On: Why bad scientific code beats code following “best practices” (2014)

[qwertyuiop924]

That's why becoming an master requires experience: you need experience to know what's good and bad, down the line.