Hey! Thanks to the whole Reddit mess, I’ve discovered the fediverse and its increidible wonders and I’m lovin’ it :D
I’ve seen another post about karma, and after reading the comments, I can see there is a strong opinion against it (which I do share). I’d love to hear your opinions, what other method/s would you guys implement? If any ofc
That real question is, what problem are we trying to solve? Then we can go from there.
In wondering about that myself. What is the problem?
Individual users having some sort of reputation is useful. I always thought it was handy on Reddit to be able to distinguish people I happened to disagree with from actual trolls. The latter always had pretty high negative karma scores, and it was good to know that there was no point in engaging with them.
You can check their post history? Karma doesn’t tell you anything, really. Mine went up tenfold one day just because I replied to what ended up as the top post in a top thread in a much bigger sub than those I normally post in. Some people spend all their time in big subs making short, smart remarks that get a lot of karma, others spend their time in enemy territory battling people they disagree with. Some toxic people have a lot of karma because they hang out in toxic subs.
The problem to be solved is how to order threads. Old skool bulletin boards just bump the most recently replied one to the top. Which works well on an old skool bulletin board as long as it isn’t too large, but very badly on a big site where a few big active threads can drown out all the others.
I don’t know what the solution is. But the numbers don’t mean anything without checking the context. Karma is useful for ordering threads/comments, and giving users a bit of dopamine when they get some attention. But there (probably) are better ways to do it.
I don’t even know that karma/upvotes are good for ordering threads or comments. It just encourages gamification, group think, and snark.
I’d say get rid of down votes, replace upvotes with emoji reacts, and sort based on reacts + replies, but that’s probably just encouraging gamification, group think, and snark, too.
Reddit, like other centralized social networks that are trying to monetize us, prioritizes time on site and generic “engagement”. Those are what generate the most money for the company.
They’re not what’s best for us as users.
Maybe what we need to do is allow users to quickly and easily hide comment chains - not just collapse them, but dismiss them entirely - and allow for user-scriptable and shareable sorting algorithms. We drop down votes entirely, because they’re just used passive-aggressively anyway, make blocking users as easy as possible, with temp blocks and notification silences at the ready, and then forget about user reputation points entirely, because they’re as meaningless as Dragonball Z power levels.
Good stuff, thanks.
The thing is, high karma on Reddit doesn’t mean someone has a history of thoughtful engagement. Just as often, if not more, it means someone whose well timed with zingers on popular posts.
And incentivising that kind of take-down behaviour actually creates toxic communities.
I agree with you that high karma doesn’t indicate anything besides popularity, but someone with negative karma is almost certainly either a troll or a political extremist of some sort.
This is why it’s useful at the account level. It’s also useful at the post level in order to build a sorting algorithm which raises the most engaging/important/interesting submissions to the top. Within a community it is important to help define what that community is - irrelevant and low effort content is suppressed and relevant/high-effort gets boosted. Moderators can enforce this by just removing and pinning too, but that’s almost always too unilateral, and the voting system is generally better because it’s expected that then you get a representation of how people in that community feel about it. It’s a good system.
I can imagine some tweaks to help improve how karma is implemented:
-
Use Bayesan Inference to produce a ‘shit/shinola score’ for contributors instead simple up/down vote totals
-
Experiment with different recency biases for the score; you can trust that people will change over time
-
Generally figure out what you’ll be using karma for and make sure you have a way to measure how well it’s working
I’ve googled Bayesan Interference, however I don’t understand what you meant by it. Could you elaborate please :)
Here is a good general explanation of Bayesian inference.
I think @jayrhacker@kbin.social is suggesting using such techniques to predict “troll” or “not troll” given the posting history/removed comments/etc. My personal thought is that whatever system replaces karma, it should be understandable to the typical user. I think its possible Bayesian inference could be used in developing the system, but the end system should be explainable without it.
Thanks for the link. To anyone that does’t know about Bayesian inference, do check it out!
Now I have an existencial crisis thanks to the video 😂 the funny part is that thats the same thing used to detect spam email…
-
Or you could have a system where trolls and bad people are simply banned in stead of needing users to figure it out themselves
How would you find those bad actors?
