It’s cool and all, but I do worry about this instinctive reaching for AI to build convoluted answers to stuff like this, which is easily demonstrated by, say, two balls and a lamp, not to mention the myriad of astronomical simulators that already exist.

The software is being written to teach the user something, but it’s not novel software, or a novel problem, and the creator doesn’t actually care about the process of making the software, or actually thinking about how to solve the question. But they have access to an automatic wheel reinvention machine and so...

I think you're overstating your case. I never said this was novel. I didn't vibe code any of this. Two balls and a lamp were not sufficient to make this clear to me, but writing the simulator was.

The LLMs have read all of the internet, and are thus quite good at answering questions like this, and I have zero problems asking them about it. In THIS case, the answers available on the internet kinda suck, which made the LLM suck as well.

> is easily demonstrated by, say, two balls and a lamp,

...have done this many times for friends! A torch and a yoga ball work nicely ;)