In an effort to escape the data-mining AI hellscape that is modern software, I installed Linux on bare metal (a HP laptop) this weekend. I haven't run Linux outside of a VM since I got my first academic VMWare key in around 2013, because I frankly haven't needed to. VMWare, VirtualBox, WSL, Multipass, Lima, Docker Desktop, and the cloud have provided me a ton of capability without having to worry about drivers.
I ended up going with Debian Stable. Canonical seems as enthusiastic to treat me as just another training tool for AI as Microsoft, so Ubuntu wasn't on the table for me. My Red Hat developer account is broken, so I can't connect to RHEL package repositories. I'm not smart enough to run Arch. That left me with Fedora and Debian, and I have been super impressed with Debian recently. Greg K-H did a great episode on the #osspodcast about Debian running a stable kernel release, and the way the team responded to the XZ backdoor was incredible. Plus @joshbressers recommended Debian Stable to me.
I am super impressed by the experience so far. The visual installer in the netinstall image worked pretty seamlessly (I had to ask it twice to detect my NIC, but it found them on the second try). There's a hiccup where Wayland doesn't want to use the AMD GPU in the laptop (even though the drivers are loaded and the card is detected), but there's not much screen tearing in Gnome even with the Intel integrated card being used to drive a 4K display. Honestly, it works better than my work machine (M3 Pro Macbook), because I don't have to randomly restart it to get it to use a USB C monitor, and I don't need a 3rd-party utility to get windows to snap to the sides of the screen (how is that not implemented in MacOS in 2024?).
QOTO: Question Others to Teach Ourselves
An inclusive, Academic Freedom, instance
All cultures welcome.
Hate speech and harassment strictly forbidden.