That seems way better that some token system that becomes a game.
Repetitive low-effort posts and comments were common on Reddit
That it true… if they don’t “earn” anything for low effort comments, then they will diminish
They very often get hundreds, if not thousands, of upvotes, though.
They’re common on the internet and real life in general.
This!
/s
Number go up, makes brain happy
Number go down, makes brain sad ;(
Monkey sees negative numbers, neuron activation, monkey leaves Lemmy
Monk tap where Apollo used to be on phone. Monk end back up on lemmy
There are few things Karma system helps with that come to mind.
For others:
- Reputation
- Activity
For you:
- That endorphin XP boost when you level up. Makes you more likely do engage after the first hit.
- Gives you an idea how your comment has been received by others.
Presumably there are other things as well, these just quickly came to me.
That is a good way to think about it. What is the need from the reader’s perspective and from the poster’s.
One would certainly read a post with low upvotes from a author with high reputation if you are interested in the specific magazine. I wonder if the reputation should not be topic bound and not just general. That would be useful from the reader’s perspective.
Some kind of implementation of what you said would solve Reddit’s problem of mods reposting and deleting content untill it “goes viral”.
The exciting thing about this space is that much of it is undefined. It is all about the protocols and the main features at the moment. The 2nd generation tools will be born out of what we discuss now and think about now.
How do you make sure a user is not trapped in his special interest bubble and still gets to see content that has everyone excited? How will we make use of the underlying data, on both posts and users to suggest and aggregate content.
I think there will be more than one solution eventually, different flavours of aggregators running on the same underlying data.
So much possibility. And we control it. If you don’t like the way your lemmy instance or kbin aggregates, choose another site or build your own. The data is there.
-
The first problem is people tend to follow the hive mind. If it’s downvoted, they will also downvote and vice versa. They also will believe a comment with lots of upvotes and won’t fact check.
-
The second problem is people will abuse a karma system. Bots can increase the reputation of an account to make them seem more trustworthy
-
The third problem is that the current system let’s you see who is downvoting/upvoting. People take it personally when they are disagreed with and will retaliate since they can see those users and stalk their account
I don’t think these problems warrants a change in the current system. The transparency is a crucial feature. Seeing the number of downvotes serves as a great red flag to warn readers that a comment might not be true even if it has a larger number of upvotes.
This does take away the anonymous part of your social media voting experience, but the ability to manipulate the platform is greatly decreased. People that get riled up about disagreement will need to chill and you will need to block those individuals that can’t.
I think this will allow the development of a more mature community by taking away some of the anonymity
The third problem is that the current system let’s you see who is downvoting/upvoting. People take it personally when they are disagreed with and will retaliate since they can see those users and stalk their account
I actually really like this. I’ve been downvoted a bunch, my kbin karma sits at negative, but it’s kinda neat to see that I haven’t been downvoted by complete assholes (based on their history) – makes me appreciate that we might just have different view about a thing (or I’ve acted like an asshole to no surprise). Nonverbal communication can be a powerful thing.
Do I think it’s feasible to leave as it is if this whole thing explodes in popularity in a new magnitude while Reddit sinks? No I don’t think so.
You sound like a normal person who doesn’t take shit personally – some people really, really do take negative feedback on social media the way that you might someone keying your car, and I worry about the repercussions of downvoting the 'wrong" person who might seek reprisal. An anonymous downvote button feels like an “oh, fuck off” button, a public one feels like “fuck YOU for real” to me.
This is indeed a new debate that I’d like to have 😂
Well said there’s probably more of a barrier without the anonymity which in itself can be also unhealthy for the whole system (less user engagement). Though now thinking about it maybe that’s not a bad thing. Less bandwagoning on downvotes etc.
But truely angering some unhinged lunatic by downvoting who will then dox you, harass you and possibly kill you doesn’t sound great. But that’s a danger you’re susceptible to just by existing on internet.
