What to do to contain this kind of thing? Is there a chance I can take only the signaling from him?
AI-generated summary
The discussion revolves around containing a troll member who is flagging everyone’s posts. twofoursixeight and MikeNolan suggest demoting the user to Trust Level 0 (TL0) and locking them at that level to prevent further flagging. MikeNolan highlights the vulnerability of self-moderation to abuse by a few users and proposes requiring membership in specific groups for trusted users to flag posts. Heliosurge recommends checking email domains for disposable addresses and using the Fingerprint browser plugin to help identify users with multiple accounts.
Demote the user to TL0, this will prevent the user from making flags.
Be sure to lock that user at TL0.
That’s the fly in the self-moderation concept, a few people, possibly just one, can make the system essentially unworkable. Another service (DuoLingo) would hide posts if 5 people flagged them, there were several people suspected of having multiple accounts (from different IP addresses) and they would flag posts from certain users or on certain issues within a few minutes of when they were posted. (That’s a major reason why I stopped using that service.)
Requiring membership in specific groups for those trusted enough to flag posts would be one way around that, with the default being ‘trust level 1’, but a separate by-invitation-only group could be used to control who has that ability.
If you suspect a user using multiple accounts it is good to check email domain to see if disposable. Another tool/plugin that can help is Fingerprint browser plugin.