It’s like another attempt at programming with natural language, except using a non-deterministic black box.

They should call it Wish-BASIC ™

I thought I saw it as far back as simcopter

Strange, I’ve never seen that. Have you rebooted the system to make sure it has nothing to do with open files?

I did find one thread that seems related:

https://www.reddit.com/r/btrfs/comments/lip3dk/unreachable_data_on_btrfs_according_to_btdu/

btdu is an excellent tool for finding out what’s taking up space in btrfs

Sorry for the duplicate replies. Lemmy server drama…

That’s a tricky one if you’re getting no info from the kernel. I think the reply above about system instability under load sounds promising. Throttling things down to test seems like a good idea.

Were you running dmesg on another screen or over ssh or something? I’d look in journalctl -b-1 after a reboot.

Is it completely frozen or does it respond to pings etc?

Were you running dmesg on another screen or over ssh or something? I’d look in journalctl -b-1 after a reboot.

Is it completely frozen or does it respond to pings etc?

deleted by creator

Unfortunately X forwarding doesn’t work (as far as I can tell) with vulkan.

What I’ve been doing is using waypipe (which seems very stable), with xwayland-satellite (which is not so stable) on the remote end.

I’d also love persistent sessions, so I’ve been following wprs, but it doesn’t seem to support GPU drawing at all.

Lots of interesting tech, but it’s still pretty immature.

export WAYLAND_DISPLAY=wayland-1

Like with X it’s not guaranteed to be that value, but same idea.

C at least has a preprocessor. C# has almost nothing except generators, which are a huge pain in the ass. Java seems to be similar.

Lisp is the greatest. Everything else is in between.

We should have tools and libraries that help us avoid boilerplate, not ones that help us write more of it.

In interactive add mode you can use s to split a hunk, and e to edit it. That’s usually enough for me to split things up.

Or to copy something and modify it 30 times.

This seems like a very bad idea. I think we just need more lisp and less AI.

I was a pretty experienced programmer when I first read SICP, but I still found it incredibly valuable. I’d recommend it to anyone.

I don’t disagree exactly, but I’d argue that you’re contributing to the project even if you’re just reporting bugs or helping others with it on e.g. Lemmy.

I could see avoiding all of that pragmatically in order to use some obscure, critical software, but not something you use every day and for which there are reasonable alternatives.

It’s kind of absurd. When you buy a TV, the bloated adware at least helps lower the price. Imagine paying extra for it.

Fair enough, most of that isn’t something a user should have to worry about.

VT is just Virtual Terminals. You always have one of them active, and in most distros you can switch to others by Ctrl-Alt-F1 through F12. In some distos it’s just Alt-F1.

So if you press Ctrl-Alt-F2 you should be brought to a text login. For crazy historical reasons you may have to either press Ctrl-Alt-F1 or Ctrl-Alt-F7 to get back to your usual graphical session.

Arch docs for example: https://wiki.archlinux.org/title/Linux_console

I would try:

• see if you can get logs of the resume process
• suspend from a text VT and see if that changes the behaviour
• boot into single user mode and try suspend from there
• boot an older LTS or a newer test kernel and see if it has the same problem

I just started using finamp a couple of weeks ago and this inspired me to install the beta.

If I find any problems I’ll try to get involved on the repository. Discord is a bit of a turnoff though.