If space constrained and wanting compression, why not do that at the filesystem or block layer if it's not supported in the app?

There are many reasons to dislike GNOME:

1. Very little can be customized. 2. Extensions that let you customize things are unlikely to work in the next release because the APIs keep changing. 3. GTK apps have enormous padding around everything that eats my precious screen space. 4. It's heavier and slower than KDE. Probably thanks to all the embedded JavaScript. 5. Its' "my way or the highway" approach to workflows is abrasive.

Here's a nickel, kid. Get yourself some non-Nvidia hardware.

Everything I have runs AMDGPU.

Things have come a long way in the last year or two ;)

Cosmopolitan libc binaries require a little extra config to work around the defaults most distros set for binfmt: https://justine.lol/apeloader/

If you're writing Python tools to support OS operations in prod, you need to target the system Python. It's wildly impractical to deploy venvs for more than one or two apps, especially if they're relatively small. Developing in a local venv can help with that targeting, but there's no substitute for doing that directly on the OS you're deploying to.

This is why you DON'T write system tools in Python in the first place. Use a real language that compiles to a native self contained binary that doesn't need dependency installing. Or you use a container. This has been a solved problem for decades. Python users have been trying to drag the entire computing world backwards this whole time because their insistence on using a toy language invented to be the JavaScript of the server, as an actual production grade bare metal system language

Just use Debian and switch it to Testing. Works amazingly well and you'll always have relatively current and generally stable software.

Headless remote desktop, at least for KDE, is very much not possible today as far as I can tell. It's the last thing I miss from Xorg.

FWIW, like everything else in Wayland, it's per-compositor. Here's a working headless sway in a container:

https://gitlab.com/yjftsjthsd-g/docker_sway-vnc

(This is not a defense of Wayland, just trying to share useful information)

Hmm, you mention in the README that it only works in a privileged container. This of course negates the security benefits Wayland supposedly has over X11, so it doesn't seem ideal.

It really doesn't. The security benefits of Wayland are about isolating applications from each other, which this still has.

Also that's only really relevant to this running in a container. My point was that you can have headless Wayland.

I love Wayland a lot, but as far as I can tell the available remoting solutions still cannot enable a headless LXC container to serve a KDE Plasma Wayland desktop. I spent the last couple days trying to cobble some solution together for it and failed miserably. If you know a way, I would be most grateful :-)

Me too!

He has in fact written one: https://bellard.org/ts_server/

Yeah I've seen that, but it looks like the inference-side only?

Maybe that is a hint that he does use off-the-shelf models as a coding aid?

There may be no need to train your own, on your own code, but it's fun to think about

Are you saying a LFM could be a good idea? A Large Fabrice Model?