"GIMP quality" is a very vague term. While it's clearly a very powerful tool (one type of quality) it's also impenetrable to most consumers (the lack of quality I was refering to)

Very cool... My version is getting close to be sharable. It's great we're both trying this out. What would you say have been the biggest insights in building this out?

Very interesting! By simplifying the window chrome do you mean you got rid of the title bar? How are you dragging the window with the super key?

As I spoke at KDE I'm using that but honestly, it might be better to pick something like Ubuntu or Debian as the goal is to get as many as possible to use it. I'm hopeful that if you get it to work for any of these three, the other two aren't that hard to support.

Give it a shot! I expect it'll work on other distros as well?

I"ve had some interest in building it. Still talking. Nothing even close to shipping unfortunately.

I'm not against an Niri at all! It's well loved and it solves a problem well. It's solving a slightly different problem. I'm trying to find that mix between windows that you focus on in the center and a periphery of all of your windows that you can see at the same time.

I would actually even argue Niri is the opposite of what I'm doing. It fills the screen with your window and then aggressively scrolls things off to the sides. It's also very heavily dependent on keyboard shortcuts, which is very good for power users, but I don't think approachable for novices.

Oh, I think there are LOTS of things yet to explore. As I said in my talk, it's important to play with something, be honest about what isn't working, and keep improving it.

Exactly.

I think anything that allows more advanced users to "quick switch" needs to be explored.

I never saw that. I had to did into the 'settings area' and turn it on.

I worked on the Android team for YEARS and I just learned that you can turn on "Clipboard History" on GBoard. So many useful features are just too hard to find/setup.

completely agree! I'm starting off slowly, trying to build up a learning/understanding of what works and what doesn't (e.g. the morph effect on window content). I see these demos as just the beginning. They aren't ready to "ship" just yet, it needs more exploration.

Thanks! Glad you liked it. I've learned from previous projects that a good idea never survives the first prototype: you've got to iterate it a few times. I'd like to build something to learn not ship. It'll take a few tries.

I do think we could be even more exploratory and go into a stronger "VR Style" of window management. This really was just a first attempt at the problem. I hope that once I can get a version of this actually running as a real window manager we could push it further.