Oh yes, I once had someone tell me they wanted my family to die in a fire and suffer while doing so because I didn’t think Guardians of the Galaxy was the sort of comic that would make a good movie. I was wrong, but you know, I think I was the only person who ought to suffer for that mistake.
I do tend to think that this is one of those things that the Fediverse, by not demanding everything be monolithic, is uniquely capable of dealing with because instances could provide per-community options (public up and down, private up and down, maybe an actual “fuck you” button) and people could try and figure out what works best for their unique forum.
How about, agree/disagree and informative/troll
It’s a definitely an area to watch but I’m a huge believer that transparency makes a community better regardless of size. If you being brigaded or abused it’s visible to everyone and you can block those accounts if you wanted
The ultimate hope is that social media evolves for the better
On the one hand yes, but also, this makes it much less incentivizing to downvote instances of abuse, discrimination, far right extremism. A lot of those people are not okay mentally and hiighly committed to harassing anyone who disagrees with them, I constantly hear stories of a single disagreement leading to years of harassment on hundreds of alts.
I also believe in transparency makes social media better, under certain circumstances.
In real life, talking to people face to face, there’s a huge amount of limiting factors to what you can say and do, and that’s a good thing.
Somehow using the same principles online can be beneficial. The flip side is, online also exposes everyone to everyone else. The reach, both ways, is enormous compared to what’s physical possible IRL. On top of that, IRL, moments are transitionary. Nothing stays forever. Online, everything you do is tracked so there’s no possibility of privacy ultimately. IRL, everything is ultimately private because the reach is too limited.
It’s a complicated issue, all in all.
-
Not a problem at all. I understand that we are ego-driven, but then again, the fediverse is a new working paradigm. We are here because we want to. Genuinely curious what you guys thought!
We want to discuss topics. This is a place to do that.
Simple need, simple solution.
You don’t need an extra incentive to make people talk about things if people talking about things is the thing you want. You don’t want to incentivize people who don’t want to talk about things to be active somewhere you want people to talk about things because then those people will start doing the thing your’e incentivizing them for instead of talk about things.
I personally only want people who want to talk about things here, and don’t want people who don’t want to talk about things.
I am not sure I followed, sorry
Basically they only want autistic levels of Internet “debate” and don’t want people having low effort fun.
Exactly this. You want to incentivize discussion, not the dopamine rush casino/arcade that just leads to low effort, low quality posts. If people want to be here for discussion, then they will either lurk and consume, or participate earnestly. Don’t put systems in place that reward the opposite.
Yeah, the question strikes me as, “Reddit has this thing. A lot of people don’t like that thing, but how could we still have it without people not liking it?”
I think we’re good as is.
Good point, take my:
handshake, pat on the back, slightly too long hug point thingy.
I’m against any kind of global user ranking.
It makes sense to rank content, but ranking users just begs abuse of the system. There’s always those that will try to farm the system resulting in lower quality content. It’s also an attack vector for bots.
I don’t miss the “karma” aspect one bit here. Rate my post quality, not me. On the other hand, tools for ranking users privately could be helpful. In other words a personal ranking for your eyes only would be fine.
I agree. I personally found the system was far too addictive, in the Cookie Clicker kind of way of “bigger number = happy”. I sometimes find myself missing it almost, only to remember that it’s worthless.
It also means I can more freely share my actual opinions, without that reflecting on some sort of global score if people generally dislike said opinion.
Yes, look up “Facebook demetricator”. It was brilliant. I’m glad I never used Facebook.
i do like the RES feature of personal counts though
if someone on res had a [+10] next to their name, i’ll know i personally respect their opinions, even if i don’t remember their name. similarly, if they have a negative number, i’ll know not to engage as they’re probably a troll
What we have right now in Lemmy strikes the current balance IMO. Individual comments are upvoted/downvoted. But no cumulative score.
which is the right thing, judge the opinion not the person
There is that aspect of karma of “if you’ve got negative karma, you’re probably intolerable” but I’m not sure how much that helps in practice vs just banning people. Karma can also filter out fresh accounts for high spam communities, ofc, that doesn’t work perfectly either…
Karma farming has always been one of the worst aspects of the other place. Repost bots will sustain them long after the humans are all gone.
