What if someone considers you an "asshat"? What should they do?

Sounds like the real issue is scale and anonymity then. What solution do you propose?

Self service moderation which serves as training for machine learning algorithms. Perhaps a system where messages aren’t whitelisted for the entire platform by default but instead must pass scrutiny as a canary. At first you could seed this with employees for users with no audience. During this time the sentiments of those who view the message are used to determine if wider distribution is desired.

Part of that is that the average age of participants on the internet has dropped very significantly since mobile applications become ubiquitous.