← Back to context

Comment by greazy

6 days ago

> Lately I’ve seen a push toward minimal or nonexistent code review, with the claim that we should move from “validating architecture” to “validating behavior.”

This is very bad. A programmers output is code, when we begin to measure the output of code as the primary product of a programmers work we will fall into a death spiral of unmaintainable code. I still strongly believe we write code for other humans, not for machines. Yes, yes we need to solve problems and you may argue that is the true measure of code. However this argument ignores maintenance and continued developmental requirements.