Throwaways are still an issue with banning.
Some kind of participation based scoring would just bring us back to farming and alienates lurkers.
Account age is unreliable.
Hmm… I hate leaving the burden on mods but karma has too many negatives.
It is doable for the mods when the community has 100 members, after that, it gets hard for the poor mods ;(
“Karma Farming” is only an issue to people who place excessive value on karma. For the rest of us, people posting content that people like is not a problem, it’s in fact the goal of the entire website.
“I’ve already seen this!! Why is it being reposted!” is such a selfish perspective, not everyone stalks these websites and every time it’s posted it reaches a new audience, why does it upset you that people like seeing these things?
This wouldn’t work in the fediverse anyway, as it’d even easier to fake your user karma here (on an own instance).
Person could still be banned if cumulative score is too low - it just has to be hidden. Second problem can be solved in the way forums did it in the past: user can be marked as newbie if his cumulative score is low (i.e. < 100 or something). These problems do not require visible karma system to be fixed
Agreed 105%
I agree 90%, downvotes shouldn’t have that much weight. That said, comments which are abusive or hateful probably should have long term consequences for the user, even if they are themselves not worthy of a ban. Maybe reputation can be a “strike” for number of reported comments.
To be clear, here I’m thinking of “dogwhistle” comments which individually are plausibly fine, but in aggregate indicate this person is up to no good.
On the other hand, kbin has a cumulative score, but currently implements it badly wrong. Your cumulative ‘reputation’ is calculated as “boosts - downvotes”. So if you post a thread that gets 100 upvotes, 9 downvotes, 80 comments and 5 boosts, you are rewarded with ‘-4 reputation’. Nobody really uses boost, so it is very easy to rack up negative reputation.
Thankfully, I don’t think ‘reputation’ actually does anything, but it is still kind of annoying to be ‘punished’ for posting.
I would almost say a better system would obscure usernames completely. Only show the comment text, and allow voting accordingly.
No, we need people to have some accountability or everyone’s just gonna be intolerable.
Federation already makes that completely impossible.
I don’t agree with the lack of usernames of course. There’s no community when there is no way to associate posts with individuals.
It’s not impossible. Each user is still tied to an instance, they still have usernames, etc.
But an individual can be any user on any instance. Even one of their own creation.
Accountability is important, even if we are just using usernames and avatars
The problem is Lemmy already can’t allow that. Every user is Multiple Man. If you ban or block me on one instance I can just come back from another instance. What’s more, I can just keep creating more and more instances to evade blocking or banning infinitely.
My point is simply that votes on comments should reflect merit on the actual comment, not because you recognize the posters username and dont like them so you downvote them regardless of what they say.
basically 4chan but with extra steps 🤔
How will I upvote people for having a name like rimjobsteve or a username related to the context ala r/usernamechecksout! /J
Why, by visiting the new c/usernamechecksout, of course! :)
No system. The goal isn’t Reddit 2, it’s a federated link aggregator.
deleted by creator
Upvotes/downvotes are still a useful engagement metric, for instance what should appear in user feeds. Converting that engagement into long term karma encourages reposts and bad actors though so throw it out the window.
Honesty, I don’t think I really like upvotes and downvotes at all. My favorite system is Discourse where the only sort option is old -> new and you can provide reactions (heart, thumbs up, etc…) that don’t change the sort at all. This lets you follow the discussion as it happened & gauge engagement yourself.
Best would be to give it some reddit gold… er… somehow.
I love the poor crop of lemmy.world. shit just works yo.
This could work 😂 I’m giving you the wholesome award! 🦭
fax
Web of trust. The biggest thing missing from most attempts to build social networks so far. A few sites did very weak versions, like Slashdot/s friend/foe/fan/freak rating system.
Let me subscribe, upvote, downvote, filter, etc specific content. Let me trust (or negative-trust) other users (think of it like “friend” or “block”, in simple terms)
Then, and this is the key… let me apply filters based on the sub/up/down/filter/etc actions of the people I trust, and the people they trust, etc, with diminishing returns as it gets farther away and based on how much people trust each other.
