I am guilty of using trigraphs on a day to day basis. It is useful tool to make code more readable but you really have to restrict yourself to only use it for v

I am guilty of using trigraphs on a day to day basis. It is useful tool to make code more readable but you really have to restrict yourself to only use it for very simple and easily recognizable idioms if it is to work as intended (make it more readable). Sadly, most devs (especially at large corporations) don't go out of their way to pay attention to such details and so overall I'm in favor of rules that restrict constructs that can very easily be abused. As my career progresses and I'm spending more of my time reading, evaluating and diagnosing code, I got a new perspective on being smart when writing code. Now I think, if you are smart, you will write most plain code possible. If your code looks smart then you are probably doing something wrong. There is very few problems that actually require smart looking code.