question: Will Owl become "official" tiling wayland compositor

[amatsagu]

Hey, like in title - I'm curious if this project will become something like sway, river or hyprland or it's only really meant to be custom tailored for its creator and it just happens to be public for anyone else to use.

It looks promising, it's neither overcomplicated cow like sway and has the fancy effects of hyprland without developer being a dick (sorry, not really). Some of our staff members at company use tilling window managers on laptops but as most of them are 14'' or even 13.5'', wayland's fractional scalling is must have as we rely on 125% or 150% scalling - also battery life is a thing. Sway is a bit battery draining at times but serves us well but I'm always looking for better alternatives. We tried Hyprland at some point and it's better - as long as it works... It's not something we can rely on, hyprland breaks a lot between updates.

In case Owl is planned to be officially supported and You deem it stable enough then we would be happy to give it a monthly trial. We mostly work in VSC, Gimp, Krita, Web browsers (google chrome), Termius, Discord, etc. All our apps have some level of wayland support so xwayland support is not required, in worst case we can try running rootless xorg-server like we did with Niri.

About protocols... I don't really know much about them but we would expect basic stuff like audio control (pipewire), video capture (discord and obs), brightnessctl control, touchpad gestures, drag & drop files between windows, waybar, lock screen (swaylock or gtk-lock), hot plugging extra monitors, fullscreen app support (maybe not really for games but some render apps we may use) and that's probably it.

We would be happy to be early testers, report any problems we find, give our own feedback after some time of use and maybe prepare a debian package. We all work on Debian 13 (Trixie, testing branch) - it has like 2-3 week old packages compared to Arch most of the time.

Hope it doesn't sound like we want to force anything lol, we're just curious and would like to give it a try but before that - get confirmation whether it's going to be long term public project or just side hobby of creator, not really for other people usage. Oh and all laptops rely on AMD CPUs with integrated graphics*

[formadi]

I think it’s a new window manager that is lighter than Sway, doesn’t yet have as many features as Hyprland, but stays true to the basics while being as sleek as Hyprland. Personally, I really love it.

[amatsagu]

I think it’s a new window manager that is lighter than Sway, doesn’t yet have as many features as Hyprland, but stays true to the basics while being as sleek as Hyprland. Personally, I really love it.

Sounds good. I'm not really a fan of Hyprland, like it's cool on it's own but Vaxxry (main dev) makes various questionable decisions that brings project down. They add more & more of features which is both good and not really as it stops being just your minimal tilling manager, purposefully makes more hypr-* only things and pushes "weird" ideas like uwsm. Many of those things I've seen only on hyprland, nowhere else - and i's clear a lot of it is not very welcomed by looking at low amount of stars on github.

What Sway does is like 95% there for us, nothing really to nit pick about other that it is heavy at times (noticeably worse battery lifetimes & occasional animation lag when moving quickly windows of heavy apps even on powerful CPU and iGPU).

[dqrk0jeste]

hi! first, thank you for supporting the project. it really means a lot seeing other people seeing potential in it and wanting to daily drive it. lets go one by one

Hey, like in title - I'm curious if this project will become something like sway, river or hyprland or it's only really meant to be custom tailored for its creator and it just happens to be public for anyone else to use.

yes, i do currently work on making owl for everyone, actively taking feedback and feature requests from people that have tried it. i should tweak the readme a bit, so its not misleading.

It looks promising, it's neither overcomplicated cow like sway and has the fancy effects of hyprland without developer being a dick (sorry, not really). Some of our staff members at company use tilling window managers on laptops but as most of them are 14'' or even 13.5'',

the idea behind owl is to keep it dead simple and, most importantly to me, predictable. thats why i choose master layout, because i know exactly where my toplevels are going to go, its dead simple, and thats also why owl does not have that many unique and flashy features, and thats also the reason why some feature requests wont be implemented (or will be tweaked) if they dont fit this criteria (unless there is a really good reason and a lot of people wanting it, which is still not the case as owl is still pretty niche).

wayland's fractional scalling is must have as we rely on 125% or 150% scalling

fractional scaling support has been added yesterday, 99% of work is done by wlroots so it should just work.