Finally, when I see problematic content, let me see the chain of trust that exposed me to it. If I trust you and you trust a Nazi, I may or may not spend time trying to convince you to un-trust that person, but if you fail or refuse then I can un-trust you to get Nazi(s) out of my feed.
It’s a novel idea, I can certainly see the nice implications of it, but it also seems incredibly excessive. Would you really going around flagging every user you see on a trust system? Or even enough users for the system to be moderately effective? And then expect many other users to do the same?
I honestly don’t think I’d use it, blocking people is enough for me.
I maintained a substantial set of tags for problematic users of all types so I knew to avoid commenting or engaging with them. In an given community it’s common for there to be a tiny percentage of prolific posters who are a real problem and tags with res is how I managed that. It absolutely can work and arguably twitter’s block and mute functions do a similar thing as thevpy reduce a user’s presentation considerably.
I think a thing like this will even be required in the near future. Because we can no longer trust that we are talking to real humans, therefore some trust system (including physical interaction perhaps?) will be needed to not get social networks drown in AI dystopia. I have bookmarked this comment which describes the krass scenario we might find ourselves in soon: https://lemmy.ml/comment/878882
@dedale@kbin.socialRe “physical interaction”… Are you old and/or geeky enough to know about https://en.wikipedia.org/wiki/Key_signing_party ?
I found very interesting the concept of chain of trust :) What is the friend/foe/fan/freak?
Slashdot has a friend list and foe list. Fan and freak are the lists of people who name you as friend and foe respectively.
Thanks for the reading! That is very interesting concept that I’d love to see implemented here 👌🏻
Posts should just be upvoted and downvoted with no credit given to the person who posted. Same goes for comments. In my opinion, upvoting and downvoting should just help the user find the most relevant information. Content that people upvote is the most seen. Content that people downvote is the least seen. Posters and commenters stay on an equal footing with no points system.
Maybe we could still have karma, but display it as a ratio of good:bad karma or something? Active user and most of your interactions get upvoted, green dot. New user or not active for a while? Gray dot. Established user and all your content gets downvoted all the time, red dot.
Get banned from 50+ subreddits? Your color dot gets changed to a picture of u/spez.
best idea I have heard thus far
It provides other users with an at a glance idea of your reputation, without chasing a “high score”. Could always rank users based on up/down votes, as I said, but limit the range so that as long as you’ve been active for a few months and aren’t a douchebag, your score will be maxed out.
We should keep it as is. Having an account score just amplifies a big issue with sm. The content should be in focus, not the people posting. A relevant comment should be hightened because it itself is good. In the same way we shouldn’t judge something because the user has a low karma, but because the content is bad.
The idea behind something keeping a score on a profile is good, but it doesn’t work as intended in practice. People will farm in whatever way they need to get a moral highground. Not having such a scoring system will be a good way to reduce the incentive to copy/paste content from others.
You said this far better than I could. If there’s no supply, the addicts stay away.
A relevant and good comment, even and especially if it opposes the opinion of the majority. Giving downvotes to signal disagreement, when posts are sorted by karma and very low karma posts are even hidden, leads to circle jerking and immediately kills every healthy debate and controversy in the bud. If I have a dissenting opinion, I want to argue, not be muzzled.
What about the same system, but it shows both upvotes and downvotes?
I’d prefer that. 2600 up and 2500 down is really different than 105 up and 5 down
Tbf you can probably tell the actual numbers by looking at the % reddit shows in the corner, but that’s not very intuitive
You can do that for Reddit posts but can you also see it for comments? It wasn’t shown in my client app but perhaps it’s visible elsewhere.
Yeah imo the real problem with reddit was that
A: they started fudging the votes so they didn’t really matter and they could shadowban accounts from even being able to upovte/downvote
B: stupid fucking awards could keep posts at the top even if they had like -2000
c: fascists were gaming the system with bots anyway to push their content.
I don’t think you can have anything in the same spirit that isn’t toxic and doesn’t encourage brigading by minority groups who want to cancel opinions they don’t like. The whole concept is simply glorified ad hominem.
