Daniel Frost - Refactoring the Mental Bin: An Extension in the way

This is an article that is a little bit obnoxius really. I draw up some opionions here that I don't initially, when writing code, always comply with. But I would like to comply with at all times. But I guess that's another story.

I remembered a talk by a former professional friend who had made a statement about arrays and performance. Another professional friend had once said, "stupid code is fast code." I was left pondering these two statements while reading the code.

Code readability is subjective. How someone else perceives code is different from how I do, and therefore, it's of course very important that, when working on a team, such things are handled by using some kind of tooling.

You should, of course, not be "allowed" to use one style of code while your colleague uses a different one; that will most likely lead to confusion and frustration. Program correctness, on the other hand, is something you can objectively measure — "if your specifications are contradictory, life is very easy, for then you know that no program will satisfy them, so, make "no program"; if your specifications are ambiguous, the greater the ambiguity, the easier the specifications are to satisfy (if the specifications are absolutely ambiguous, every program will satisfy them!).".

Having consistent, readable code alone should lower the cognitive load while reading that code. It's about context switching, they say. We all know how it feels when we enter a module with neatly organized methods, naming conventions, sizing, and documentation—and then switch to a different module that doesn’t fit into the mental model you were just in. You then need to switch gears and figure out what's going on.

Leave a Comment