Compared to the constraints of the physical world, the universe of computability is essentially limitless. On top of that, there are almost zero objective fact

Compared to the constraints of the physical world, the universe of computability is essentially limitless. On top of that, there are almost zero objective facts in the field. Faster and smaller are about the only factual statements that can be made about computability. Everything else is an opinion. With no physical truth to disprove wrong opinions, people get accustomed to putting forward passionate arguments without every worrying about being proven wrong. Finally, it usually takes significant effort to even have a few sample cases to compare opinions when it comes to coding. And usually it's not under scientific controlled environment. There are no two decades long arguments about "does a ball roll down a hill". Noone is taking a strong opinion against. Before it would get that far, someone would just put a ball on a hill and see the end result. You can't do that with software, the problem space is too large and there is no little bearing to an objective way to measure to results, or they often aren't generally applicable.