When someone makes a report, it can go up to 3 separate instances.
The admins of the instance of the person making the report.
The admins of the instance where the reported account is based.
The moderators and admins of the instance where the community the post was made to is located.
Where it gets complicated, is that only one report appears on each instance. And if one person marks it as complete, no one else sees it…
So, you can end up with situations where the bigot's account is on the same instance as the community they're posting in, the moderators might remove the post and ban the person from that community, but because the mods close the report, the admin never sees it, so someone that should be banned from the instance stays active, because the people capable of banning the entire account aren't aware of the problem
Or if a bot on a bigot instance is set up to automatically close all reports, none of the other instances will ever see it? That does sound like a bad implementation.
No, one report goes to each involved instance. So if the reported user and community of the post were in the same instance, a community moderator could "hide" that report from the instance admin by taking quick action. But an additional report would still go to the instance admin of the reporting user and that instance admin could choose to defederate from the problematic instance.
It's not so much even "rogue" moderators. It's just regular moderators doing their regular moderating. If they don't know that marking the report complete stops the admin from seeing it, then they'll ban the problem user from the community, and remove the post, and think that they're doing the right thing, but they have in fact, stopped the admin from seeing the report and banning them outright.
When someone makes a report, it can go up to 3 separate instances.
Where it gets complicated, is that only one report appears on each instance. And if one person marks it as complete, no one else sees it…
So, you can end up with situations where the bigot's account is on the same instance as the community they're posting in, the moderators might remove the post and ban the person from that community, but because the mods close the report, the admin never sees it, so someone that should be banned from the instance stays active, because the people capable of banning the entire account aren't aware of the problem
That's dumb, it should tell you where the report was also sent and what their outcome was
You can check it manually for now: https://lemmy.blahaj.zone/modlog/
Or if a bot on a bigot instance is set up to automatically close all reports, none of the other instances will ever see it? That does sound like a bad implementation.
No, one report goes to each involved instance. So if the reported user and community of the post were in the same instance, a community moderator could "hide" that report from the instance admin by taking quick action. But an additional report would still go to the instance admin of the reporting user and that instance admin could choose to defederate from the problematic instance.
Ah okay that's less problematic. But could it still cause problems if a rogue moderator decides to be trouble for the instance admins?
It's not so much even "rogue" moderators. It's just regular moderators doing their regular moderating. If they don't know that marking the report complete stops the admin from seeing it, then they'll ban the problem user from the community, and remove the post, and think that they're doing the right thing, but they have in fact, stopped the admin from seeing the report and banning them outright.