Hot take alert:
This is a stupid opinion.
First, the article reads like an AI wrote it, but assuming that’s not true, the Linux space absolutely does NOT need more tiling window managers.
Quite frankly, I’m amazed there are still as many actively developed ones as there are.
The VAST majority of Linux users have little to no interest in a tiling WMs, and the basic tiling features of Plasma, Gnome, and soon Cosmic are fine for most of the users that want to try it out. The few that really want hardcore full tiling are almost always already very experienced power users who know what they want and how to get it. They aren’t going to be put off by their favorite distro not having built in support for tiling WMs.
In fact, most of them are already using distros that are able to be heavily customized to their liking, like Arch, NixOS, and Gentoo.
How many users do you think want to run Linux Mint or PopOS but with some hardcore tiling WM?
Linux has a massive amount of variety in all areas, it’s already hellish for new users to pick a distro from the forest of suggestions, do we really need even more tiling WMs on tip of the dozen+ ones that already exist and serve a tiny percentage of Linux users?
I disagree. What is wrong with a fully featured batteries included desktop environment that has proper tiling support (not just partital drag the window to the edge of the screen support). Lower the barrior to entry so that more people can make use of this powerful way of working. The main reason that tiling is considered hardcore is becuase it has mostly only been available on minimal configure them yourself window managers. But tiling does not have to be for the fully DIY only crowed.
IMO the basic tiling support on gnome or KDE are not good enough. So I am forced to use something minimal but TBH I am sick of needing 100s of lines of config to get a basic environment setup. Cosmic seems like it will be a good answer to this post as its tiling support looks far more fully baked than other full desktop environments and hopefully we will see more people wanting to try out tiling once it reaches a more stable point.
with kde 5 i used bismuth, now with kde 6 there is polonium but still a long way to go.
https://github.com/Bismuth-Forge/bismuth https://github.com/zeroxoneafour/polonium
Did you even read the article? Lmfao why are you going on a rant about hardcore tiling WMs - you missed the point of the article so bad.
I bet you don’t even use a Dvorak keyboard.
You got me!
they are only ‘hardcore’ because of the poor desktop environment integration.
No, they’re hardcore because tiling WMs are hardcore. Most users don’t want to use them, they don’t care. And they wouldn’t be significantly more likely to use them if more distros had them as a default.
Y’all complain that users still occasionally need to use the terminal for certain tasks on Linux, but you think those same users will be totally interested in spending hours writing Perl or JSON configs and memorizing dozens of keyboard shortcuts for every function they used to use the mouse for??
but you think those same users will be totally interested in spending hours writing Perl or JSON configs and memorizing dozens of keyboard shortcuts for every function they used to use the mouse for??
Of course not. This is the argument for a tiling desktop environment. The only reason people need to do all that ATM is because of the current tiling window managers. Not because tiling window management is inherently complex to understand. You can have a tiling window manager with a GUI configuration and that better supports the mouse while still supporting keyboard shortcuts. Then users can incrementally learn the shortcuts - like they do with floating window managers - to gain productivity in their day to day tasks.
They might not be for everyone, but giving everyone the choice is also not a bad thing. Most people I have seen that try a tiling window manager do end up liking it and quite a few hate to go back to floating ones. But not all of them can be bothered with the amount you need to configure the current ones.
So what is wrong with trying to make a easier to configure, use and generally a batteries included tiling desktop environment? This is essentially what it looks like Cosmic are doing - they support both floating and proper tiling without needing complex configuration or needing to learn loads of shortcuts.
You don’t have time be on arch, gentoo, etc to heavily customize your system - you can customize on Ubuntu’s. That the beauty of Linux.
I’d argue that we need more /wayland/ tiling WM. But you are right that most desktop users (ie. OSX, Windows) just want something intuitive that makes sense. They don’t want to learn i3wm. But once you do learn it, it’s hard to go back
We do have them.
Most popular DEs already support tiling with extensions (Gnome and KDE).
KDE actually added native support, although pretty limited so far.IMO the tiling support in KDE and with gnome extensions does not look great. It cannot replace someones workflow that has been on a true tiling window manager. It is a benefit to those that have been using floating window managers for their whole life but I cannot now go back to them. Cosmic is the first desktop environment that looks like it has true tiling support (that can rival a tiling window manger) and not just drag a window to a side/area of the screen. Though I have yet to really try it out.
