in different social networks I often see a table of fediverse alternatives to centralized social networks like twitter = mastodon and so on, but I noticed that the alternative to reddit is piefed and not lemmy, can someone explain what kind of fediverse project this is, and is it different from lemmy?🤔
If someone has nine downvotes to one upvote, they’re a troll? Or even just ninety downvotes to ten upvotes? Given how easy it is to make multiple accounts, an enterprising user could easily sway that figure for new users or people who don’t post often.
Even if you argue the system isn’t that bad, it’s built on a factor that’s easily manipulated, and therefore can be manipulated. It shouldn’t exist in the first place.
It’s based on total lifetime votes, isn’t it? If so, then I’d say yeah. A 9:1 downvote to upvote ratio is insane. Back when karma was public I never saw a legitimate user with karma that bad, not even on the most argumentative and controversial posters (hell, finding a user with sub-60% upvotes was crazy enough). Anyone who averages a ratio that low across all contributions is almost certainly not someone you want around.
I could see a new user getting hit with the tag if their first post is extra spicy, but it would go away after just a few regular comments and I’d hope the system doesn’t kick in until a certain post threshold anyway.
Re: vote manipulation, it should be easy to spot bad actors given that Fediverse votes are public. Admins can see voters directly and users can use a site like lemvotes.org to check if votes are legit. A swarm of bots hitting a single user with nine times their total votes up until that point is the kind of thing that would be obvious to instance admins and grounds for bans/defederation.
That’s what banning and blocking are for, those are features already up to the users, admins and moderators. So what’s the point of the flag in the first place?
In order to notice a flurry of users downvoting the same person consistently, someone would have to be investigating. I’m a mod and I’m not sure how to begin to tell if they happened to be sockpuppets. This means tools would have to be built to monitor where votes are coming from and flag users who vote together (if they don’t already exist.) That increases mod duties to patrol a feature because it can be gamed, when, again, there are already tools for users to decide for themselves who they interact with.