>>13637039I'm not a physicist or engineer, I'm a statistician, so I don't really build anything (other than "models"). My opinion on the whole "build from first principles" idea is that you can't just follow some general idea and hope that it always works. In fact, a lot of things that are true don't really make sense, and sometimes things that make sense are not true. I say this purely based off of what I've seen. I've worked with theoretical Economists who have very elegant theories about how some social phenomenon works, but then the data doesn't really fit it at all.
What I'm basically trying to say is that you shouldn't pay too much attention to maxims like the one you mentioned, and just analyze case by case what is the most appropriate thing to do. The unfortunate reality is that we have limited minds in an infinitely complex world. But I'm a nobody, so obviously make your own opinions (and also don't know jack shit about building things).