• 0 posts
  • 65 comments
Joined 3 years ago
Cake day: June 12th, 2023
  • Regarding the CLAUDE.md, didn’t work at all.

    Not at all? I do know that LLMs tend to have a “recency bias” and lose earlier context over time, especially when Claude summarizes a session to create a new one seamlessly. I’ve run into issues with it wanting to run git commands despite me explicitly telling it that it should never do so… Keeping sessions short seems to help.

  • That’s a much nicer read! I’m interested in some of your Claude experiences though.

    I tested Claude heavily, and I have to be direct, it consistently fell below my quality bar. Even with strict, detailed guidelines, the code it produced was riddled with subtle issues, off-by-one errors, improper error handling, and a complete disregard for my idiomatic Go patterns.

    Which Claude model were you using? Opus 3.7 & 3.8 tend to do a very good job at things like that. It catches my off-by-one issues much better than I do… They are, however, a bit slower as you point out. But my local models on an NVidia 3070 with only 8Gig of VRAM are garbage…

    Single-letter variable names might be fine for tight loops, but when you’re generating thousands of lines of code via AI, i, j, and k become impossible to trace.

    I find that if I drop in a CLAUDE.MD some style guides that it does a good job of following them. In particular I hate the Python idiom of prepending “_” to “private” variables and functions. If I put a note in there it stops it from doing it.

    As an aside - i, j, and k for loop variables is a very old tradition going back to FORTRAN where i, j, and k were always integer variables. So i is the outer loop, j the inner loop, and k the inner-inner loop. If you have more nested loops than that you’re doing it wrong…

    But if you don’t like it the style hint should help.

  • I’ll start with the unnecessarily combative tone with your reader. Then the self-contradicting absolutes (claude sucks, but my solution is x% as good as the lowest-end claude). Then the fact that it’s really just an untargeted rant against “everything” that doesn’t really justify the anger.

    And you hit my pet peeve about saying “containers aren’t running on bare metal” when they run “on bare metal” exactly as much as non-containerized processes. That’s the point of containers.

  • Yes, but some do things better than others.

    Ehhhhhh, kinda. “better” is highly subjective. Distros are “bundles of software” and a philosophy about how things are installed / when they are installed / what their default settings are / etc.

    A lot of people, especially newbs and less technical folks, grossly misunderstand what those differences are and what they mean.

    CachyOS is btrfs by default, and does btrfs better than most.

    Bullshit. Not that it’s btrfs by default but that it does it “better” than anyone else is ridiculous. It uses the same kernel driver and user-land tools as my Pop_OS install which is based on Ubuntu 24.04 and which, believe it or not, is running btrfs just fine.

    Btrfs is a bit more complex than your vanilla ext4

    Kid - I remember when ext4 was released. Very exciting to have a journaling filesystem at the time…

    Bazzite excels at gaming related things. Alpine at lightweight stuff, Nix at inmutability, etc…

    “excels” at meaning “has steam installed by default” and “makes nvidia drivers easier to install” you mean.

    My Pop_OS laptop runs game just fine.

    All these distros with their various design goals and bundled packages are variations on a theme. Like different Lego sets that include parts from a common box. But when you use the “btrfs” block you use the same one everyone else does.

    And none of them will deal gracefully with a failing disk that the OS has been told not to ignore errors on.

  • I hate when people here complain of “bloat” since often it just means “there’s this one thing I don’t use and it irrationally offends me by its existence”.

    You can uninstall just about anything you don’t want. So very much not like Windows in that regard.

    Are you running out of space? Out of memory? Or simply offended that some functionality you don’t like exists?

  • So - I setup that model according to the docs and gave it this prompt:

    Write me a highly optimized n-queens solver in go. It should take advantage of parallelism (what little there is) and output only the solution and how long it took.
    

    After 10 minutes it gave me code that didn’t compile.

    It took another 3 mins to fix the compile error and the output is not correct.

    As I said - LLMs on 8Gig VRAM just aren’t worth it.

  • It just doesn’t seem logical to me to lock me out of the whole system because a non-essential part is not working/present.

    It’s a fair criticism - but the system doesn’t know they are non-critical. You’ve, effectively, told it they are critical in fact.

    You can add a nofail option to the drives that aren’t needed to boot in /etc/fstab.

    e.g.

    UUID=9d1f7f1b-2a78-4b92-a8d5-0b05e7273920 /mnt/backup  btrfs   defaults,nofail,subvol=@backups 0 0
    

    I use this for a disk that’s external and sometimes not connected.

  • just want it to work,

    Unless you’re doing software development and using lots of tools that need dbus or ever plan to do anything other than “use a browser and flatpaks”. Then you get to figure out how to fight with distrobox to make things interoperate with your desktop environment and other distroboxes.

    And I doubt bazzite would have dealt as well with a missing disk.

  • Because they’re clearly looking for a way to appear to be doing the job without having to do the job.

    There’s no way this forum can help with this question without knowing a lot more about the company, what they do, what they want/need, etc.

    If they truly are looking for help from this forum of ignorant strangers then they’re an idiot and should just be fired anyway.