See above comment thread from another discussion, first reply is from me.
Limited space to express your intent, and the quick response cycle it engenders are not conducive to productive discussion. It's hot-takes and terse replies that people interpret as hostile even if they weren't meant as such all the way down. Take that out and what makes Twitter any different than every other platform? Though, maybe they're large enough and popular enough they could weather that change.
There's many good points here. This is the best. They need to assume the content kinda sucks and let people control as much as possible, instead of thinking that AI is doing a good enough job for them.
Couple of questions:
1.Normally when you report something as a violation of the rule and the moderators determine it is not a violation that might not be objective. "Hate Speech" for example is not objective even if you provide examples.
- How would you prevent people from being punished unfairly.
- Right now a false report has no consequences. Wouldn't reporting go down since people would want to avoid a punishment?
2. In detecting mob patterns how would you determine if it's not just everyone getting on to tweet about a current event and if you rate limit it doesn't that mean some people would be censored and others not simply based on time?
3. Related to the other limits you mentioned, such as number of retweets. What prevents people from opening multiple accounts or bots?
4. Assume the above three are solved wouldn't this put the business at a disadvantage due to both decreased ad revenue and competitors who don't have these rules?