- also battery life is a thing. Sway is a bit battery draining at times but serves us well but I'm always looking for better alternatives. We tried Hyprland at some point and it's better - as long as it works... It's not something we can rely on, hyprland breaks a lot between updates.

i may be biased here, but i feel like the battery life really good when running owl, much better then in Hyprland for me, tho i will say it again, it may be just bias.
as owl aims to be dead simple, one of the things i will try to do is to avoid conflicting config options as much as possible, and NOT break your config ever.

In case Owl is planned to be officially supported and You deem it stable enough then we would be happy to give it a monthly trial. We mostly work in VSC, Gimp, Krita, Web browsers (google chrome), Termius, Discord, etc. All our apps have some level of wayland support so xwayland support is not required, in worst case we can try running rootless xorg-server like we did with Niri.

i would say owl is now failry stable - it should not crash unless some weird edge case happens (please let me know if it happens). in the past month i have been using it exclusively and it has been rock solid. tho there may be more issues found by others - i basically only use a terminal and a browser lol.

About protocols... I don't really know much about them but we would expect basic stuff like audio control (pipewire), video capture (discord and obs), brightnessctl control, touchpad gestures, drag & drop files between windows, waybar, lock screen (swaylock or gtk-lock), hot plugging extra monitors, fullscreen app support (maybe not really for games but some render apps we may use) and that's probably it.

all the things here are currently supported other than lock screens - i will add the support for them soon(-ish), as they need some work. screenrecording works. discord does not even launch for my under any wayland session - i use vesktop client and screensharing there works (only whole output sharing is supported, but thats a limitation of xdg-desktop-portal-wlr, there may be a workaround there).

fullscreen apps are supported tho currently owl does not allow apps to be launched immediately in fullscreen mode, i dont find that predictable, i would rather fullscreen them myself (i can add a config option to allow it). speaking of that, i want to add a server-side keybind to fullscreen clients, so it is the same for every app.

We would be happy to be early testers, report any problems we find, give our own feedback after some time of use and maybe prepare a debian package. We all work on Debian 13 (Trixie, testing branch) - it has like 2-3 week old packages compared to Arch most of the time.

that would be awesome, i really appreciate it! owl is not hard to build, but packaging it for debian would really help. here is my arch PKGBUILD for a reference if you end up doing it.

Hope it doesn't sound like we want to force anything lol, we're just curious and would like to give it a try but before that - get confirmation whether it's going to be long term public project or just side hobby of creator, not really for other people usage. Oh and all laptops rely on AMD CPUs with integrated graphics*

no worries! amd should play nicely.

i also wanted to let you know that i plan on adding more eye candy to owl by utilizing this project when it becomes ready. i have talked a bit with the people behind it and that should probably be this month. by using their fx_renderer owl will be blessed with rounded corners, shadows and blur. exciting! there will be a testing branch first because it may involve some bugs!

again, thank you for the support, and please ask if there is anything else you want to know or need help with :)

[amatsagu]

So, I think I'll keep lurking around the project for little longer before I start using it - having ability to lock our screen with swaylock or gtk-lock is important to us due to security reasons - work laptops contain various, protected data. We obviously only allow laptops to go outside if they have full encryption.

I can't really help with code as I only know how to program with golang, typescript or bits of rust. I only know very basics of C and like nothing when comes to wayland or wlroots inner side. I'll try helping with packaging tho. I can't promise it gets to official debian repositories as that's actually quite painful to achieve but we can just make releases here on github and manually attach *.deb files, similar to how Vencord, Google Chrome or few other app does.

btw. I like that you want to create something minimal, I really dislike when hyprland forces new standards like uwsm out of nowhere or starts using QT stuff as required lib. Stuff like animations, blur, rounded corners are always nice to have but priority is core functionality and optimization if you ask me.

I'm not super active on github so it may be easier to catch me on discord if needed - invite/message amatsagu (390394829789593601)

[dqrk0jeste]

will try adding ext-session-lock-v1 asap

.deb files are fine, i dont expect to get owl to the main repo. i plan to soon(-ish) finally start versioning owl, so we can have releases.

[dqrk0jeste]

having ability to lock our screen with swaylock or gtk-lock is important to us due to security reasons

@amatsagu added :)

[dqrk0jeste]

can i close this?