The line between outright fraud, bad methods correctly implemented, messy data, and implementation bugs is fuzzy. Trying to criminalize anything not very very clearly #1 quickly turns into a case of “show me the man and I’ll show you the crime”. You think groupthink in academia is bad just wait until professional disputes lead to jail time for the loser.
The fact that some areas are gray shouldn't prevent us from demanding legal consequences when the fraud is gross and deliberate, as appears to be the case here.