Absolutely. The real reason accounts accrue karma on reddit is to keep you engaged. People get addicted to big numbers. It’s just toxic. Upvote and downvote posts and comments but don’t keep a running tally on people’s accounts.
Do you guys think there could be a different system that could work?
Maybe something akin to Reddit’s award system as a way to thank people for being positive contributors? I think that could also be gamed but a way to say thank you rather than a status symbol meaningless number I think is a net positive.
Not that I can think of.
Subs should be able to force sort by controversial for comments and/or posts.
Any damn fool can come up with comments that are universally approved of, or universally hated. They aren’t interesting.
The phrase ‘trivially true’ applies - “This crime was a bad thing, and the people responsible shouldn’t have done it! I am very angry at them!” may be emotionally satisfying to say or to cheer on, but it doesn’t add a damn thing to the conversation, any more than “hur hur suck it libruls” does.
There isn’t a term for the inverse of ragebait, but there needs to be. All the le reddit moments - the tedious meme-chains, forced in-jokes, etc.
For subs where you want interesting discussion, you want to sort both to the bottom. It’s the posts that divide opinions that are worth talking about, almost by definition. If a post has a thousand votes but the total is close to zero, well hey, that’s probably worth seeing and engaging wth.
Let people vote with their heart, use upvotes/downvotes however the fuck they want to instead of constantly nagging and whining about it - and then use that to detect and de-prioritise mediocrity.
It wouldn’t be appropriate for all subs, but for some places, I think it’d be a huge improvement.
Top 3 most upvoted comments always being unfunny puns was getting insufferable on Reddit. Everyone was trying to be a fucking comedian - that’s what was popular and got upvotes.
The early Reddit you could have long, interesting arguments with people and you’d both be getting upvoted because you’re both making interesting points.
It honestly feels like my brain is waking up from a digital coma since coming to Lemmy from Reddit. My own personality and opinions don’t feel pointlessly supressed and sanitised.
I really like this solution. Instead of making things more complicated for users or trying to control their input, observe their natural behavior and then respond to it.
Judging by the controversial comments on Reddit, I don’t know if I want to engage with 50/50 up/downvotes for any significant amount of time. I think a 60/40 ratio might be a bit more palatable while still keeping it engaging. I’m not convinced an algorithm like this is the best course of action though.
“tell the truth” always brings the best reaction
Unfortunately, anything you replace karma with will have the same problems that karma has. Any indicator of comment or user quality will be readily gamed by anyone with any skills whatsoever in automation.
Yes and no. Toxicity will always be around indeed… but we can definitely lessen its effect.
thing is… in the end, karma doesn’t serve as that anyway (indicator of quality). It’s so easy to karma farm by (re)posting content (sometimes even stolen) in multiple communities.
In NSFW communities, at least on Reddit, I see SO MANY posts that doesn’t fit the community they were posted in, but being upvoted anyway because… well… it’s nudity
Karma may not be an actual indicator of quality, but it is often used as such. That’s the reason why all the bots exist in the first place and they are oddly enough* also the reason it’s not a good indicator.
People look at top, People like to filter out the bottom.
Look at the alternatives. Page views? They’d be instantly botted. Engagement? Instantly botted. There’s literally not any way to indicate that the crowd likes something or that something is of interest that can’t be replicated in a hot second. Karma is the closest thing we have to a sorting filter that content creators are doing the right thing or an indicator to content consumers that something might stand out from the crowd.
I’m sitting here farming /r/interestingasfuck trying to make the /c/interestingasfuck viable and 2/3 of the highest ranked crap is garbage, The thing is, even 1/3 of it being real saves me from having to sort through thousands of page of crap to find decent stuff.
edit* missed a word
I much prefer how Lemmy approaches this; upvote and downvote count per comment, no tally of total points.
Way less people trying to Karma farm then and repost content for fake internet points that don’t mean anything.
Absolutely nothing. Reducing people to a number and ranking their value based on that is inherently wrong.
