• sugar_in_your_tea@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    3
    ·
    1 year ago

    I never said we should eliminate votes, just that comments are more useful than votes alone. If you down vote, leave or upvote a comment that explains why.

    The problem with votes on their own is brigading, as in people down vote stuff because it’s unpopular (at least to a very mobilized and motivated group). The vote itself doesn’t explain what’s wrong with the content, only that a lot of people clicked the button.

    So I’m in favor of requiring the user to either leave or upvote a comment for the down vote function to count.

    • wildginger@lemmy.myserv.one
      link
      fedilink
      English
      arrow-up
      5
      ·
      1 year ago

      I dont really agree that you need to explain what was wrong with the content. And brigading happens with comments too, arguably to a worse end.

      Explaining every vote you make, or requiring a yes vote before you allow a no vote, also defeats the purpose of voting at all. I get it can feel bad to have a comment downvoted but… Like… Its really not a big deal. Especially in a system that has even less weight on karma. Sometimes the group didnt agree with you, and thats a normal part of community interactions. You shrug and move on.

      • sugar_in_your_tea@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 year ago

        feel bad

        That’s not the point at all though. The point is that it hides good content that a motivated group wants to silence. We had precisely this problem earlier in Lemmy’s history where posts critical of China were heavily down voted, not because of quality, but because the group didn’t like the message.

        Requiring a comment gives context to the negative reaction. It’s not a silver bullet, but it should increase the barrier to hiding content, hopefully enough that good, controversial content stays visible.

        I’m actually working on a Lemmy alternative that uses a web of trust instead of votes to prioritize and moderate content. Reddit has shown the limitations of voting, and I’m more interested in interesting content than content the majority likes.

        • wildginger@lemmy.myserv.one
          link
          fedilink
          English
          arrow-up
          4
          ·
          edit-2
          1 year ago

          Comments do the same thing, by drowning the one opinion in a sea of alternate opinions, and is directly incentivizing only interaction via people with the time to type up a comment. You arent preventing brigades, you are reducing the number of users who arent capable of attempting to brigade at all.

          Especially since your version of brigading is literally how communities work. If the group doesnt agree with an opinion, even opinions you do agree with, the opinion is going to be drowned out. You cannot police “opinion quality.” Because such a subjective thing is good when it agrees with you and bad when it doesnt.

          Good luck and I look forward to seeing it, but to be frank it sounds like you want to build a personal group chat, not a social media site. And like any web of trust, it relies on the integrity of the central member. Which isnt a defense against brigading, just a defense against brigading that doesnt come from the central member or their points of trust.

          E: mind, not that theres anything wrong with crafting your own supported super chat. Just that its less social media, and more a hyper evolved chat among friends and friend-of-friends

          • sugar_in_your_tea@sh.itjust.works
            link
            fedilink
            English
            arrow-up
            1
            ·
            1 year ago

            Comments do the same thing

            Maybe at a very high level, but comments have the very obvious advantage that they provide something that moderators can block. Lemmy does have open voting logs, but I highly doubt any decent moderator would feel comfortable blocking people based purely on how they vote, and they’d only actually look if there was an obvious problem (e.g. maybe they need to consider blocking an entire instance).

            directly incentivizing only interaction via people with the time to type up a comment

            This only applies to negative interactions, you would always be able to upvote a post.

            I think there’s an argument for hiding the voting buttons inside of the comment thread so users can’t just drive-by vote without actually looking at the comments, much less the linked content, but that’s not what I’m arguing for.

            You cannot police opinion quality

            You’re absolutely right, but you can increase the effort needed to downvote something. A downvote tends to have more weight than an upvote, so it should require more effort as well (e.g. a post with 8 upvotes and 0 downvotes would probably be ranked higher than one with 20 upvotes and 12 downvotes).

            it sounds like you want to build a personal group chat, not a social media site

            No, I definitely want a social media site, I just want everything distributed, including moderation.

            Basically, I want something like BitTorrent, but for social media instead of files. That way there’s no central authority for pretty much anything, so moderation pretty much has to be opt-in (otherwise you’d pick a different client with different moderation). Ideally, you’d select a moderation team that would filter out bad stuff like CSAM, but not filter out high quality content that you simply disagree with. So you’d pick a diverse set of content moderators to trust, and content would only get filtered out if a certain number of them flagged it. You could use the tools to create an echo chamber for yourself, or you can use it to expose yourself to diverse, high quality content that may challenge your beliefs (my personal preference).

            That said, things tend to work differently in practice. At the very least, I’m not going to release it until I have a way for users to review the quality of the moderators they pick.