> Curiously, authors that receive no feedback, are more likely to leave the community entirely.<p>Ding ding ding.<p>I think, aside from the issues with downvoting, upvoting by nature slowly creates hivemind, circle jerk communities that quickly start to lack any real critical thinking.<p>Go check out just about any subreddit and it becomes very apparent what illogical silly ideas dominate. If you are logical but uncontroversial you'll get ignored which is just as bad as the controversial opinions which get downvoted to hell. You start to wonder if the confirmation bias we are all building towards is more dangerous than just being uninformed.<p>There's lots of research on voting behavior in political science, I'm slightly surprised more hasn't been down with it.
I work for a gaming website that lets people vote on the games and comment. It's mostly a younger audience (under 18) but there's still a good amount of all ages.<p>Anyway, we released a game one day and some of the first comments were really negative, even though the game was objectively a pretty darn good game. Knowing the game was good I was awaiting the nice comments to show up, but they didn't. It was just a flurry of negativity.<p>I talked to some other employees and they agreed the game was good and something felt off. We all jumped into the comments, gave good reviews of the game and started discussions with all the negative reviewers. Then the positive reviews started showing up, and the games rating did a complete turn around to overwhelmingly positive.<p>I feel like if we hadn't intervened, the game would have continued to have negative reviews and end up with a bad rating. There's a kind of hive mind mentality going on, and the first public comments have some sort of psychological effect on the rest of the players.
From the article: "Cheng and co began by compiling a dataset of the comments associated with news articles on four online communities: CNN.com, a general news site; Breitbart.com, a political new site; IGN.com, a computer gaming news site; and Allkpop.com, a Korean entertainment site. The data includes 1.2 million threads with 42 million comments and 114 million votes from 1.8 million different users." I would respectfully suggest that the comments from news readers on news sites may not be a dataset that generalizes to discussion and news aggregation sites like Hacker News, especially because Hacker News starts us out with a set of user guidelines[1] particularly geared to encouraging sustained user participation in thoughtful discussion. There is also a welcome message for new users here[2] (I forget how it is implemented, as I am such an old user that I think I joined HN before it was implemented) that packages the rules in a slightly different format.<p>In other words, most of the time when I am downvoted here, what I ask myself is, "How could I have rewritten that comment to make it more clear or more persuasive," or sometimes, "Was that really a constructive addition to the discussion here?" I'm not generally spoiling for a fight or crushed in my self-esteem after being downvoted here. As far as I know, that happens to almost everyone here sooner or later. It sure has happened to me over the years<p>On my part, I downvote some comments, for reasons suggested in the welcome message, but I try to make sure that my upvotes predominate over my downvotes. And I don't keep an enemies list. If someone makes a thoughtful comment, they get an upvote, period, even if they disagreed strenuously with me just hours before. The idea here is to help good comments gain enough attention to promote thoughtful discussion in all good senses of the word "thoughtful." Maybe Hacker News is just plain different from most news discussion websites.<p>[1] <a href="https://news.ycombinator.com/newsguidelines.html" rel="nofollow">https://news.ycombinator.com/newsguidelines.html</a><p>[2] <a href="https://news.ycombinator.com/newswelcome.html" rel="nofollow">https://news.ycombinator.com/newswelcome.html</a>
I'm a math teacher and I saw a weird hivemind effect recently. A student was showing a proof, but I felt one of the steps seemed wrong (I wasn't sure). So I asked everyone else if it was right. They responded with a chorus of "yes" "of course" and a little laughing. Eventually the proof ended up showing something that was not true. So I went back to that step and again asked for confirmation. Again most of them agreed. I wrote the formula to be sure: log(A+B) = log(A) * log(B). Eventually one student with a different text book had looked it up and found it's wrong after all. How had most of the others been so confident that it was right? Surely they hadn't actually learnt it wrong. They must have been making their decision on some other factor, perhaps the fact that all their vocal classmates agreed.<p>At least they weren't giving negative feedback to the one who made the original mistake though. So I guess he won't continue to make worse mistakes as this article suggests.
I think the reason this behaviour disagrees with traditional ideas about negative reinforcement and punishment is because those theories were built on hierarchical relationships (someone's in charge and doles out conditioning action). If the person receiving the punishment accepts the right of the person doling it out to do so, even if they resent it, they will accept its validity more readily. Also, if they feel unjustly punished, they will focus their ire on the authority rather than the community as a whole. This kind of thing can easily be seen in forums where there is a hard line between moderators and moderated.<p>When people feel attacked by their whole community (which is what a slew of negative votes feels like), it seems like they'd be far more likely to lash out at that community as a whole, and that means throwing out increasingly antisocial posts and attacking people with downvotes themselves.
