Using the fediverse is still a better love story than xitter or its spinoff.
Comment on Bluesky experiments with dislikes and 'social proximity' to improve conversations
lvxferre@mander.xyz 2 days ago
Charitably, these tweaks sound like another way Bluesky is trying to give users more control over what they see on the platform, in the same way it does with things like notifications. Less charitably, you could read the “social neighborhood” concept as a way to entrench users in their “filter bubble” rather than address larger moderation issues.
Potentially both simultaneously.
I think that a few key factors would be:
- will the algo deciding who’s in your “social neighbourhood” be publicly available? Or is it a matter of “trust me = be gullible trash”?
- which will be the rules deciding it? Poor rules can backfire really bad, encouraging mob mentality instead.
- are they going to address the poor moderation of their own platform, regardless of the above?
Either way the prospect is good. Still a better love story than Twitterlight.
technocrit@lemmy.dbzer0.com 2 days ago
lvxferre@mander.xyz 2 days ago
I should’ve made the sarcasm more obvious; my bad. (“Still a better love story than Twilight” kind of implies something is still bad or unremarkable; it’s like saying “wow, this meal tastes better than shit”.)
Yes, the Fediverse is genuinely good. And better than both Bluesky and Twitter.
OpenStars@piefed.social 2 days ago
I like the way that PieFed implements this.
“Highly contentious users” i.e. those who are consistently heavily downvoted by “trusted instances” (I don’t know the actual thresholds but imagine someone who receives 10x more downvotes than upvotes - and e.g. hexbear.net can be federated with but not “trusted” so that downvote brigading can be eliminated, unless ofc they use their non-HB alts but while nothing is perfect, every ounce of protection does help:-) are labeled, but there is currently no way that I am aware of to actually remove their content. Still, it helps to see that automatically-applied label as you scroll down, so that you can skip past it or at least realize that a reply is going to fall on deaf ears. People’s reputations precede us irl so why not online as well, where it is so much easier to measure?
Individual content - posts and comments - that are highly contentious, according to user-defined thresholds, can be either automatically collapsed or even hidden. I personally disable both of these, but if someone wants to not see highly contentious content then this makes it happen for them. Similarly there are keyword filters - again nothing will ever be perfect but if you want to see less of e.g. Musk or Trump, then this is a method to help reduce the incoming flood of content related to such.
Communities have access to “community-specific” voting patterns. I know less about this aspect but generally the entire community or perhaps an individual post could be limited to community-specific rules, like a member can vote but a non-member drive-by commentor might be disallowed under certain conditions. Not every community should be this way and I hope most won’t enable these features, but they are necessary sometimes - e.g. a community for and by women needs to exclude all the “don’t you know that I am such a nice man"-splaining that will inevitably arise.
Anyway I love the hierarchy that distributes the work of moderation all the way from instance admins (for e.g. illegal content) through community mods (who have access to software to help them) and ultimately powers the end-users to control their own recipient of content, which they can change over time - e.g. rather than leave social media entirely they could enable some of the contentious user and/or keyword filter controls and thereby attain for themselves a break from the noise and hubub that the entire internet tends to prefer to throw at us all the time.
In contrast, whatever little moderation that Bluesky has is obviously insufficient - the problems of outright monotonization spam and high contentious users seems to have overwhelmed whatever capacity there was to handle such.
PieFed has really high me hope for the entire Fediverse.
lvxferre@mander.xyz 2 days ago
#3 would probably not work in a microblog service due to the structure, but the other two could be easily ported. I could see them popping up in Mastodon, for example; but Bluesky? Not really - it doesn’t want to empower users, it wants to herd them.
rozodru@pie.andmc.ca 2 days ago
your last point is key. Bluesky has notoriously horrible moderation and a CEO that refuses to acknowledge it and mocks users that do.
This reeks of Bluesky letting “the babies have their bottles” so to speak but I feel it’s a waste of time. Bluesky users already heavily depend on lists to build their own communities in a way. It’s almost to the point on there where if something you dislike shows up in your feed well that’s your fault and the communities as a whole for not almost shadow banning an account via the god knows how many block lists users have created. “disliking” a post isn’t going to do anything. it’s stupid.
Hell Bluesky users have been trying to get some transphobic journalist banned from the platform for well over a year now and the CEO of Bluesky has essentially said “lol, nope, you all suck. don’t like it? stop posting in protest.” When Bluesky users are offered an alternative with better moderation that THEY can control i.e. Mastodon they tut their tongues and roll their eyes saying “eww the linux of social media”.
so again, let the babies have their bottles. it’s not going to do anything because Bluesky has clearly said they don’t want to do anything.
OpenStars@piefed.social 2 days ago
Not true - it seems designed to increase advertising revenue for the CEO:-P. That’s… “something”, technically? 🤪