I see, that sounds sensible. But I believe you've also mentioned a mechanism where if 5 people I follow, block someone, that user will also be blocked for me. Has that stopped?

Replies (2)

Correct. So separate from the automated anti-spam filter (which is an Amethyst thing), there is the usual Nostr reporting system with NIP-56. In that system, people can report you for Spam, Scams, Explicit Content, Impersonation, and/or Illegal Behaviour. Anyone can report you or your posts at any time. Reports are saved on relays and might last forever. Many relays use that information to delete posts they don't want on their servers. Your reports only affect the people that are following you. When you report content, you add a warning sign for your followers. BUT that warning sign only shows up if 5+ accounts that a user follows reported the same note. So, alone, a reporter cannot do much damage. But if there is some agreement among the people a user follows, that user will see a warning post that shows who, among the people he/she follows, reported the post and a "Show Anyway" button. At that point, they can agree with their follows and also report that content or simply hide the reported user from their view entirely, OR they can disagree and report the reporter. If the report is not for a post, but for a user, a similar 5-reports-among-follows happens. BUT, for users, we cap the report in 24hrs. So, if your user (and not a post) gets reported today, tomorrow the followers of the reporter will start seeing your posts without the warning again. The reporting system does not hide posts from global feeds or chats, it's only between the user and the people that user follows. For instance, I have 100s of reports on my user. Many of them were deleted by relays that evaluate reports. But I believe 108 are left out there. Those reports only affect the followers of the reporting accounts. Since they are mostly random accounts, it doesn't affect anyone.
Thank you for the explanation. I think it is mostly the 'reported users' which I find a bit confusing and iffy. But you are saying all clients do this? Seems pretty weird to me that a user would be hidden if they said something somewhat (seen as) controversial or unpopular, and pretty easy to abuse as a system to block/silence users if there ever were to be bad actors collaborating (in a worse case scenario). Or just a very realistic scenario you disagree with the personal or political opinions of 5 prominent users...? But this is protocol-wide you say? That makes me feel a bit hmmm something lol:/ Am I completely misreading this?? I'm mostly an egirl but I'm wearing my investigative journalist serious hat atm