I make things: electronics and software and music and stories and all sorts of other things.

  • 1 Post
  • 183 Comments
Joined 3 years ago
cake
Cake day: June 14th, 2023

help-circle

  • I freaking love niri

    It’s easy to lose windows that are off-screen

    By default. They should really change the defaults, but in config, you can make it so if there is a window to the side hidden, a piece is always visible.

    It’s called struts.

    layout {
        ...
        struts {
             left 16
             right 16
         }
    }
    

    It ensures a piece of the window is always visible

    Plus with a bar like from QuickShell or WayBar you can see all open windows anyway up there.

    To achieve this, I’ve been using Niri with the tilemod script. It helps make Niri behave more like a regular tiling.

    The things you show in the gif are already possible with niri via Super + [ and Super + ] (default bindings) for placing into columns

    It has limitations, but everything in the gif at least is already there. Niri actually has quite a bit of support for base tiling.

    I can’t go back to non-scrolling ever. I even use a similar setup on my work machine (a Mac) via PaperWM.spoon, tho it isn’t as well integrated (bc it’s a mac and you’re forcing it to not be a mac)

    EDIT: One tiling feature of niri I love is that with Super + W you can switch to a stacked mode like in a tiler. For KiCAD PCB editing, you have three windows - project, schematic, and pcb. It is essential that at the very least the project tab be stacked behind and out of the way imo, since there is no minimize


  • Not sure if it helps, but traditionally the max was 80 columns by 25 rows on old computers, and that became a standard.

    Nowadays we can be much bigger with a terminal emulator window, but 80x25 is still the “standard” and most terminal emulators default to that size when you open their window.

    It’s not the max size possible, but if you’re making software for terminal, that really should be your target size imo








  • I wonder how many Arch haters have actually used Arch and ever ran into a breaking issue.

    In my 8 years of daily driving Arch, I have had 1 ever, and it was only bad bc I didn’t know of BTRFS yet.

    That happened 3 years in, and I went on a new distro-hopping spree in between, trying Fedora, NixOS, and others, and they had actual problems. Not “oh an update broke my system and was fixed in a day.”

    So now I’ve been back on Arch for more years.

    This whole “Arch is always breaking and takes so much maintenance” is a myth. In reality, Arch is the most stable, because it packages everything Vanilla. Everything works the “standard” way. On Debian, things decay, and then some stuff that’s new breaks with some stuff that’s old. On Fedora, there is a Red-Hat tax: configs in weird places, strange scripts, software missing from main repos (COPR sucks btw). On Arch, it’s just “normal” Linux; and that works


  • I’m not really talking about teaching and guiding though.

    If you look at the meme, one of the responses is something like “PRs welcome.”

    I think that’s appropriate in certain circumstances.

    You are correct, there is nothing wrong with a lack of knowledge or dumb questions. I’m referring more to how they ask their dumb questions.

    Some newbies come with hostility, pride, and entitlement, not ignorance.

    They request new features, say something sucks bc of blah blah blah, and then get mad at people responding rudely back. But what else should we respond with in this kind of situation? They aren’t looking to learn. They are looking to degrade projects, users, and maintainers.

    They aren’t looking for answers; they’re looking to be mad about something and confirm their opinion that “Linux sucks” or some other such bias.

    So yeah, when someone comes humbly even with bad assumptions, absolutely correct and guide, but there’s no point guiding people who don’t actually want guidance, ya know? Pearls before swine

    So sometimes these aren’t bad responses







  • Pros:

    • Better UX for system. On Windows and Mac you’re stuck with the old-style window movement desktops which suck and are chaos with lots of effort to maintain when trying to do anything productive. Even with third-party tools it’s nothing like being able to have something like Niri. It’s just better on the Linux world.
    • It actually meets your needs instead of making your change your needs. The customization is insane, but usually you don’t even have to go very far to be happy.
    • No bloat, no forced features like AI (tho you can get your own AI stuff if you want), less privacy concerns, better tools generally

