• 0 Posts
  • 36 Comments
Joined 1 month ago
cake
Cake day: January 28th, 2026

help-circle

  • There’s nothing I can say to convince you Ubuntu is sucky if you don’t personally experience problems with it/have the occasional “I wish I could do this, but I guess it’s not possible” thoughts—spoiler: it’s usually possible, just not with that distro; this is true for all distros though, there will always be things they cannot do that others can.

    That being said, my biggest gripe with Ubuntu, besides canonical’s geologic-paced attempts to make it profit-driven (which was what I was thinking of when I mentioned user-friendliness), is Snaps. I understand the motivation behind them, but the implementation was just sucky for many nitpicky reasons that I don’t really care to enumerate rn.

    I also just don’t see much value in downstream for my needs. I can basically turn Debian into a Ubuntu clone if I want to, and my packages will still largely be supported by upstream maintainers. For bespoke distros it makes sense, but those are usually ill-maintained or hobby projects that update at glacial speeds.

    I’m not saying that Debian >>>>> Ubuntu. It’s better just enough that I’d recommend Debian before I would recommend Ubuntu.


  • I had very few issues with a GTX 970 and i7-4790k. The only issues I hear about with either any more is the linux kernel not supporting some of the features of newer GPUs (e.g. I know ray-tracing was a pain-point at one point).

    I don’t like recommending distros based on such a general use case, mainly because every distro can be tweaked and configured to exactly what you want. Instead, you should research the different mainline distros that have been around for decades—Arch, Debian, Fedora, Gentoo, Guix, NixOS, OpenSuse, Slackware—and see what they’re about, what sets them apart from others, what the maintainers’ philosophies are, and what kind of package management system they work with. Once one sounds better than the others, look into it and try it out.

    #Dos and Don’ts:

    Don’t try a niche distro. They are harder to troubleshoot and less likely to be actively maintained.

    Don’t use Ubuntu. It’s just a suckier version of Debian. It used to be user-friendly Debian, but now Debian is more user-friendly than it.

    Don’t dual-boot with windows. This just solidifies your reliance on windows, especially if you’re the type to give up on problem-solving issues that you didn’t have in Windows. It also can cause issues with making Linux unbootable.

    Do try a live usb with persistence before you commit entirely. It’s not exactly the same as a complete install, but it’s close enough to let you know how the OS feels and what hardware will or won’t work with it. Some people say try a VM first, but that won’t have direct hardware access.

    Do problem solve the little things. Anything that irks you or bothers you or just slows down your workflow. It doesn’t have to be an actual bug or glitch, just anything that could be better. This not only solidifies the feeling of ownership over your OS—you no longer have to settle for anyone else’s lousy design choices—it teaches you the resources for troubleshooting larger issues.

    Do plan around things not being plug and play at first. Want to test if a game runs on Linux? Great, set aside a couple of hours beforehand: first to install steam and set it up, then to figure out Proton, then to troubleshoot the game not even booting up, then to fix any glitches or whatnot, then to get your controller working. This won’t always be the case, but it will irk you a lot less when it is if you expect it. The more you make time for solving these issues now, the less time they’ll take up in the future (either they’ll be gone, or you’ll immediately know how to fix them, or your troubleshooting will be more streamlined).

    Do set aside time to learn about Linux “under the hood.” You don’t have to become a computer scientist, but it will save you a lot of headaches, show you cool things you can do, and make your computer a smoother experience. It especially helps if you take the time to learn as they come up: e.g. installer asks you what “bootloader” you want, but you’re not sure what that is, what it does, or why it’s necessary? Now’s the best time to take a little learning detour.

    Do ask questions on forums.

    Don’t listen to the people who shame you for asking.

    Do listen to the people who try to show you a better way of doing things, even if it’s not your way.














  • On KDE I couldn’t get Steam to put my game library on my second harddrive. It would open up the file finder, then simply ignore whatever folder I picked (regardless of drive and folder permissions). I was able to recreate the issue on Gnome under wayland, but X11 works fine. I even tried making a symlink to the other drive in my home directory, no dice. Tried flatpak steam as well as valve’s installer script; nada.

    Interestingly, it seems that the “pick a folder” button in Steam opens up a contextual file search window in X, but just a regular nautilus instance in Wayland. I’d say that this is the problem (the regular nautilus/dolphin instance not reporting back to Steam what folder I selected), but it works for moving to different directories, just not drives (in both DEs). Same thing happened on Fedora, so it’s not just “Debian is too outdated.”

    But let’s be serious, if I wanted to spend a lot of time tweaking and tuning my graphical environment to be exactly what I want, I’m not settling for Gnome nor KDE. I’m not gonna go with Cinnamon, XFCE, LXQt, LMDE, MATE, nor any ecosystem. I’m going with a window manager and mixing and matching every single program/element myself.

    I use i3 on my laptops. I would use Sway (because I don’t have to care about Steam), but for some reason it’s like 5x as resource hungry on these machines (constant freezes and stuttering).




  • A few answers say “they aren’t private by design,” but don’t really go into the “why.” There’s the obvious “it’s an electronic tracking device, duh” reason, but there’s also a more nuanced reason:

    Airtags are able to be picked up almost anywhere because they connect to the nearest bluetooth-enabled Apple device, and then send location info across the internet to you. Without this functionality (the ability of any and every Apple device to locate it), they wouldn’t have any way to send their location back to the owner.

    Your best “privacy respecting” alternatives are something that uses meshtastic (and hoping there’s enough repeaters near you), something that uses cellular data and GPS (which is about as privacy-respecting as Airtags are), or just a key finder/beeper (which only works within a small radius)


  • This article was more constructive (suggesting alternatives) than destructive (leveraging critiques), but it did link to several critiques/vulnerabilities with OpenPGP.

    Unfortunately, half are about implementation issues (granted, it’s made more difficult to implement something correctly when it’s as convoluted and all-encompassing as PGP)—which are hopefully not applicable to Delta due to their 3rd party, applied cryptography audit—and the rest are obsolesced by the 2024 updates to the standard—RFC 9580, the so-called “crypto-refresh.”

    Do you have any critiques that address the current state of the PGP protocol’s security?