I just had my first experience blocking an instance, and it made my realize now nice the lemmy content curation experience is vs the centralized model.

Recently I started noticing a lot of posts from that I just found annoying. There was nothing inherently wrong with them, they just came from a culture I don’t understand and so I found them cringey. Since they all came from one community, realized most of them come from the same instance. I just added that instance to my blocklist and the problem is solved!

Now think about in the centralized model. I would be forced to either just accept that these posts are in my timeline, or block each community and user individually. The instance gave me an easy way to manage my content.

I also appreciate that instances can manage the blocking for their users. So the most horrible stuff I don’t even see. But it also preserves free speech, as those users who want to say horrible things can do so in their own instance, and most people will just block it.

Anyway, just impressed again by the fediverse!

  • SorteKanin@feddit.dk
    link
    fedilink
    English
    arrow-up
    4
    arrow-down
    1
    ·
    edit-2
    12 hours ago

    they can harass you by continually sending you replies for WEEKS and WEEKS at a time (this actually happened to me twice, once with hexbear.net and another time with lemmygrad.ml).

    Honestly you really shouldn’t be on an instance that federate with those places to start with.

    You may be interested to read some of the ways that PieFed is advancing democratization of moderation.

    Personally I’m not a fan of the way PieFed uses upvotes and downvotes to basically do statistics on users in order to profile “bad actors”. That feels like karma from Reddit all over again.

    There can be many legitimate reasons why a user might downvote a lot, and a user being downvoted a lot is not necessarily problematic. They may just not be following the “hive mind”, and honestly we could use more of those users.

    • OpenStars@piefed.social
      link
      fedilink
      English
      arrow-up
      2
      ·
      edit-2
      7 hours ago

      After kbin.social went defunct the smaller instances I kept moving to (StarTrek.website and discuss.online) had not defederated from those two instances yet at that time. They do now, but a year ago it wasn’t as prevalent a choice to protect people from such things happening to their users. The Lemmy community has matured a bit, the word has gotten out, a case built to do so, and admins became receptive to what they had once spurned - e.g. I petitioned discuss.online to defederate hexbear.net after it was revealed that the admins there were caught lying to admins of other instances. But it took a lot of effort and time to get to this point, and still today most instances remain federated with lemmy.ml, which is where a bunch of Hexbear alts (e.g. Cowbee) decided to continue their trolling efforts after so many other places defederated from Hexbear itself.

      I hear what you are saying about PieFed, especially from just what is mentioned on that page regarding tools that community mods would have access to, but in practice what I’m seeing so far really is tons better. For one thing, mods on Lemmy right now have to choose between the binary options of removing content vs. allowing it to remain (similarly at the instance level it can either defederate from another instance vs. not do so), whereas PieFed offers additional options that will allow the content to remain and then place the choice of what to do into the hands of the end users. Some users may e.g. want to avoid controversial content, and so like a spam filter automatically collapse content with a certain ratio of downvotes to upvotes - which preserves the ability to see it, but putting it one step away exactly like a spoiler feature for the post. Or maybe the user will instead choose to have the content removed altogether, with the auto-hide feature? Some of my own content would have gotten removed this way, and I definitely see how things can get misunderstood by downvoters, but at the end of the day, it still gives choices to the user that otherwise would have rested solely in the hands of a moderator. Personally I have both the auto-collapse and especially auto-hide options turned off, but it’s worth noting that they are there if people want them.

      There are also other niceties such as keyword filtering - so e.g. if the user wants to remove all posts with the keyword “Musk”, then they could. Perhaps they shouldn’t, but they can, if they wanted.

      I find that user labelling really is different though, than any of the filtering options above. Examples include new users with accounts newer than let’s say a couple of weeks, a user who almost exclusively posts but never comments (likely an unregistered bot?), someone who downvotes 20x more often than upvotes, or receives 50x more downvotes than upvotes - again, their comments aren’t removed (currently I am not aware of a method to even make that happen), just labelled with an icon next to their name. These icons can help someone decide whether or not to respond, or how much detail to put into it if so. Essentially these are just measures of a user’s “reputation”, so this is the numeric version of the type of info that people use anyway? But at each moment the choice lies with the user to either pay attention to or ignore those icons. And yeah the precise formulas to determine these icons are constantly tweaked to improve them, so that’s a thing, surely.

      Ultimately though, whether the end-user makes proper usage of the tools given to them or not, either way, I think it’s awesome that these tools are given to the regular users, rather than constraining them to work solely when in the hands of a mod. That’s “democracy”! I guess whether democracy itself is a good thing or not is a whole other discussion altogether…🤣

      Edit: also the tools are definitely in their infancy - e.g. to avoid karma farming, communities should be able to use community-specific metrics so that actions taken outside of a community do not necessarily affect filtering options inside of one. Which iirc is coming or at least people are aware of this issue. PieFed itself is still mostly in the works:-).

      • skaffi@infosec.pub
        link
        fedilink
        English
        arrow-up
        1
        ·
        4 hours ago

        I don’t think there’s anything democratic about publicly singling out some users as pariahs. If a user is disruptive, just get rid of them. If they’re not, leave them be. If the jury’s still out on this one, well, don’t bias the jury against them.

        • OpenStars@piefed.social
          link
          fedilink
          English
          arrow-up
          1
          ·
          10 minutes ago

          Democracy means allowing the people to decide, as opposed to authoritianism where a central group (mods, admins, facilitators or whatever) makes all the decisions for people. PieFed still allows for authoritian control, mind you, e.g. on PieFed.social (the flagship instance) both hexbear.net and Lemmygrad.ml are defederated. But in addition to that, this opens up a new possibility where someone who is perhaps controversial yet not over a hard line can be allowed to remain, yet merely labelled.

          And then people can decide for themselves, rather than solely an authority figure, what they wish to do about it. You are free ofc to debate whether this is “good” or “bad”, but either way it does seem more “democratic” to me, bc it places the power and therefore responsibility into the hands of the user to decide.

          Which is sorta how our brains work anyway all the time, so these icons act as a shortcut to help jog people’s memories or realize something about the target that they would eventually figure out anyway.

          The point is that the icon means… whatever you want it to mean. A “new account”, a “potential unregistered bot”, a “noted controversial person” - the authorities no longer control these meanings, only the recipient. Hence “democracy”.