I think both of you are in more agreement than opposition.

I understand the comparison but you can't exactly mute people with minimal effort in real life. Additionally, the threat of rape in person is significantly different than anonymously online from a legal perspective because the person making the threat knows who and where the target is at the moment the threat is made.

At a high level I don't disagree with most of what you're saying. The point I'm making is that there's a pretty large gap between "something should be done about online harassment" and "this is our plan for stopping online harassment". Most calls for action appeal to the first without much concern for the second, and the solution is the difficult part, not identifying the problem.

Sure, but who decides who the assholes are? What standard of proof do they use? What happens to people who almost meet that definition but don't quite? Some cases are pretty straightforward but many have too much grey area for a simple concept like 'ban all the assholes' to hold up without a huge amount of effort that often can't be provided effectively.

Evidence is key and that can be difficult to get. Additionally, players often edit clips to get the reaction they want from moderators or the community as a whole. The video in the linked article could potentially be a great example of this. The streamer obviously wants you to think they were attacked unprovoked but is that really what happened? We have no context for what led to the recorded exchange. Did the streamer refer to him using a racial slur for failing to hold what they thought was the proper position during the round immediately before the video begins? That doesn't excuse what the player said in response but it does change the context significantly. If you ban the rape guy and then he releases a video showing the streamer saying even more outrageous things do you reverse the ban, ban them both, or give them both a warning not to be dicks in the future? This is not an isolated incident either. Multiple situations just like this happen every day on platforms with any significant number of users. How do you give each one the time it requires to be resolved correctly? Will you have any players left if you ban everyone who offends someone else?

The point is, policing people's behavior is very challenging. There are tons of ways to abuse any system you can design unless you record literally everything your users do and that comes with it's own set of moral and logistical issues. The simplest and most universally applicable solution is to enable users to block other players themselves. Making that impact matchmaking is kind of dumb in my opinion as it just opens up the potential for abuse as you noted. Set up a few very simple ground rules and then let players sort out who they want to be able to communicate with based on how they treat them. That's the only solution I can see that is realistic and sustainable but it requires users to take an active role in maintaining the community, which they should be doing anyway if they want it to be the kind of place you're describing.

For most people gaming is a social activity. The popularity and prevalence of multiplayer games vs single player games bears this out. Playing single player games is not a viable solution to avoiding harassment for people who are interested in the social and/or multiplayer aspects of gaming. Muting people who are dicks is a viable solution and that's exactly what I've been advocating for in this discussion. Many others seem to think they shouldn't need to be involved in the process and game devs or other communication platforms should do all the work for them. I don't think that's a realistic suggestion.