This is not a new issue, but we’ve had reports from some communities that they are experiencing a lot of repeated downvotes from the same set of people.
This is how it typically plays out, using AI images as an example:
- A section of the lemmy user base really hates anything AI generated.
- Instead of blocking AI generated image communities, they down vote those posts every time they see them.
- The posts in those communities effectively have to overcome a “handicap” of down votes each time they are posted. This harms community growth and discoverability.
The admin team would like to know how our community would like us to handle this issue, since it isn’t clear to us what is the best approach, and we would like a consensus view.
Some option for consideration:
- Encourage/allow community mods to ban persistent down voters from their community (note that we currently have no specific rule in place for this, so it is currently allowed).
- Pros: prevents future down votes; essentially “unsubscribes” from the community on their behalf
- Cons: could potentially be abused by mods who want to eliminate all down voters and “game” the system
- Have a policy of ignoring the persistent down voters
- Pros: allows people to continue to express their dislike of [insert topic]-type posts
- Cons: means that communities on topics that are not of interest to (or are actively disliked by) the majority of users will continue to be penalized in the lemmy post feeds.
- Leave it up to the discretion of the individual community mods
- Pros: self-determination and community based approach (i.e. only applied when needed)
- Cons: potentially inconsistent approach to down voters across the instance
Feel free to come up with more options, but these are the three main alternatives I could come up with.
We are interested to get your thoughts on the topic so we can come up with a policy for the instance. Please leave your comments below on your preferred option and the reasons for your choice.
Edit: apparently community mods can’t currently see the voting breakdown in Lemmy, only instance admins can, so this adds further complexity to the issue.
A solution I thought of is that we could potentially create a bot that checks if a downvoter is subscribed to the community they are voting in. If they downvote more than 3 posts a day in such as community they get a 3 day ban, second time they get a 7 day ban, third time they get permanent ban.
Also if they downvote more than 70% of a community’s posts they get permabanned either way. Downvote happy subscribers are clearly trying to game the system, so they don’t get two chances to correct the behavior before a permanent one.
Could have the rules be customizable or not, basically it would check those two statuses and impose a ban if correct conditions aren’t met or the user goes overboard with downvoting. Users who would want to use it in their community could simply add the bot via photon or the API.
punishing people for downvoting AI slop
Now that would be quite the turn… Downvoting AI slop is the basic human decency I’d expect from Lemmy.
One person’s “slop” is another person’s favourite content, which is exactly the problem with “activist downvoting”. Those persistent down voters are effectively forcing their own personal feed preferences on everyone who uses Lemmy instead of simply blocking the communities they aren’t interested in. AFAIK it’s a problem with a lot of niche communities, not just with AI related communities.
effectively forcing their own personal feed preferences on everyone who uses Lemmy
But… that’s how voting literally works. Hence one of the smartest ideas I saw once for Lemmy (and for the Fediverse in general) was to substitute upvoting/downvoting for “voting on tags”, such as being able to tag a post as “fake news”, “inspiring” or “AI generated” and have people vote on those instead of on the psot / comment proper. Alas, I don’t know what ever happened to that proposal, and could never find it again to track it down.
But, see, that’s the thing and that’s why there’s an important difference between blocking and downvoting:
If I block something bad, like say fake news or fascism or AI, I block it only for me, it’s only protecting myself; but if I downvote it, I also help protect my peers. If we want to make community, that’s very important. And like any measure, it can be gamed, but so long as it’s the one option we have, we gotta use like that. I expect AI slop to be batch-downvoted; if I didn’t, I’d be back at Reddit.
This is a very paternalistic attitude imo. Your peers don’t need “protecting” from AI content any more than they need “protecting” from pornography, for example. Isn’t it preferable that they get to decide for themselves without you trying to sink those communities in the feed?
If I block something bad, like say fake news or fascism or AI, I block it only for me, it’s only protecting myself; but if I downvote it, I also help protect my peers. If we want to make community, that’s very important. And like any measure, it can be gamed, but so long as it’s the one option we have, we gotta use like that. I expect AI slop to be batch-downvoted; if I didn’t, I’d be back at Reddit.
Keep in mind that this is an AI friendly instance and those communities indeed are clearly marked and have the right to exist, your attempt to suppress them from others including their target audience would be viewed as hostile, and therefore it is fair and reasonable for their moderators and admins to ban you from the communities, and possibly even this site for the interference.
I understand you have feelings about AI as a concept, but try to understand how it looks to people who don’t, including those admins and moderators. Imagine if a person operated a community about Digital Audio and music on Digital formats, and someone who passionately hated Digital music as a concept and thought it was killing Analog downvoted their entire community as a group of Anti-Digital activists. Do you think their behavior would be seen as some kind of righteous activism, especially by the mods of those communities, or do you think it would be seen as an act of hostility to try and suppress a community that is clearly marked, and has the right to exist.
