration still counts.

Who should run Omarchy

To be fair, and to be useful; plenty of people should, and the math runs the opposite way from mine:

  • Developers tired of GNOME/KDE who live in the terminal. If your day is Neovim plus tmux plus a browser, Omarchy’s keyboard-first tiling deletes real mouse-travel friction, all day, every day.
  • Anyone who ever bounced off Arch. The install story alone is worth it: Arch’s freshness with none of the wiki-diving. If the Arch Wiki ever made you close the tab, this is your second chance.
  • Laptop-first keyboard people. Tiling shines on one screen with no mouse. Train, couch, conference talk; Omarchy is built for exactly that posture.
  • People who want taste without the ricing homework. Twenty-four-color theme system, everything consistent on boot, switch themes with one command. You get 90% of a riced desktop with 0% of the dotfiles archaeology.
  • The curious. It installs in minutes, dual-boots with Windows, and uninstalls cleanly. The cost of trying is one evening. That is a rare property for something this opinionated.

The summary, in two lines

Omarchy in 2026: the first Linux desktop in years that feels like a product instead of a parts bin; opinionated, beautiful, and coherent out of the box. If tiling is your shape, stop reading and go install it.

For me: I like my windows floating and my mouse under my hand. So I stole the best idea, Quickshell, Wayfire, Arch, and went off to rice my own perfect desktop. Thanks for the inspiration, chef. I sent the tiling course back, and the rest was delicious.