92
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this post was submitted on 26 Aug 2025
92 points (100.0% liked)
Fediverse memes
1857 readers
87 users here now
Memes about the Fediverse.
Rules
General
- Be respectful
- Post on topic
- No bigotry or hate speech
Specific
- We are not YPTB. If you have a problem with the way an instance or community is run, then take it up over at !yepowertrippinbastards@lemmy.dbzer0.com.
- Addendum: Yes we know that you think ml/hexbear/grad are tankies and or .world are a bunch of liberals but it gets old quickly. Try and come up with new material.
Elsewhere in the Fediverse
Other relevant communities:
- !fediverse@lemmy.world
- !yepowertrippinbastards@lemmy.dbzer0.com
- !lemmydrama@lemmy.world
- !fediverselore@lemmy.ca
- !bestofthefediverse@lemmy.ca
- !fedigrow@lemmy.zip
founded 11 months ago
MODERATORS
Downvotes are useful to make bad content sink. Without them, the bad content has the exact same score as fresh new content, content that failed the Fluff Principle, etc. And you do want the bad content to sink; if you don't reduce its visibility, some clueless muppet is bound to interact with it, usually generating more bad content.
That's why I'm not sure if the best solution is to outright remove downvotes. It feels to me like throwing the baby out with the dirty water.
Instead I feel like splitting its role into 2+ buttons might alleviate the issue. Perhaps a simple "disagree" button, or a more complex Slashdot-like system, dunno. Either way, giving people way to say "I disagree!" without interfering on the main purpose of the button - sorting content.
This could also solve another issue with downvotes I don't see people mentioning often: you're often downvoted without knowing why.
At least in the default interface, the sardines comment would show +12 -62, so you know at least 11 people upvoted it.
If we’re using votes to rank content then downvotes are redundant because now you have to upvote „right” stuff and downvote „wrong” stuff. Assuming everyone is waging the same kind of information warfare then downvotes won’t anything… but we’re not. Those that downvote willy nilly just want to have more say in things than others who don’t have energy to religiously clean website from „wrong” content. You’re not responsible for safeguarding users from „wrong” content unless you’re reporting rule breaking one. If you don’t like what’s being said but it doesn’t break rules then reply and explain why is it wrong, let others upvote if they agree.
Tildes solved this already. They have regular upvotes and they have labels for offtopic/noise/malice. Being able to use labels is reserved to users with good standing and can be applied once only. Noise downranks things without removing them, malice is essentially same as reporting them. Notably, there is no label for „wrong”.
In practice they are redundant because most people vote based on opinion, so both become the same (agreement gauge). However ideally they aren't redundant; upvotes are to be given to things that stand out, and downvotes to things that detract from the discussion (noise, trolling, etc.)
Some see this as an abuse of the system, not as its normal usage. I'm not sure on the dividing line between both things, though.
The problem with that is Brandolini's Law: even if we ignore "intention" (whatever this means), it takes far more effort to address bullshit, assumptions, oversimplifications, "ur sayin dat cuz ur..." etc. than to come up with it. And if it takes too much effort, people won't do it.
As such, a system can't rely solely on replies to let users show each other "hey, this post/comment is bad".
You can rely on stricter moderation; but that comes with additional costs.
Incidentally my proposal to fix downvotes isn't too different in spirit from what Tildes do.
So, people want to up/downvote based on opinion, right? Let them do it. But give people other ways to quickly show some piece of content is bad, and why. Effectively splitting the downvote button.
I don't see how downvotes help filter content. It makes sense at first, but either people are sorting content by New, in which case votes do not matter, or they are sorting by Top and will get only the "good" content. Several instances already have downvotes disabled. I don't see any complaints from their users about "bad" content having the same scores as "good" content.
lemmynsfw had to disable downvotes because gay content posted in gay communities was being downvoted. It wasn't being downvoted for quality, but for not being what the majority of users wanted to see. That doesn't mean all users now have to see gay content they don't like because they can't downvote it. It's still easy to filter using the block feature. Again, I've never seen users there complaining about being unable to filter good from bad because they can't downvote.
I've seen posts and comments with -100 votes often get lots of interaction from people who can't stop themselves from arguing with a troll. Sometimes there's only 1 or 2 comments under a post so the score doesn't even change its visibility at all.
The way to say "I disagree!" is with the reply button! Votes don't prove who is right and who is wrong. I've never changed my opinion because of downvotes. Sometimes I even agree with a downvoted comment because I form my opinion based on arguments, not votes.
I also like seeing different opinions. Yours gave me a lot to think about! It'd be a shame if people didn't post their thoughts because they feared being downvoted for it.
Think quantitatively. Ideally "meh" content should still be easier to see than the bad one.
In their situation (as admins of an instance where downvotes were consistently misused), I agree with their decision. However I still think something needs to be done on a software level.
Note this is prone to selection bias.
If it wasn't downvoted, you probably would've seen way more interaction with it.
(Additionally I think people who argue with trolls should get 1d~3d bans. Just a "stop it, you baka!". Including myself. But that's an aside.)
I mentioned this in the other comment, but basically: if the reason you disagree is due to some issue in the content (e.g. it's an oversimplification, assumption, or plain bullshit), it takes more effort to address it in your reply than to generate that content with the issue. As such a quick-and-dirty way to voice "hey, something wrong with this" is necessary, even if some people abuse it.