prxm

i see. i guess its just my view that is different. i believe the value is in the process of producing, not driving an agent to get something to exist. because if im not learning anything new, is it really me that is producing? as long as you just keep prompting an llm at every opportunity you get, you will be at the exact same spot you are right now and may be easily outraced by someone else who just as well prompts an ai. i believe learning is perpetual no matter the amount of experience. now, are we going to achieve agi? will this really matter then? i guess that is another topic for later.

what is your metric for "productivity" exactly? generating more and more lines of code? we have always had a way to do that which is called copy pasting. now what we have is a tool that vaguely understands the problem at hand and performing the copy pasting in a stochastic manner. do you get your llm to bring a novel solution to a problem that hasnt been explored before? otherwise, i seem to fail to see the value in this process. what is it that you are getting out of it that makes you "more productive". do you even learn anything new when you are not even the one doing the work. what is the point of anything at this point.

This is one of my favourite activites with LLMs as well. After implementing some sort of idea for an algorithm, I try seeing what an LLM would come up with. I hint it as well and push it in the correct direction with many iterations but never tell the most ideal one. And as a matter of fact they can never reach the quality I did with my initial implementation.

> Zig is explicit

i never got this point. whats stopping me from writing a function like this in zig?

  fn very_bad_func() !i32 {
      var GPA = std.heap.GeneralPurposeAllocator(.{}){};
      var gpa = GPA.allocator();
      var s = try gpa.alloc(i32, 1000);
      s[0] = 7;
      return s[0];
  }

the only thing explicit about zig approach is having ready-to-use allocator definitons in the std library. if you excluded std library and write your own allocators, you could have an even better api in rust compared to zig thanks to actual shared behaviour features (traits). explicit allocation is a library feature, not a language feature.

can you elaborate on what you mean by "move the borrow around registers", an example maybe? what other things are you experienced that are struggling doing in rust?