I’ve been using Lemmy for a couple years now, and unfortunately I’ve noticed a significant decline in the niche communities that were originally active. When I first joined I saw much more variety when browsing the All feed. But over time, the communities I liked have faded as shitposting and meme communities have come to dominate the platform.
I think this shift has changed the culture of Lemmy. There seems to be more of a herd mentality now, where people downvote reasonable opinions they disagree with. The discussions don’t feel as nuanced. Some people have even been attacked for innocuous comments that don’t align with the prevailing groupthink.
The niche communities that made Lemmy special are fading away, and the resulting monoculture makes me less inclined to participate. I want a platform that supports substantive discussions in my interests, not just memes and shitposting.
I don’t know what the solution is on a platform level, but a culture shift is needed if Lemmy wants to retain users like me who valued the diversity of opinions. I may have to move to a platform that allows better filtering and proportionality between niche interests and funny or stupid content. I want Lemmy to succeed, but right now I’m finding myself drawn back to Reddit because the niche communities there seem more active. I’ll keep checking in, but Lemmy needs to recapture its original spirit if I’m going to make it my main home.
Slashdot had a good idea for this wherein votes were combined with a small set of reasons for a positive or negative vote. I think the use of those reasons was optional there, but modern systems might do well to adopt mandatory reasons for votes. If the reasons for negative votes are limited to things like “off-topic”, “hostility”, and the like, it could cut down on inappropriate downvotes.
Of course some people might still select inappropriate reasons, but a small nudge like that can have a big impact on aggregate behavior.
Should still include “i dont like it” as a reason
It is a very (the most?) common reason for downvoting and if you force people to chose a reason but don’t include it, they will just lie and the whole exercise will be rendered pointless.
And you know even though it’s not your personal preference, I think there are situations where it’s really just helpful to know “a lot of people agree” or “a lot of people disagree”. Not everything is about having a long debate with many sides. Sometimes the most popular thing is the best thing and the least popular is that way for a reason. Or it can provide useful context to understand the comments. Like if I am posting to ask advice about how to fix something and several options are presented but one of them has 5x the upvotes, I am thinking that might be the best one.
And it can tell you about the community. Like if I go into a community and I see someone says something nasty/dangerous/stupid and it has a similar votecount to other comments, I would think “I guess that sort of thing is acceptable here”. Whereas if I see it has lots of downvotes I might think “this comment is not representative of the general community here”. Voting based on like/dislike allows the community to express approval/disapproval when things don’t meet the threshold of moderator action; especially in very permissive communities where mods do not wish to take a heavy hand.
Further more, agree/disagree votes cut down on identical “me too” type comments. They give people a way to show approval without needing to make a comment and sometimes that is appropriate.
How do you think that would improve the quality of discussions?
It would let users choose the option that best describes the reason for their voting. You then do a bit of trickery where you don’t tell the user those votes don’t count and ignore all votes with that reason. That would allow removing useless downvotes.
But this being FOSS it would have to be a public secret so the irresponsible downvoters wouldn’t know about it.
Voting reasons and weights for each reason could be made configurable on a per-server or per-community basis without showing it to end-users, but to have federated voting work correctly it couldn’t be completely secret.
Most of the people thoughtlessly downvoting things aren’t going to put in the effort. Coordinated inauthentic voting is a separate issue my proposal does not attempt to solve.
I remember this system. I had to apply to do it after my account was old enough, then they’d give me a little bit to rate at first. Then IIRC they gave me more to rate after it was clear I wasn’t abusing it.
They had a guideline page I had to read before I started to rate comments and I don’t think those attributes were optional. So, comments got a primary attribute associated with their rating.
I wasn’t able to rate comments that I saw as I browsed but rather it was a collective rating system where volunteers were served comments (with expandable context) to curb the tendency to downvote just because you disagree with something.
At the height of Slashdot the discussions on there were incredibly educational and thoughtful and that rating system worked very well.