baptou12

Humans can share context outside the code, while LLMs need it to be more explicitly structured inside. And if we keep the slightly verbose good practices that already help humans understand the intent, we kind of get the best of both worlds.

I agree, but “write code like a human will maintain it” can also be limiting: if LLMs reduce the cost of maintaining more explicit or verbose code, we should use that to raise the standard, not preserve compromises made for human convenience.

Good invariant !

Interesting, but this feels very economic.

It seems possible to optimize for these metrics and still end up with a society where I wouldn’t want to live if I were randomly placed in it (high stress, low trust, poor quality of life).

Would you add anything to capture that?

so social rating as a key signal ?

  But to me, what matters is the actual change to main/master at the time the thing is merged — that is what affects the team.

Totally agree, if the developper goes in wrong product direction at the beginning and add 10 additional commits afterwards: team + 6 month actually don't care about it.

  I certainly see no point in merging it in when it based on an ancient ancestor

If you want the CI to run, before the merge on the result it will produce on main branch ?