Have you seen @BryanFinster's experiments? Or @AdrianCockroft?
I'm suspicious of "fewer than X" rules since in the wild these values are power law distributed. Those "outliers" may be harder to reason about but the alternative may be worse.
I agree that these time-enduring principles (like bounded complexity, tight/loose coupling, maintainability) feel a part of the domain rather than the actor - and Information Theory and Computational Complexity are constraints on whomever's "writing a program".
I have been surprised these last few months by the way CC drops little comment blocks into its modified code. And I have almost always foudn them useful. That surprised me too.
Having said that, I do find its ability to create overly detailed git commit messages a bit of a PITA. They have a balance between "why" and "how" whereas mine are almost all "why". I haven't spent much effort tailoring that behaviour, but I know that is possible. I have banned it from doing any pushes; I do those myself
Have you seen @BryanFinster's experiments? Or @AdrianCockroft?
I'm suspicious of "fewer than X" rules since in the wild these values are power law distributed. Those "outliers" may be harder to reason about but the alternative may be worse.
I agree that these time-enduring principles (like bounded complexity, tight/loose coupling, maintainability) feel a part of the domain rather than the actor - and Information Theory and Computational Complexity are constraints on whomever's "writing a program".
I really loved this one Dave.
I have been surprised these last few months by the way CC drops little comment blocks into its modified code. And I have almost always foudn them useful. That surprised me too.
Having said that, I do find its ability to create overly detailed git commit messages a bit of a PITA. They have a balance between "why" and "how" whereas mine are almost all "why". I haven't spent much effort tailoring that behaviour, but I know that is possible. I have banned it from doing any pushes; I do those myself