This is a follow-up from my previous thread.

The thread discussed the question of why people tend to choose proprietary microblogging platfroms (i.e. Bluesky or Threads) over the free and open source microblogging platform, Mastodon.

The reasons, summarised by @noodlejetski@lemm.ee are:

  1. marketing
  2. not having to pick the instance when registering
  3. people who have experienced Mastodon’s hermetic culture discouraging others from joining
  4. algorithms helping discover people and content to follow
  5. marketing

and I’m saying that as a firm Mastodon user and believer.

Now that we know why people move to proprietary microblogging platforms, we can also produce methods to counter this.

How do we get “normies” to adopt the Fediverse?

    • BeAware :fediverse:@social.beaware.live
      link
      fedilink
      arrow-up
      0
      ·
      edit-2
      2 months ago

      @AterNox @dch82 blocking and reporting work fine.

      However, people from corporate social media won’t report posts because in their experience, it either doesn’t get taken seriously or the admins ignore it. Corporate social media sites don’t exactly act on reports in a timely manner.

      I’m on my own instance, I moderate for myself. I don’t want slurs to exist on my instance at all. However, if I don’t see them with my own eyes, I cannot ban the user.

      PS. I’m talking about banning users that are harassing others on the instance level. These are user actions. I am an admin. I run my own instance.

      @fediverse

      • cm0002@lemmy.world
        link
        fedilink
        English
        arrow-up
        0
        ·
        2 months ago

        I’m confused, do you mean like automated enforcement rules/algorithms like big SM has? I.e. if user gets reported for breaking Y rule X amount of times ban user for Z amount of time and forward to admin for further action?

        • BeAware :fediverse:@social.beaware.live
          link
          fedilink
          arrow-up
          0
          ·
          edit-2
          2 months ago

          @cm0002 no, I want automated reports.

          A user using the n word, full on with the hard R, isn’t gonna be a good post. It should be automatically reported to me so that I can judge context and take action.

          If a user doesn’t report it, I won’t see it.

          I’m on my own instance, I am the user.

          If I don’t report it, nobody sees it.

          That’s dumb.

          @fediverse

          • cm0002@lemmy.world
            link
            fedilink
            English
            arrow-up
            0
            ·
            2 months ago

            Ah, makes sense now, that is dumb. I can totally see why they would have issues with automated enforcement, but what you described I don’t see why anyone would be against it lol