• Ephera@lemmy.ml
    link
    fedilink
    English
    arrow-up
    1
    ·
    3 hours ago

    These days they’d write “AI” and you’d be none the wiser, because even if it involves a machine learning model (which is definitely not a given either), they will still have used quite a lot of coding and algorithms to tie it all together.

    I like to mentally substitute these terms with “magic”. Because ultimately, it just means that they don’t want to explain it to you, but they need to tell you something, so you don’t ask about it.

  • intensely_human@lemm.ee
    link
    fedilink
    arrow-up
    1
    ·
    3 hours ago

    This is actually how I would start out if I were developing this algorithm. I always start with the highest-level code first, then flesh out all the details in function definitions.

    • Ephera@lemmy.ml
      link
      fedilink
      English
      arrow-up
      2
      ·
      3 hours ago

      Well, even at that level of abstraction, it’s a bit weird, because goingToCrashIntoEachOther and dont() both need the information from where a collision is going to take place, so you’d expect something to be passed into dont().

      Well, and it’s easy to dismiss this stuff as implementation details, but that if-statement needs to run as part of a loop. This loop should probably be on a separate thread, so it doesn’t get blocked by other stuff going on. Which means access to the motors needs to be behind some form of mutex, which it needs to be able to acquire fairly quickly. And then, yeah, those implementation details quickly add up to become the part that’s actually complex.

    • ByteJunk@lemmy.world
      link
      fedilink
      arrow-up
      5
      ·
      7 hours ago

      It definitely should be, but at some point in time, very intelligent people though that this was a Good Thing:

      bReadLine(bPort,&arru8NumberList)