Keep it simple, the current Lemmy system works fine. Spambots and particularly disruptive people should just be banned anyways, a gamification system would not solve any issue on that front.
While I still would like to see an alternative to Karma that’s less problematic, I agree with the idea that gamification will not solve issues. If anything, it creates a “KPI/score” people want to desperately meet for the wrong reason.
But what about the dopamine hit from upvotes though…
Let’s keep the upvotes to the post/comment only, do not show the overall of a user and don’t take it into account in any algorithmic decision. Let community managers see the ‘karma’ of the user in their respective community maybe, but beyond that it’s a feature that only had negative implications on Reddit
I asked chatgpt, just to see what would filter out…
I would like some advice on designing a content-sorting and content-filtering mechanisms for discussion forums, in order to avoid or mitigates some of the problems that current systems are prone to.
One of the social problems with discussion forums results from the concept of upvoting or downvoting content.
It can be argued that voting content up or down serves to sort content by quality, allowing high-quality content to be seen and engaged with, while discouraging low-quality submissions.
However, in practice, Goodhart’s law ends up applying: when a measure becomes a goal, it stops being a useful measure.
By using popularity as a proxy metric to determine quality, and by rewarding popularity with positive attention, this ends up selecting for content with superficial emotional appeal - ragebait, memes, facile/obvious comments pandering to common sentiment, puns, etc. - and not ‘useful’ content that is thoughtful, incisive, analytical, or important-yet-unpalatable. Ironically, content of this nature is also low-quality in a different way.
Worse, this ends up training users to produce content of this nature at the expense of thoughtful, interesting, incisive discussion, reducing both the production of quality content, and the quality of the collaborative sorting that users perfom by way of voting.
A cumulative ‘karma’ score for users, being a sum total of their upvotes and downvotes across all of their submissions, while ostensibly encouraging engagement and admirable behaviour… can again end up being gamified, and lead to users making large numbers of popular but low-effort submissions in order to maximise their score.
This can also produce ‘filter bubbles’ or ‘hive minds’: subcultures or communities that reject and discourage disagreement or criticism, and end up with a positive feedback loop leading to increasing orthodoxy with little grounding in reality. In some cases, the communities involved can become radicalised or toxic as a result, as more-moderate voices are suppressed from the discussion, and increasingly-extreme views become required in order to gain attention. Of course, it’s reasonable and useful to let groups based around common ideals preserve an individual identity separate from the mainstream of discussion, but perhaps a way to regulate excessive self-selection would be useful.
Doing away with any kind of quality-sorting mechanic altogether is not a very useful alternative; a forum where spam, antisocial, or irrelevant submissions are given equal weight and visibility as interesting or thought-provoking posts tends to discourage users from participating or even reading.
As a further complication, users do not generally follow instructions or policy set out for voting on posts, and will tend to use them to indicate approval/agreement regardless.
Human moderation is an important part of any discussion forum and essential to the guidance ans shaping of any community, but the workload does not scale well; it should be considered a limited and intensive resource.
It would seem there are a few challenges to overcome:
1: Better-define the concept of ‘quality’ for discussion-forum posts beyond immediate, superficial emotional appeal - what is it that makes discussions feel satisfying, and drives deeper engagement? 2: Find a way to programatically ascribe a metric based on this concept, via the engagement of other users, in any form including upvotes, downvotes, tagging, commenting - and statistical or derived data such as the number, depth and overall score of replies to the post, and perhaps even analysis of the text of replies. 3: Ensure that any such system is difficult for individual users to exploit or game; for instance a naive reply-count metric could be trivially exploited by a user posting a large number of replies to a single comment.
4: Limit the effect of self-selection feedback loops, to prevent radicalisation while still allowing common ideals and identity to flourish in their own spaces.One idea I’ve had would be to sort all content by ‘controversialness’ - that is to say, posts with a high number of total votes, but a low-magnitude score once downvotes are subtracted from upvotes. I believe this would tend to deprioritize low-effort posts such as ragebait or pandering to common sentiment, along with junk, spam and boring or irrelevant content, and thus bring ‘interesting’ content to the top. I believe it would be difficult to game for attention, or for other users to have undue effect on its visibility. However, I’m not sure whether controversialness would be a good metric to drive engagement, or what large-scale effects sorting on this metric might have on a community.
