Re:Oops! We Automated Bullshit
by trimethylpurine on Friday November 17, 2023 @09:27AM
(#38304045)
>For instance when you ask it for a snippet of code and it gets it right.
Does this happen often for you?
I've logged my success rate with Go and C.
32/37 responses failed to meet the requested criteria. 15 wouldn't even compile.
Getting the bot to do it right takes longer than writing it myself.
Maybe that's just me?
I wouldn't think so. Word prediction and writing code aren't nearly the same thing, when I think about them. Sure, they both require language. But one of those also requires reasoning. As worded in this paragraph, could an LLM tell me which one? Sure. Sometimes...
Does this happen often for you?
I've logged my success rate with Go and C.
32/37 responses failed to meet the requested criteria. 15 wouldn't even compile.
Getting the bot to do it right takes longer than writing it myself.
Maybe that's just me?
I wouldn't think so. Word prediction and writing code aren't nearly the same thing, when I think about them. Sure, they both require language. But one of those also requires reasoning. As worded in this paragraph, could an LLM tell me which one? Sure. Sometimes...