• 1 Post
  • 323 Comments
Joined 2 years ago
cake
Cake day: March 22nd, 2024

help-circle

  • But why?

    Wouldn’t it be better to spend the same effort writing ffmpeg modules and interfaces in Rust?

    keeping external dependencies to a minimum

    This is… concerning, too.

    Media processing code is difficult. It’s not even a pure coding problem, and often involves human perception, extensive, expensive experimentation and esoteric, buggy hardware APIs . Hence the whole point of ffmpeg is basically integration of external libraries, with immense amounts of labor already put into each.

    There are some Rust libraries they could pull in though. I guess it’d be reasonable to focus on newer formats/codecs that have Rust implementations already, and let ffmpeg handle weird legacy formats.






  • brucethemoose@lemmy.worldtoProgramming@programming.devLLMS Are Not Fun
    link
    fedilink
    arrow-up
    14
    arrow-down
    3
    ·
    edit-2
    14 days ago

    Mmmmm. Pure “prompt engineering” feels soulless to me. And you have zero control over the endpoint, so changes on their end can break your prompt at any time.

    Messing with logprobs and raw completion syntax was fun, but the US proprietary models took that away. Even sampling is kind of restricted now, and primitive compared to what’s been developed in open source.


  • brucethemoose@lemmy.worldtoProgramming@programming.devLLMS Are Not Fun
    link
    fedilink
    arrow-up
    34
    arrow-down
    9
    ·
    edit-2
    14 days ago

    If you think of LLMs as an extra teammate, there’s no fun in managing them either. Nurturing the personal growth of an LLM is an obvious waste of time. Micromanaging them, watching to preempt slop and derailment, is frustrating and rage-inducing.

    Finetuning LLMs for niche tasks is fun. It’s explorative, creative, cumulitive, and scratches a ‘must optimize’ part of my brain. It feels like you’re actually building and personalizing something, and teaches you how they work and where they fail, like making any good program or tool. It feels you’re part of a niche ‘old internet’ hacking community, not in the maw of Big Tech.

    Using proprietary LLMs over APIs is indeed soul crushing. IMO this is why devs who have to use LLMs should strive to run finetunable, open weights models where they work, even if they aren’t as good as Claude Code.

    But I think most don’t know they exist. Or had a terrible experience with terrible ollama defaults, hence assume that must be what the open model ecosystem is like.