• Lvxferre
    link
    fedilink
    English
    arrow-up
    1
    ·
    11 months ago

    Distributed system? Madman! You’re going a step further! Mad respect for that, seriously. Now I want to see your project to get successful.

    Regarding moderation, did you see this text? I feel like it’s perhaps worth a try; I don’t expect it to devolve into web of trust-like “feuds” as there’ll be always people working as links between multiple groups, but it also prevents the “first come, first served” issue that you mentioned.

    • sugar_in_your_tea@sh.itjust.works
      link
      fedilink
      English
      arrow-up
      2
      ·
      11 months ago

      Great article! I was thinking along these lines, so I’m glad to see a formalized version of it.

      What if participants could automatically block the malicious peer, if they discover that the peer has been blocked by someone the participant trusts?

      That’s essentially what I’m after. Here’s the basic mechanism I’ve been considering:

      1. Users report posts, which builds trust with other users that reported that post
      2. Users vote on posts, which builds trust with other users that voted the same way
      3. Posts are removed for a given user if enough trusted people from #1 reported it
      4. Ranking of posts is based largely on #2, as well as suggestions for new communities
      5. Users can review a moderation log periodically (like Steam’s recommendation queue) to refine their moderation experience (e.g. agree or disagree with reports), and they can disable moderation entirely

      And since content needs to be stored on peoples’ machines, users would be less likely to host posts they disagree with, so hopefully very unpopular posts disappear (e.g. CSAM).

      So I’m glad this is formalized, I can probably learn quite a bit from it.