GNOME Devs Are Working on a New Window Management System
GNOME Devs Are Working on a New Window Management System
I really enjoy how GNOME handles windows currently already.
Between having the ability to move and resize windows with Super + (mouse left|right), switching between windows of the same application with Super + ``, workspaces and Super + type` to search, there is very little to desire.
Unlike tiling VMs, this makes sense out of the box for 99% of the apps out there while providing a really quick way to get where you need quickly.
There are already ways to have tiling and a DE.
On GNOME, there’s PaperWM, although it’s not quite traditional tiling either.
On KDE Plasma 5.27+, you can use Polonium. For versions before 5.27, Bismuth.
And on Xfce or LXQt, it’s often possible to use them with a traditional tiling WM, like i3wm, bspwm etc…
Krohnkite went unmaintained a while ago, which is when Bismuth forked from it. So, Bismuth is basically a straight upgrade. The dev implemented tons of features, which you may or may not need, but I think, there were also some fixes for stability and Plasma version compatibility.
Polonium came about, because Plasma 5.27 introduced a (manual) tiling system of its own, which partially broke Bismuth, but also meant it made sense to develop a new KWinScript, which makes use of this native system.
As such, it is a step back from Bismuth. I think, it’s roughly comparable to Krohnkite in terms of features now, but still a very young project, so not as stable yet…
The 9to5 article is poorly written. In the first paragraph 9to5 says a new window system is "scheduled to replace" the current one, but this is not true. In fact, in the blog post the article cites the Gnome developer Tobias Bernard explicitly says "There’s no timeline or roadmap at this stage". The Gnome developers are merely experimenting with a new window management system and at this early stage it's impossible to know what the finished product may look like if these experiments go anywhere at all.
Here's a link to the original blog post where Tobias explains their dissatisfaction with existing window management systems and discusses the techinical challeneges developers face.

Window management is one of those areas I'm fascinated with because even after 50 years, nobody's fully cracked it yet. Ever since the dawn of time we've relied on the window metaphor as the primary way of multitasking on the desktop. In this metaphor, each app can spawn one or more rectangular windows, which are...
Agreed. It’s the best blend of keyboard driven window management and recognizing that users might also use the mouse from time to time. I got my wife to use and default to tiling with Pop!_OS.
The only problem is Pop!_OS is a shitshow of dependencies being built on Ubuntu. I had an update last night that reinstalled snapd and LibreOffice and Firefox even though I intentionally uninstalled them in favor of the flatpaks. Cosmic DE, and presumably re-basing Pop!_OS on nixOS (given a dev comment) can’t come soon enough.
Yet somehow, through only apt updates, it brought back LibreOffice, Firefox, and snapd.
IIRC, it was something to do with ubuntu-minimal or ubuntu-release meta packages, which I never intentionaly installed.
I’m probably the only person who uninstalls the Firefox and LibreOffice packages and replaces them with the flatpaks, but this seemed like an oversight and dependency hell that comes from using the derivative of a derivative distribution.
I experienced the same thing (had previously uninstalled libreoffice, but it came back after the update). I didn’t get snapd back fortunately (though I do use Firefox packaged by Pop).
Part of the change is that Pop!_OS is moving away from ubuntu-minimal and ubuntu-standard meta packages and towards their own metapackages as shown in this this recent commit.
After the update, I simply uninstalled libreoffice… hopefully it doesn’t return in the next update :]
Yes, the solution for me was to remove those ubuntu-* meta packages, reinstall what I needed by hand then update. Simple things like ftp, telnet, time, etc. had to be reinstalled.
I was kind of nervous on the reboot since a plymouth theme was removed in addition to adding a newer kernel with the amd microcode patch, but it came up fine.
xdg_toplevel::set_max_size/::set_min_size, I’m sure the rest of the ecosystem will be on-board for this.
Exactly, for this community is to blame. People mostly are against even minimal and anonymous surveys, telemetry and stuff. So, all they can do is just assuming that people want something or not.
Usually they are talking to active community members, whom, we all know that programmers and technical people.
IMHO, they need a bit more data to decide on
And yet it seems to me only GNOME has this problem, and it has been there since Torvalds still publicly executing everyone in mailing list. XFCE, LXQT, hell, even KDE only has minimal complain about unexpected behavior. It seems to me that in a concerted effort to predict as much user behavior as possible, GNOME created this non existent “average user” that conforms to no one, and created this mess on their own.
Also, we are mostly against nonconsensual, non-explicit, or opt-out type of feedback. As far as I concern, efforts to point out to GNOME devs their faults are many to the point its a meme. It is also, not unrelatedly, a meme that GNOME denies these complaints because “the average users wouldn’t get it”) . I think it should be clear enough by now.
On one hand I’m interested in seeing how well it works and what they do with it, on the other hand…