Reddit -> kbin.social -> kbin.run
-> kbin.earth

4th times the charm, right?

  • 4 Posts
  • 135 Comments
Joined 6 months ago
cake
Cake day: August 20th, 2024

help-circle


  • OK so I think you might be joking but in case you’re not:

    1. “They don’t need to understand DEs” and “Please teach people.” Well which is it? is it intuitive or does it need to be taught? It can’t be both

    2. That was just an example. Your solution doesn’t solve the problem I’m describing as a whole and I think my point still stands. Search might be common to most DEs but that doesn’t change the fact that they all work slightly differently, and if you want to know how to do something that can’t just be searched for, you need to know what DE you’re using. Which means knowing what a DE is. Not to mention, a user coming from a Mac wouldn’t think to just hit super anyway. It’s cmd + space there.

    3. It’s not the “proper” way, it’s just “a” way. There is no “proper” way do to this kind of thing. I would even argue that it’s not even the “best” way because you’re not learning how to navigate your OS/DE if you do it that way.

    This is exactly the kind of facetious bs “ugh, it’s not hard, just rtfm, noob” response the op is talking about







  • I’d be lying if I said I didn’t do a little of that in my younger years, but I’ve calmed down a lot. These days I generally advise caution when someone tells me they want to switch to Linux.

    I personally don’t actually think any one variant of Linux is that much harder to use than Windows or Mac. I think the difficulty comes from two things:

    One, I think people forget how much learning is involved in those OS’s as well. If you’ve ever tried to teach an elderly grandparent how to use “the computer” then you know first hand how much of this specialised knowledge you can take for granted. Simple things like knowing where to look to change mouse sensitivity as an example, are really challenging to any new user of any OS.

    Two, there isn’t just one variant of Linux. It’s biggest strength is also it’s greatest weakness here. It’s amazing that you have so many choices for your desktop environment, but that comes with the major drawback of users needing to understand what a desktop environment is, and why Googling “how to change mouse sensitivity in Linux” is probably not going to return anything useful. You have so much choice in Linux for every little thing. Down to a level of granularity that most Windows or Mac users wouldn’t even realise they’re not getting a choice in. Alsa vs pulseaudio, xorg vs wayland, not to mention the plethora of package managers. Hell even drivers for your video card: proprietary vs open source. And yes, some of those examples boil down to the old way vs the new way, but ALL of this is added complexity, which results in a steeper learning curve for a new user.

    So yeah, Linux is hard to use. The learning curve is a cliff, and anyone who thinks it’s perfect is kidding themselves! ESPECIALLY for the user who just wants to play a few games, and maybe do some browsing. We’ll never get the year of the Linux desktop with this mentality!

    I do also try to warn new users about this. It is a whole new ballgame, and it will take some effort to get up to the same level of comfort you have in Windows. It really is best to not just jump in to the deep end, and fully wipe your system on day 1.

    Start with a VM, then dual boot, and once you’ve stopped booting into WIndows in frustration, then you’re ready to commit.

    One thing I promise though, it is 100% worth the effort








  • re: The warning/grammer checking system.

    What you’re describing is called a linter, and they’ve existed for ages.

    The only way I can really think of to improve them would be to give them a full understanding of your codebase as a whole, which would require a deeper understanding than current gen AI is capable of. There might be some marginal improvements possible with current gen, but it’s not going to be groundbreaking.

    What I have found AI very useful for is basic repetitive stuff that isn’t easily automated in other ways or that I simply can’t be bothered to write again. eg: “Given this data model, generate a validated CRUD form” or “write a bash script that renames all the files in a folder to follow this pattern”

    You still need to check what it produces though because it will happily hallucinate parameters that don’t exist, or entire validation libraries that don’t exist, but it’s usually close enough to be used as a starting point.