Can you offer any advice or commentary on the situation, or identify any important flaws in my analysis? Can you suggest any alternatives that might work better?
===================================================================================== ChatGPT You’ve outlined several important challenges and considerations when it comes to designing content-sorting and filtering mechanisms for discussion forums. The issues you’ve raised are indeed prevalent in many online communities, and finding effective solutions can be complex. I’ll provide some advice and commentary on your analysis, as well as suggest alternative approaches that might address the problems you’ve highlighted.
Better defining quality: You’re right in emphasizing the need to define quality beyond superficial emotional appeal. One way to approach this is by incorporating multiple dimensions of quality, such as insightful analysis, thought-provoking content, well-reasoned arguments, and valuable information. Encouraging constructive and meaningful discussions should be a primary goal.
Ascribing metrics based on user engagement: Rather than relying solely on upvotes and downvotes, consider implementing a more comprehensive metric that takes into account various aspects of user engagement. This could include factors such as the number of replies, the depth of discussion, the length of time spent reading the post, and even sentiment analysis of the comments. By incorporating a broader set of metrics, you can obtain a more nuanced understanding of user engagement and use it to evaluate content quality.
Avoiding gaming and exploitation: To mitigate the risk of individuals exploiting the system, it’s crucial to design mechanisms that discourage or penalize excessive or low-quality contributions. For example, you could implement rate-limiting measures that prevent users from making an excessive number of submissions within a certain time frame. Additionally, considering the quality of the engagement (e.g., the relevance, thoughtfulness, and contribution to the discussion) rather than just the quantity (e.g., sheer number of replies) can help minimize gaming attempts.
Addressing self-selection feedback loops: To prevent echo chambers and radicalization, it’s important to strike a balance between allowing communities with common ideals to flourish and promoting diverse perspectives. You could introduce mechanisms that encourage exposure to alternative viewpoints, such as algorithms that prioritize showing posts to users with different preferences or even implementing mandatory or randomized exposure to different topics or communities. Additionally, fostering a culture of respectful and open-minded discussion can help create an environment where diverse opinions are valued.
Regarding your idea of sorting content by controversialness, it could be an interesting approach to promoting engaging discussions. However, it’s important to carefully consider the potential effects of such sorting on the community. Controversial topics and discussions can be divisive, and if not managed well, they may lead to increased conflict and hostility. Implementing safeguards, such as active moderation and community guidelines, would be essential in maintaining a healthy and constructive environment.
In conclusion, tackling the challenges of content sorting and filtering in discussion forums requires a multifaceted approach. By defining quality beyond superficial appeal, incorporating comprehensive engagement metrics, preventing gaming and exploitation, and addressing self-selection feedback loops, you can design a system that promotes meaningful discussions while minimizing the negative effects associated with current mechanisms. It’s crucial to iterate and refine these systems based on user feedback and community dynamics to ensure the best possible experience for forum participants.
This is a great example of how aggressively useless chatgpt is at this kind of thing, the response is literally just the prompt rephrased and packaged like an answer lol
Thanks for the great comment. This multifaceted approach sounds intriguing, but may produce a black box effect, wherein users don’t even understand the system or whether behaviors they engage in are contributory or detrimental to the community as a whole. How technically difficult would such a system be to implement and furthermore to properly convey meaning to users about their engagement with it? How could it be implemented to provide the psychologically rewarding effects, like karma does, to encourage further engagement without gamefying the system to the point of reaching the pitfalls you outlined of the current system?
It’s a shame, but any sort of number-based system will most likely end up with the same problems as karma. Not having the numbers add up is a good start though, since upvotes and downvotes are only really useful as ‘in-the-moment’ indicators of good vs bad content.
Let’s keep it how it is, so that we don’t have another social credits system that doubles as a dopamine factory.
Another element is that total upvotes don’t need to be shown on your profile. It can be on the comments/posts alone.
I like that.
