marino

I like the idea but with a more gradual approach.

Instead of trusted and not trusted users (1/0, yes/no), something like levels (0 to 100). Users trust level can be growing with use and positive flags, and decrease with false positives or with time.

When a user mark a post as abuse, his/her trust level is added to the 'abuse level' of the post. As an example, when the post 'abuse level' reachs some threeshold, a warning to users can be shown, if level reachs higher, a moderator can be warned. If the level reachs even higher threeshold, post is hidden.

If a moderator reviews the post and find it is a positive mark, trust level of reporters is increased. If it's a false positive, trust level of reporters is decreased.

This can make that a single user with a false positive won't make the post hidden or tagged, unless is a user with a very high trust level. At the same time, a post can be marked as abuse if a few users with medium levels find it abusive.

Trust levels shouldn't be so easy to reach and I think users shouldn't know the exact level they are.

Just an idea :)

Very interesting. I didn't know beehaw.org at all but I really like what I'm reading.

I think it's a great idea to add this footnote.