This. It’s not well-advertised in KDE – I accidentally discovered it through a key combo – but it was good enough (i.e., Win 11-level) in KDE 5 to make the switch painless on desktop. Where both have issues is apps insisting there are arbitrary dimensional minimums for functionality and refusing to adhere to positioning. This is most egregious in messaging programs.
On KDE you can try out Polonium, which integrates with the native tiling support and makes it awesome!
I’m currently using KZones, actually, it’s not automatic, but it works pretty great.
https://github.com/anametologin/krohnkite
This looks quite nice as well
If you’re still on X11. Krohnkite didn’t support Wayland all that well last time I checked.
Makes sense. I am using Bazzite, and while it uses Wayland by default, I saw that some games refuse to run on Wayland (F.E.A.R 2 in my case). So I just went back to X11.
I want to like Wayland but it’s making it difficult for me. Oh, I am using Nvidia, so that could be a reason as well.
Heh, I was confused because I switched full-time to Wayland a while ago and that was never an issue for any game… but then you mentioned Nvidia. RIP
I saw they switched recently to partially open source drivers, so hopefully it’s gonna be better for you soon.
Only the kernel module is open source, and it’s just a wrapper for closed source blobs.
In actuality the open source drivers just kill all support for the 10 series, and otherwise do nothing to fix Nvidia’s utterly fucked up driver problems.
@quarterlife @sanpo About Nvidia’s gpus being proprietary and shit… I’m on AMD because, well, there’s no alternative. Still, I’m struggling with most FOSS AI being developed specifically for Nvidia. That’s non-sense. Develop FOSS for non FOSS-hardware.
I got this card for free so didn’t really have a choice. My next purchase will highly be an AMD GPU, unless Nvidia does some magic (which I highly doubt)
After I switched and got used to a tiling WM, it’s really painful to have to use a floating one.
Or: Tiling Window Manager distributions instead
I think what this guy need are more distributions with a pre configured setup of a tiling window manager, like a desktop environment. Some distributions have such i3 spin in example. I wouldn’t mind if we had more of those. It would definitely make it easier for people to try or get into one. Such as the theme and their applications, shortcuts and widgets setup in a consistent way to all their other desktop environments.
Before I switched to EndeavourOS, I was a Qtile user. Back when EndeavourOS still had a community maintained Qtile spin, I thought to myself, why not? So I tried it. What can I say? It was Qtile, but with all the stuff that makes up an EndeavourOS. It was ready to use. I still reconfigured a few stuff, but it was a nice and pleasant experience right from the start. It was an eye opening experience, in which I believe such a tiling window manager spin makes sense.
Switching a whole distro is an extreme way to try out a tiling window manager. It would be far better if desktop environments supported it so it is a simple toggle a user can turn on or off and not having to upend everything to get into.
It’s not an extreme way at all. I was thinking along the lines like testing in a virtual machine in example. Also one does not have to switch whole distribution each time they try. Installing a distribution is often easier than installing the tiling window manager and configuring it hours and hours, finding out details how to do stuff, to get it into a state that is comparable and styled the same way as the other distribution spins.
This dos not mean one has to use that, you are still free to install and configure it from scratch. Some distributions have it and its great.
I have been working on a Wayland Tiling WM and have thought about expanding more into the DE space. While it’s finally starting to get to a good spot, it’s pretty daunting to consider all of the other items that need to be developed for a fully-featured DE. Especially when it’s something I’m doing as a side project after my day job. I think for anything bigger it’d require financial backing, for which open source projects are still struggling to find a good solution.
I want a good tiling DE solution so much. I really hope COSMIC will fill in that gap.
Couldn’t tiling just be done with an app like how PowerToys FancyZones does it on Windows? That way anyone could just install it when wanted.
Those always feel like a half baked hack when compared to a true tiling window manager. At least all the ones I have tried on my work mac I have not found any that are good enough and all have weird edge cases or break in weird ways.
It would need some sort of way to hook into the compositor. PowerToys has it easy because they can just add the necessary APIs to the Windows compositor if it doesn’t already have them. And I feel like compositors would just implement it directly instead of designing an API for it because that’s less complex.
I use tiling now since a year. And it’s so painless and straightforward if you know the key combinations. People watch me working got headache because of the speed.
My setup was easy because, instead of building all these settings I used an existing one.
My desktop install was fedora-sway but I added an easy hyprland script from this guy.
https://github.com/JaKooLit/Fedora-Hyprland/
Fully functional hyprland setup installation in under 30 min.
