This article is interesting but they are probably overthinking it. I think it's 80% or more due to the "road rage" effect. That is, people who aren't face-to-face don't humanize the other they are interacting with. There's also the "alcohol" effect, which is to say that someone speaks recklessly when they aren't afraid of getting punched for it.
The freedom to downvote. A downvote should force the downvotee to give a reason why.
A reason seperates the signal from the noise. The original poster can dispute the downvote given; which Reddit and HN, slashdot are all prone to. Karma points only encouragesusers to vote negative without thought if the output is already negative. As the same with positive.
--
Edit: My current downvote is a good example. I'm at 0.
I have been downvoted. I do not know why. Has some brigade decided to target my post? Downvoted because of my username? Unhappy with the idea? Does someone a personal vendetta agaisn't me?
An explanation to why would provide civil discussion and seperate the bots/power-trippers from those who do want to contrbute civily. The conversation was on point, contributing civily. Nothing unconstructive was within my text, yet it leaves me baffled to why I was downvoted. Was it soley because I clash with someones view and they don't want to challenge it? I like to earn my "internet points" in a solitude fashion but it only takes one bad actor to screw the show.
Forcing the user to give a reason to why, well that would be a different story. Users further down the line could then decide if that downvote was the worth. If that user is a being a bad actor. For now, it's marked as 0 and you can watch the trend.
For a long time I thought HN was free of the kinds of bad faith, intellectually dishonest interactions that you find in other popular forums. I was wrong, it's just a different crowd (one that isn't so clearly left or right). There is a vague "orthodoxy" and the downvotes will flow if you speak outside it. Recently, my account has apparently been flagged such that I'm only allowed to post a few comments per hour. I'm also suspecting that some users employ some kind of black list or tagging system to punish accounts that have demonstrated wrongthink. shrug It's still better than Ars Technica.
"Disagree" is actually a suggested downvote reason at Ars, with predictable results. Then they published a number of articles that turned out to be perfect bait for flame wars, trolling, and upvote/downvote wars. Comments dripping with twitter-style vitriol and outrage. Engagement up, quality down.
Perhaps they did it on purpose to watch the trash fires burn and the ad impressions roll in, but it really destroyed the comment section SNR and stank the whole place up. I've mostly quit reading Ars except via HN.
I’ve never used it, but the way others do, here and on Reddit, I think of it as the “disagree” button, rather than “downvote”.
It can be frustrating when people use it in lieu of a comment, though. It means that, of all the arguments at their disposal, the person decided to expend only the minimal effort required to produce something that still qualifies as a reply. The communication via binary responses is a great illustration of the decay of modern discourse, if you believe in that sort of thing.
Forums have envolved in to social lands known as Reddit,HN, Twitter and the such. The only element that's shifted is how the way the content is presented.
Twitter is a glorified SMS/MMS message
Reddit is a glorified PHPBB forum
HN is glorified usenet.
Specific forums hold specific coversations which the above platforms are the same. HN is targetted towards the tech edge, reddit is targetted towards the average casual indiviual and Twitter/TikTok for the teenages but you still get the same output on all three.
As reddit to HN, The "downvote/upvote" mechanism is old, broken and blatent. Reddit introduced fuzzing to hide the real numbers and as a method to hide abuse. HN had the upper edge of only allowing selected users to downvote after a threshold but that itself feels bias. How many engineers read HN, and how many colleagues know each other with the power to conspire? They all have a problems of their own. HN for example has an problem that if you comment calling on a "fanboy" group, whether it's Elon, Apple, Google in discussion you end up getting lynched for it.
The karma-idealogy has been done and it worked when the internet was growing, however the internet has grown immaturely and we've never shifted away. We still live in an "upvote/downvote" generation where to the point now that folk are paid to farm for internet points. Controversial opinon, but it's civil.
Yeah HN down votes seem totally random to me. On reddit I can usually guess why I was down voted (advocating a viewpoint that goes against that subreddit's bias), but on HN I have no idea. I've been upvoted on things I would expect to get downvoted on and vice versa. It's so random I just ignore it as noise.
Most humans act pretty decently when they encounter other humans in real life.
However, this changes rapidly when they are using technology. Many people who are pretty decent otherwise can say horrible things on the phone or email. I have seen this a lot of times. So it is pretty important to think about ways for people to disagree without descending into some stupid flame-war.
Some online conversations remind me of a night at a typical pub.
Early in the evening, people are getting along just fine. By full sundown, some people are becoming boisterous. By the end of the night, fists are flying because everyone has offended everyone else.
Basically your typical online comment thread. HN does an amazing job at curtailing this typical behaviour. You could say dang is a good bouncer.
I haven't seen many (any?) online systems that seem particularly well-suited to arguments.
Nevertheless I like it when sites try to call out clear statements of position, supporting and opposing arguments, citations for evidence for claims, and rebuttals, while trying to reduce logical fallacies such as ad hominem arguments, guilt by association, etc..
It also seems hard to have a productive discussion of controversial ideas without some degree of third-party oversight and moderation, but moderation usually favors the biases of the moderators and usually generates animosity among those whose comments are suppressed.
>block users who derail conversations and use emoji to convey emotions in text
Blocking people who use emojis seems extreme. In fact, an emoji-based rating system could improve conversation by replacing the urge to comment lightly with a "thoughtful" or "informative" label. They needn't have any weight other than their visibility.
Some old-style forums implemented this to cut down on noisy "I agree" posts, and it worked. Maybe reducing conversation surface area reduces chances for conflict.
> We found that people like some features that are already present in social media, like the ability to delete inflammatory content, block users who derail conversations and use emoji to convey emotions in text.
I mentally inserted an Oxford comma before the last "and." Thus:
People like some features that are already present in social media, like the ability to:
* delete inflammatory content
* block users who derail conversations
* use emoji to convey emotions in text
The lack of an "and" between "delete inflammatory content" and "block users" unambiguously makes it a list of three items. The oxford comma, while useful, is not necessary here.
According to your "obvious" interpretation, the sentence is:
"We found that people like some features that are already present in social media, like the ability to delete inflammatory content, block users who do X."
Does that sound like a grammatically correct complete sentence?