We see the need for decentralised anti-abuse and reputation systems as the single biggest existential threat to the growth of Matrix.org (which otherwise risks being almost worse than SMTP). It's a problem that the whole decentralised web community has in common, and a solution could benefit everyone equally. We've experimented with data entries in Stellar ledgers for this already, but it needs a lot more thought & work. We tried to issue a call to arms at FOSDEM this year to try to get people thinking more about it, with a bunch of possible ideas very similar to this post (see slides 27 to 32 of https://matrix.org/blog/wp-content/uploads/2017/02/2017-02-0...) and it would also be absolutely awesome if the W3C social groups could help drive a solution.
That's great to hear that the Matrix community cares about this so much, and is thinking along similar lines! (The fact that you're already exploring the Stellar approach makes me feel a little bit less crazy for thinking some of these things.) I agree, I hope we can work together... it really does face everyone in this space.
I linked your comment and slides from the blogpost now, btw!
I think the only practical solution is an old one, of using multiple spaces - a bunch of relatively closed spaces that you have to get invited to, with more open spaces that you might get invited from.
Closed doesn't have to mean secret, just that it uses a whitelist for who can post.
Slap on some access controls to my idea here and you might have a decent start;
I have to agree. I don't think that 'global moderation' really works as a concept. What someone views as just a joke might be incredibly offensive to me. I have no problem with rape jokes or 9/11 jokes for example, but I never ever want to see any sort of gore, that's completely offlimits for me. Whereas some people (see 4chan) might not care about either, any many care about both.
