Spambots are abusing this fact

.ml admins fix pls

  • velox_vulnus@lemmy.ml
    link
    fedilink
    English
    arrow-up
    38
    arrow-down
    1
    ·
    9 months ago

    The spammers are taking advantage of open fediverse instances - and it’s not one, but instances chosen at random. You may think that banning instances would solve it, but it will block other legit users.

    • m-p{3}@lemmy.ca
      link
      fedilink
      arrow-up
      8
      arrow-down
      1
      ·
      9 months ago

      Some of the automod bots are also filtering and blocking posts based on the content as well (RegEx, confidence rating based on the age of the account and upvote/downvote ratio, presence of rarely seen domains in URLs), so that helps a bit.

      • MrKaplan@lemmy.world
        link
        fedilink
        English
        arrow-up
        5
        ·
        9 months ago

        Most of these are unfortunately limited to local instance moderation, which means all instances that don’t run these bots don’t benefit from them.

        The posts have already been modified in ways that they aren’t as easy to reliably filter anymore, though still possible with fairly low false positive rate.

        To add to this, depending on how content is removed, removal may or may not federate properly. on Lemmy.World, we’ve been removing content in a way that reliably federates, so while a lot of this spam does arrive in !asklemmy@lemmy.world and !selfhosted@lemmy.world, the removals on Lemmy.World should federate to all other instances (0.18.5+).

        The other portion of the spam is mostly on Lemmy.ml, in !asklemmy@lemmy.ml and !opensource@lemmy.ml, and not all of their removals have been done in a way that federates.