But there are at least two problems. One is subjectivity. As in "terrorist" vs "freedom fighter". The other is that there are lots of assholes out there. Billions of them. It's like spam. You just gotta filter it out.
I think everyone can agree that the most egregious examples of harassment are bad. Sending someone a new picture of gore from a newly created Twitter account every hour for years (or whatever example is trotted out every time this comes up) or scouring DNS records for their address and swatting someone? Yeah obviously that's bad, we all know that's bad. Those are harassment to the level of being criminal offences anyway, there are ways to deal with them.
And I think that everyone agrees that telling someone that their ideas are 'stupid and naive' is perhaps a little hostile, but not to the extent that it needs to be censored.
So what's the middle ground? Are we allowed to swear? Are we going to apply the Americans' ideas of what is offensive to the internet? Because I sure as fuck don't come from America and I sure as fuck don't appreciate being told what is and is not acceptable from a country where it's illegal to say the word 'shit' on TV but it's legal for the Westboro Baptist Church to harass people mourning at the military funerals of gay soldiers. You can sort your own shit out before you start policing the language and behaviour of people in other countries talking to other people from other countries over the international medium that is the internet.
If you want to filter offensive language from your twitter replies, go ahead. Twitter and Facebook are already incredibly filtered thought bubbles, that ship has sailed. It's not going to get any worse if you can automatically filter out everything with language you find 'offensive'.
But this 'let's build a future internet without assholes' stuff is pathetic. What I consider an arsehole, you might consider someone just defending their pronouns or whatever it was that kid stabbed his teacher for the other day.