Set windows version to Windows XP via winecfg, install, set back to 10.

You can have it set so it fails to boot with secure boot disabled. Not part of my threat model but AFAIK it’s default even for Windows FDE.

I am but with self enrolled keys. People wildly misunderstand secure boot, it’s more for kernel/boot level malware and so it can be used with module signing.

The Epstein class appreciates your brave defense

Thanks for the update.

I don’t like how the developer handled it and I have deep disdain for LLMs but is a project mostly by a single dev the place to focus attention on?

Low key though and I hope nobody on hexbear dot net sees me tell you this. Running Fedora on your computer at home doesn’t really provide any support to red hat the company and its government and military shit. It is also unlikely that they are overtly backdooring their distro specifically (its more likely/true to say that they have helped backdoor almost all distros via systemd.)

Of course not, the backdoor is on NSA developed SELinux /s

I didn’t post this to tell people to switch from Fedora, just that it’s monstrous behavior by the parent company. Of course they use Fedora as their downstream beta testing of features but that’s a different conversation.

Use Ubuntu or stay on Mint. (really, they are fine, I have been around long enough for when the hyped distro for new users was Mint instead of Ubuntu). No one hypes the stuff that works like OpenSUSE (well, someone mentioned it to you, think it’s a good distro but they are EU Red Hat).

deleted by creator

I had no idea Solus still exised.

I think I messed up formatting. Context is Red Hat helping military kill people with AI, scrubbing the info off their site when it gained attention.

https://web.archive.org/web/20260402155236/https://www.redhat.com/rhdc/managed-files/ve-compress-the-kill-cycle-detail-693397pr-202402-en_3.pdf

Radio Free Asia probably. Maybe BBC. Very legitimate sources of course.

The reason some programs were paused is because they can’t handle the volume of slop submissions that seem real, take time to triage. It’s not because AI is “good at finding bugs”.

Curl author complained about these submissions repeatedly https://daniel.haxx.se/blog/2025/07/14/death-by-a-thousand-slops/

HackerOne is the least innocent of all, having a slop bot to ”help" triage and submit bug reports.

It would be possible, I considered implementing such a proof of concept years ago for fun (when blockchain was not exclusively cryptocurrencies). But then I remembered I don’t like DRM so I never released anything about it.

Lisp is so back

A chemist who drank a gasoline additive to prove it was safe knowing it wasn’t

I think it has to do with the slippery slope. When fields like work phone were added they served actual uses at the time.

They may be functionally identical but this is happening under an increasingly more invasive surveillance system and people have hated systemd eating every linux component since forever. Now that it’s a soft dependency on major desktop environments it just feels like a rug pull that they get to make these changes

It upsets me more that he’s letting AI bros contribute to systemd than anything he did before. I think he even used one of the slop machines to review the PR.

They do. Most programmers think they’re above average (there were actual statistics on this, maybe from stackoverflow survey) and are mediocre enough that they find it useful/faster long term.

I’m statistically likely to be mediocre myself, but I would rather try to improve than relying on LLMs. Every single coworker I work with who is actually above average hates the forced AI usage.

Even on /c/selfhosted a lot of projects were being advertised that were blatant AI slop

This is a sort of reputable script for removing most crap https://github.com/christitustech/winutil