Don't get me started, I have seem repos that I'm fairly sure never ran in their presented form. A guy in our lab thinks authors purposefully mess up their code when publishing on GitHub to make it harder to replicate. I'm starting to come around on his theory.