    Microsoft Windows is actually a complete piece of crap

    Unfortunately, cons of switching:

    • Have to use non-standard apps for certain things like office files or DAWs for audio production
      • Tho, these are less bad these days. In fact, some are better like I’ll take Blender over Maya any day, and I use OnlyOffice even on Windows, bc why pay for the “Copilot App” (formerly MS Office) when OnlyOffice is just as good and is fully compatible?
    • Some games don’t work (esp multiplayer), and sometimes for no good reason other than the devs don’t like Linux users, e.g. Bungie.
      • Also far better than it used to be. Very few things I miss out on
    • MacOS clearly has the better app distribution system. A single folder with all necessary deps save a couple core libraries. Simple, effective, can still be put in a store. Instead we have an obsession with sandboxing or overcomplicated packages. AppImages were so close to being right. But nope. We can’t have nice things sometimes

    I’ve been using Linux-only since around 2019 (having used it alongside Windows for gaming before then) when Proton finally started getting good. I’m also an engineer, artist, writer, gamer, musician, maker, and more, so I feel like I have touched a lot of the different ways in which computers are used. I’ve used several distros for extended periods, and my fav is Arch (tho Nix is a close second; it’s just not quite ready for primetime)

    Linux is absolutely a viable alternative, but you have to know what you want from your PC. How do you want it to function? Pick that choice. It’s not ice-cream flavors where all are equal and you have to decide; form follows function here. Decide what you need and then build your own system from that - bc Linux is yours. Refuse to be spoonfed slop no matter if the slop is from Microsoft or from a Linux distro. Slop is slop. Cook your own meal. It tastes better. If you don’t like mushrooms, don’t get mushrooms.


  • FreeBSD is not Linux, it is Unix.

    See the “BSD?” That stands for “Berkley Software Distribution” which was the Unix that was used at UC Berkley back in the day. It uses a Unix kernel not the Linux kernel.

    There is certainly overlap between the worlds of modern Unix and Linux, since much of the Unix world benefits from the popularity of Linux and FOSS as a whole, but it still is its own thing.

    Namely, the biggest difference is that FreeBSD is a complete OS. All components other than a user’s personally installed apps are made by (or at least integrated and maintained by) the FreeBSD team itself. Linux is technically just a kernel, and a Linux distribution, while similar to a Unix distribution, is made up of many many many moving parts made by lots of different people, each piece with its own goals then steered into working together.

    Now, there are some exceptions. There is a bit of a blur now. These days, the BSDs have more 3rd-party parts and the Linux world has become more governed and polished, but they are ultimately different things with different goals.

    There are plenty of other BSD descendents including OpenBSD and to a lesser extent Darwin (aka the macOS core, which runs an XNU kernel based on the Mach microkernel with a BSD subsystem beside it)



  • The future of AI in Ubuntu [Except it’s Internet and Slackware in 1996 instead of AI and Ubuntu in 2026]

    As 1996 progresses, internet-based tools are becoming more and more ubiquitous. Adoption across the tech industry has been mixed, both in terms of which projects are embracing “Web” technologies, and in how companies are structuring their adoption. As a result, I’m frequently asked about what Slackware will do (or not) to incorporate networks.

    In this post I’ll detail how the internet will play a part in Slackware’s future, my framework for classifying internet features in the OS, and how Slackware is currently approaching adoption internally, because I think that will help paint a picture of our intent.

    The bottom line is that Slackware is ramping up its use of Internet tools in a focused and principled manner that favours open network tools with license terms that feel most compatible with our values, combined with open source contracts. Internet features will be landing in Slackware throughout the next year as we feel that they’re of sufficient maturity and quality, with a bias towards private networks by default.

    Internet features in Slackware will come in two forms: first as a means of enhancing existing OS functionality with networking in the background, and latterly in the form of “Internet native” features and workflows for those who want them.