De-escalating social media conflict

From my time on twitter, it definitely seems to me that the algorithm is designed to show the most conflict and vitriol possible. I don't really have friends who use Twitter; I'm mostly looking at accounts with thousands of followers. When I click on a tweet from Trump, I mostly see left-wing people attacking him, but when I click on a tweet from Hillary or AOC, I mostly see Trumpers attacking them. Every damn time. I know that that is not an accurate portrayal of the overall responses. I know Trump gets a fuckload of praise for just like taking a dump on the keyboard. So the algorithm is not showing me tweets I'll like/agree with, it's showing me discord, anger, and hatred.

I recently came across this tweet from Megan Amram, a comedian I follow: https://twitter.com/meganamram/status/1273446578926219265 in which she posts a letter of apology for some recently-surfaced offensive tweets from years ago. If you notice, most of the responses are not accepting her apology at all. I was not aware of any of this until seeing this tweet, but just from reading this letter, they do seem to have a point. Several points.

I think that one takeaway from this is that in your proposal, the offending tweeter just hits an "I'm sorry" button and that's it, which is fine if they just said Mario isn't Italian. But if a politician denigrates an entire race or accidentally tweets a dickpic he meant to dm to an underage girl...well actually there's just no response for that--but you know, if it's a famous person saying something unacceptable, then really there is a need for some explanation, repentance, and at least their own words.

Also, I think in such instances, their followers/the public has a right to respond. With more than just hitting a forgive button. Perhaps shutting down replies but allowing people to retweet with comment would be okay.

I think the thing that's missing here is celebrities/politicians/corporate people/etc. versus the little guy. Here's another example.

I once got into a bit of a heated thing with Jameela Jamil. It was late at night, I'm a random nobody with 500 followers (and only because I was a small-town reporter briefly) and more than a little buzz on, and I reply to an international celebrity's tweet. This is already super long, so I'll just say I think one word I used was misleading but it wasn't hurtful or out of line or anything. Never in a million years would I have expected to be in my pajamas alone in my apartment at midnight having a tense back-and-forth with this woman whose attractiveness was a running joke in a popular prime-time TV show. It wasn't long before I apologized, and then she said that we should both delete our tweets now "to avoid you getting piled upon" and immediately deleted hers. But she strategically left up the tweet about deleting the exchange, so then a flock of admirers, having no idea what had occurred, came to praise her for what a saint she was for doing that for what must have been some horrible evil troll.

The point is that running afoul of a woman put on a pedestal by 1.1 M people on Twitter could have been pure hell for me. I'm mentally ill (trauma issues and such) and experience debilitating shame on a daily basis, and I should not have been drinking, so a tweet storm of judgment and hatred could have actually caused a serious crisis.

In conclusion: I love any idea of helping people to be more caring, to reduce the vitriol, ugliness, and cruelty of online social interaction, but I think your plan is kind of missing the bigger picture, and also assumes that the problem is only the humans. You guys are the hackers here, but I am quite certain that twitter is designed to promote the hating, the dramatic, the denigrating, so that people have to keep coming back to show those idiots what big idiots they are, and to prove how right you are.

So, we should start by not assuming that. Especially when large numbers of people agree with us.

For example stop shadow banning people without warnings, make dead posts always visible but with disabled comments, unless vouched for, make moderation visible, I would like to know why something disappeared, create public spaces to resolve disputes between moderators and users: it's easy to misunderstand something when you are from 1% SV and the other person is from a developing country, for example, moderators are wrong too, hiding mistakes is not de-escalation.

Last but not least, changing the rules around "you're posting too fast" I share my public network connection (and public IP) with tens of other people and usually can't comment because it says I'm posting too fast.

Someone could be very frustrated by this behavior and become increasingly angry, this is not good de-escalation either.

I've never seen an "I've made a mistake" here on HN for example

How do I know you're certainly not from Southern Europe?)

There are a lot of unreasonable people out there. There are also people who get a kick out of causing a scene. All it takes is one Buzzfeed (or whomever...) article about this exact situation to make other advertisers run to the hills.

It's what YouTube's Adpocalypse was all about.