I think the concept of being able to see your "points" at all is non-constructive for communication. Why should it matter to you if other people up or down vote your comments and submissions? So through a knee-jerk feedback mechanism you can best mold your future communication to fit with what will get you the most points?<p>If the purpose of commenting and submitting is not to amass the most "points" possible then they shouldn't be displayed.
<i>Not only do authors of negatively-evaluated content contribute more</i><p>So empty vessels really do make the most noise.<p><i>That points to an obvious strategy for improving the quality of comments on any social network site. Clearly, providing negative feedback to “bad” users does not appear to be a good way of preventing undesired behaviour.</i><p>I'm not seeing such a strategy from the paper's findings. It suggests several ways to limit bad behavior, but since positive feedback doesn't seem to improve quality of comments or posting frequency, I'm getting the strong impression that the overall tone of an online community is set early on in its life, and while it can easily degrade over time there is not a whole lot you can do to improve it.<p><i>So how can unwanted behaviour be stopped? “Given that users who receive no feedback post less frequently, a potentially effective strategy could be to ignore undesired behaviour and provide no feedback at all,” say Cheng and co.</i><p>This is pretty difficult to implement, though, since it only takes two people to start a flamewar. All I can think of is very proactive moderator pruning, which would a) become a full-time job and b) creates problems of its own as people start howling about censorship.
This was posted last year.<p>Previous discussion: <a href="https://news.ycombinator.com/item?id=7760857" rel="nofollow">https://news.ycombinator.com/item?id=7760857</a>
Although downvotes don't help provide feedback to the users, allowing them can provide an immense amount of feedback for the operators of the community.<p>You need a certain amount of positive feedback relative to negative feedback, and if you monitor that ratio, you can have a good pulse on the quality of your community. If you are not maintaining a high ratio of positive to negative feedback, you can take action to increase positive feedback in meaningful ways. You can also use provided negative feedback to tailor interactions. Someone downvotes a particular thread? Don't notify them of any future discussion on that thread if you otherwise would have, etc...<p>See: <a href="https://en.wikipedia.org/wiki/Critical_positivity_ratio" rel="nofollow">https://en.wikipedia.org/wiki/Critical_positivity_ratio</a> In my opinion this ratio applies equally as well in social networks and online communities.
Online arguments tend to be long, degenerate, and quickly form opposing camps that will war in other threads too. I think that's (informally) very well known. And a down-vote is a potentially significant indicator that an argument is taking place.<p>But I think the authors are a little naive if they think the site-owners are not aware (and in some cases, counting on) this.<p>For many sites, it seems to be the arguments and battles that drive activity. As if to the site-owner, a (moderate) ding-dong battle is "increased engagement, even from the people who disliked the content" rather than something they necessarily want to stamp out.
If only voting represented quality and usefulness. Unfortunately, it's most of the time emotional. Throw in conformity and political correctness, and you have a powerful tool serving both.
> “We find that negative feedback leads to significant behavioural changes that are detrimental to the community,”<p>I really hope the sites that have downvotes don't take this to heart. There are massive benefits that the authors completely overlooked.<p>It allows the community to rapidly bury garbage content, and because of this, the pool of potentially good content (with few upvotes) is much smaller and less daunting for the users to sift through, which allows <i>more</i> good content to rise to the top.
I've seen this in the "social network" of code reviews where I work. A "downvote", i.e. a negative comment, often has the opposite effect of the one intended by the reviewer. The code author will dig in their heels and over time contribute lower quality code. Alas, I don't have the luxury of simply ignoring lurking bugs.
This was fascinating. One of the interesting challenges with feedback that I have observed is that often separating feedback on the comment/post/idea vs feedback on the author is hard to distinguish.<p>I have this thought that instead of 'up' or 'down' a less directed 'agree or disagree' might be less painful on the poster.
I don't buy it. They just found a correlation that users with a downvoted comment happened to also make other bad comments (what a surprise.) They didn't actually downvote or upvote comments on randomly picked users and see what effect it had, which would have been way easier than the machine learning approach they used.
Voting is basically used as a form of censorship in almost every community on the Internet, including HN. If you go against the grain, you will be removed and ignored.<p>I guess it's a mirror of real life: people generally only want to be around people that think like they do.