We can argue about the ethics of Corporate AI and AI in disinformation, however it is clear that communities like the stable diffusion art community and AI discussion communities are clearly marked and indeed have the right to exist. If you don’t like them you should block them and if your friends don’t like them either they should. Trying to suppress communities that have the right to exist and are not doing anything illegal or against policy because you dislike them isn’t something heroic, or viewed positively, it is considered an act of hostility against them, no different than the Vinyl guys brigading the Digital music communities in the analogy.
If you believe that these communities though violate your own instance’s policies, it would be suggested that you report them to your instance admins, either via the report system or sending a DM to them. Otherwise you should block them and let other people block them or ignore them as they choose.
Ok, but if that’s your opinion then you should be fine with the same happening to “traditional” art communities. We should be allowed to brigade them in return, preventing them from showing up in feeds until they die off and move elsewhere.
If you don’t like AI, move to an instance that isn’t friendly to it. If you hate something, you don’t belong on an instance where it is accepted.
I would agree with this, except on the merit of moving instances, community blocking exists for the purpose of curating your feed to remove content you dislike.
I know a lot of people do not take the time to get to know the instances of the communities they subscribe to but they probably should, since the values of the communities on an instance are greatly shared with the values of the instance itself. Instances aren’t just random servers either, they’re like larger meta communities with their own culture and value. And generally if you don’t jive well with the instance you probably should steer clear of the communities it homes, as they will mostly share its values. That’s why the instance sidebar is displayed beneath the community sidebar on all remote communities.
Also I would like to take the time to say that their comment blanket comparing AI tools (including/not excluding those developed as an open-source or community effort) with fascism and fake news, feels extremely disingenuous and like a bad faith comparison. I’m not saying they should get in trouble for it but it feels like a very bad faith attack on our community. This kind of reminds me of the time that AI banner generation was announced on [email protected] and a lot of anti-AI trolls crawled out of the woodwork to complain, or even call for action against the community. I don’t know I just think people need to know how to read the room and know the space a bit better. @[email protected] @[email protected] Any thoughts you’d like to add on this last part guys?
I think you pretty much summed it up, matey.
If only we had an option to disable downvotes from non-subscribers!
Another solution would be to set it up via a bot. I could make a bot that you invite to your community and it automatically bans anyone who downvotes more than some % of posts, and allow that to be configurable by the mod. Not quite sure I can get that info from the API, but if not I could try my hand at a plugin.
If only we had an option to disable downvotes from non-subscribers!
Now that would be great!
And yes, I agree it would be preferable to have something that mods can opt into and configure to their community’s needs, only if needed.
Have you considered turning off downvotes?
That is definitely something to consider, though eliminating downvotes could be problematic since it messes with the way feeds get sorted, and also doesn’t stop downvote trolls from attacking from instances where it is enabled. I’ve seen a lot of this type of abuse on blahaj.zone and beehaw and the vots still get federated on remote users, they just aren’t visible on the instances which disable the downvotes.
You can’t disable downvotes per community only. Only instance-wide
I think they meant Instance-wide, which is something to consider, @[email protected] decided to disable them on Lemmy.blahaj.zone because downvotes are often used aggressively against people who are openly trans or visibly queer, and also for similar reasons we’re having problems, people like to use them to suppress things they personally disagree with, that includes behaviors like mass downvoting of communities and and subjects, up to even using multiple accounts do downvote multiple times. It’s also sometimes used as a method to harass or punish users, by downvoting every one of their posts and comments.
Downvote brigaders suck, they never contribute anything to the discussion, their only goal is to try and punish people, it was one of the many shitty problems on Reddit that we were told to just deal with. While it is mostly just annoying to get downvoted, so is spam in your, and I don’t see spammers being defended or people arguing for their right to post spam or arguing what does and doesn’t constitute spam. Nor do I see mods arguing that banning spammers violates “personal freedom” so I say that just like we do for spammers we should take out the trash.
personally I think that when it comes to people who will downvote everything in a community they either dislike/hate the community or are trying to manipulate or suppress its position. I think in these cases mods should ban the individuals doing it temporarily or permanently. Vote manipulation of this nature can be annoying for larger communities but it can be detrimental for small communities that don’t have much of a userbase and will get suppressed in the feed.
I do see the potential for abuse in mods that may attempt to game the system but in cases where mods abuse it they could be formally reported for it to admins, or informally reported in [email protected].
It might be good to have specific instance wide policies against this type of malicious voting behavior as well since people who do this mess up the natural organic flow of the feeds, and also can end up hurting or suppressing communities by doing it.
So where’s the line defining malicious downvotes?
Someone who downvotes 60% of a community’s posts? 75%? 90%.
Youre setting yourself up as vote arbiter, telling users they can only vote the way you want. That’s just as problematic, maybe more so.
Yea, there are people who consistently downvote stuff, but isn’t that how votes work? Like another commenter, I too downvote a lot of AI garbage, because it’s garbage. Am I a malicious downvoter, no longer permitted my opinion?
The last few weeks I’ve consistently downvoted any political posts in communities where they clearly don’t belong (my opinion). Am I a malicious downvoter for that?
