• 0 Posts
  • 24 Comments
Joined 2 months ago
cake
Cake day: February 8th, 2025

help-circle











  • I’ll bait. Let’s think:

    -there are three humans who are 98% right about what they say, and where they know they might be wrong, they indicate it

    • now there is an llm (fuck capitalization, I hate the ways they are shoved everywhere that much) trained on their output

    • now llm is asked about the topic and computes the answer string

    By definition that answer string can contain all the probably-wrong things without proper indicators (“might”, “under such and such circumstances” etc)

    If you want to say 40% wrong llm means 40% wrong sources, prove me wrong









  • Looks like we still differ. If something is more complicated than what I may think, then there are some possibilities:

    • I have not learned to language properly yet (this is where I stand with C++, so no matter how many times I’ve got segfaults, now is not the time for me to say language is bad)
    • I have chosen the wrong tool (writing a videoplayer in assembler? something went way wrong here)
    • tool is actually bad (my rant above goes here. in the sake of making some things easy and simple, something basic in the language got screwed up irrevocably)

    And if I managed to try reading from a closed handle, or to access a memory that I am not actually allowed to use, or… (could not get more examples out of the top of my head), it is not the job of the language to slap my hands, as long as I follow the syntax. Most of the time (if not all the time) this means I have not accounted for something that my code allowed to happen - so my responsibility to deal with that

    What I keep hearing about Rust is still in the lines of too-much-fucking-care (that’s besides obviously dumb rule of “no more than one owner of a variable at any moment” which then had to be worked around because not everything can be done this way. please correct me if I am wrong here)