But then the security software and all its libraries has to be trusted, and isn’t a kernel module.

Steam Controller One.

Still have no idea why Microsoft did that.

If that’s the case, throwing more people, especially juniors, into the team won’t make a difference.

By the way, having a team full of people that “don’t want to work” is a sign that they lack trust in the project or leadership. Maybe management should work on that rather than spouting right wing linkedinisms. :P

For friends that I know, they can just ask me and I’ll give them a login secret. I don’t have the headspace to manage fedidrama.

Most actual Linux code changes come from large companies implementing or improving drivers for their own hardware.

The Linux foundation mostly manages the Linux “brand”. That is, all the logistics and infrastructure required to run a huge project with many stakeholders.

If you’re missing deadlines and getting customer complaints because of a new hire, that’s a failure in management, imo.

(Of course, that’s not saying management will take responsibility)

Doing a very quick search on this and it looks like the firmware updater uses webbluetooth. This knife has bluetooth hardware in it.

My gut reaction is that there are a lot of ways, once you have root access, to have your changes persistent. For example, modifying/replacing binaries, adding new ones before old ones in the path, adding startup scripts, modifying config files. Kernel modules seem to be an overcomplicated way to go about it, especially since (afaik) it would need to be compiled against the specific kernel version.

Ehhh… Reads like the standard “we’re happy with you tweaking it, but don’t want to be liable for you breaking it.”

If you try to sue them, they can point to this section of their docs and say you didn’t follow their recommendations.

“Why yes, can we order 1kg of kilogram and a 1m metre please?”

I thought we weren’t supposed to be using Mint anyways because it uses Systemd? /s

Wait, Windows doesn’t even have system wide dark mode yet? I thought that was standard in 2026.

Hell, even Gnome has it.

Once you’re booted secure boot is inactive. If there was a security benefit to only loading signed modules, then distros would have that enabled by default regardless of the secure boot status.

Iirc, requiring modules be signed is a requirement Microsoft put on the shim bootloader rather than Lunux’s choice. I could be mistaken here, I’m not too sure on the specifics.

Regardless, if someone has the ability to load or modify modules on an encrypted Linux install, they can just steal Firefox’s cookie jar and cut out the intermediate step.

He charges no interest and doesn’t seem to have a payment deadline. You could honestly just leave at any point and never pay him.

Feels like you’re scamming him rather than the other way around.

Emacs.

With all the vimmery going around nowadays though, I feel like I’m on the losing team. ;_;

I get that perfect is the enemy of good, but you also need to have defined threat models. Secure boot protects against people covertly taking your ssd, putting it in their own device, overwriting the OS with one that looks identical but is a key logger, and then putting it back in your system. Yet systems with secure boot have no tooling to stop attackers from just… Putting a hardware keylogger inline with the keyboard.

Finally, year of the Windows desktop.

If someone has access to your device enough to modify your bootloader they could also just install a hardware keylogger or hidden camera and get your password that way.

Nothing some estrogen can’t fix.

What advantage does secure boot have compared to full disk encryption? The only examples I’ve seen have been contrived evil maid attack that fails under scrutiny.