Sorry for confusion. I wasn't trying to say that _one_ comment got downvoted to -12. His 2 top-level comments ("and this submission is just farcically wrong" and "This isn't correct at all.") ... were at various levels of grey so those 2 comments contributed -4 or -5 or more to the -12. I do agree his tone (especially the [dead] comments) contributed to negative scores but it can simultaneously be true that downvoters thought he was incorrect.
If only we had vote arrows and vote reasons, then we would know the disposition of the voters and their interpretation which is associated with the downvote. Slashdot has had this forever, and it works great with their user moderation and meta-moderation system.
> Not only upvoting the story but also downvoting posters to minus-12
You've misunderstood how downvotes work. That person's account has multiple downvoted and flagged comments in other threads. (You can see these if you have "showdead" set to "on" in your profile.) I don't think an account gets to -12 on one downvoted comment, I don't think that's currently possible.
Based on some of my comments that have been nuked from orbit, I thought there was a per-comment score limit of -3 (so a net -4) for karma scoring purposes.
A comment goes near-invisible at -3 or -4, so most individuals won't read it or downvote further.
A motivated brigade can still keep downvoting, e.g. more than 20 downvotes on a single comment.
Brigade downvoting could incentivize future self-censorship. But it's also a strong signal for the Streisand effect, which can be surfaced by a bot and reposted elsewhere.
Has that always been the case, or was that added in recent years? I've seen more than 4 downvotes on a comment in approx 2015 era. It was noticeable because it wasn't about the comment content, but all my comments on one story were being downvoted. By the time it stopped an hour or two later, total karma was down more than 40 points. Good to see there is now a floor on downvotes.
Yeah it was definitely the case back when you could see other people's comment scores that there were often examples far lower than -4.
Don't know when the change was made, this subject has made me notice just how long ago I first joined HN and the last time I thought about how the voting works was definitely years ago.
OT: In that same comment thread is also a banned user (first time I saw this) where I'm surprised they still have a positive karma count, given the absolute toxic comments this user (troll really) has been giving for months.
Edit: Maybe shadow-banned for the most part of 3 months.
Those comments were downvoted, but comments elsewhere on the thread (pretty far down) which provided anecdata supporting his point were not.
I think the bigger question we should be asking ourselves is why people who had counter claims weren't higher up in the thread. May not be a fault of the community, I'm not sure.
Also funny (worrisome?) in retrospect is people that believed the Tweet, thought Apple was behaving fairly, and argued that it simply wasn't a big deal.
I don't think that is either funny or worrisome, at least to those somewhat familiar with credit card processing.
The way credit card processing fees work is that there are two kinds:
1. Per transaction fees, and
2. Percentage fees.
Not all per transaction fees apply to every transaction but if one does it is the same amount on every transaction. Not all percentage fees apply to every transaction, but when they do the amount is a percent of the charge amount, with the percentage the same for every transaction.
The big transaction fee is the "authorization fee". For a small merchant this might be $0.25-$0.30. If you are bigger they might give you a lower fee. I work at a company that is small now, but had a brief period of high sales and we were given $0.20 which thankfully has stuck. I've read that this can go as low as maybe $0.15 and perhaps even lower for really big merchants.
Another transaction fee is the AVS fee, which is $0.02. You are hit with this on any transaction where you supply address information and ask the card processing system if that matches the on-file information for the card. MasterCard has a $0.0195 fee called the NABU fee, which stands for something like "National Brand Use", which we apparently pay for the privilege of using the name MasterCard.
For the percentage fees, there will be a certain base percentage you are always hit with, and then various others that apply to some transactions. For all but very large merchants the base might be around 2% or so. I'm not sure what it is for the very largest.
(The above is if you have a merchant account yourself and directly use a payment processor like Authorize.Net or Merchant e-Solutions (MES). If you go through someone one step removed they might simplify it into one fixed fee per transaction plus either a fixed percentage fee or a tiered percentage fee system where a given charge gets one of the tiers. Apple almost certainly does not go through an intermediary like that. They either go through someone like Authorize.Net or MES, or maybe even are big enough to have access at an even lower level).
Anyway, the thing about this is that when you process a refund all that is guaranteed is that the credit card user gets their money back. Those fees you paid to have the charge processed? Some of them stick, such as the authorization fee. Furthermore, the refund itself is a transaction that will incur some of the per transaction fees (I don't think it incurs any of the percentage fees but I'm not absolutely certain--credit card processing bills are often extremely confusing and I've never been able to figure out from looking at the bills from at least 5 different processors we've used exactly what fees apply).
Assuming that most Apple app sales are $0.99, and taking into account the processing the refund will incur at least some fees, and allowing that Apple certainly gets better rates than I've ever seen, I can still easily believe that the net fees Apple pays on a combined $0.99 charge and $0.99 refund are in the ballpark of $0.30.
And since the whole freaking credit card system is set up, or at least appears to be from a merchant's point of view, to make sure that of all the parties involved in a transactions--the customer, the merchant, the merchant bank, the payment processor, the credit card companies, and the issuing bank--it is the merchant that bears the costs when anything happens that is going to cost someone money, nearly everyone who accepts credit cards as payments reacted to the claim that Apple keeps the 30% with "...and in equally unexpected news, the Sun rose this morning".
What scares me is that most of us measure truth in terms of "how many people upvote/like this tweet/thread/post," which further promotes fake news that go viral. It works like this:
1. Say something that clearly will make the news, regardless of it being true/false.
2. By the time people try to debunk it, it has already made its impact and served its purpose.
3. Even when the truth comes out, since it's usually boring, it won't make the news that much.
4. Congratulations! Now you have further strengthened the Internet "echo chambers," which means those who heard the false news are not even interested in hearing the truth.
Another thing that worries me is that "quiet people have the loudest minds" (by S. Hawking), which I interpret as the silent, lurking community members who don't stand up to correct a wrong (e.g. out of fear of getting downvoted/grayed_out, or simply because they don't want to get involved in discussions, etc.) IMO, they are just as blamable as those who publish lies.
What scares me is that most of us measure truth in terms of "how many people upvote/like this tweet/thread/post,"
I certainly do not equate votes to truth, and I hope most HN readers do not. Voting on HN is just a way to avoid adding shallow comments. I always assumed that most of the people on here are used to treating the popular opinion with a grain of salt.
I believed this particular story because it seemed like something Apple would do, which maybe is a bigger problem in Apple's image. Or maybe I'm just cynical.
HN at least realized a mistake then posted updates to counteract the wrong information.
Imagine if this happened in more mainstream, more polarized communities where this Tweet would get posted, immediately and undeniably be taken as fact, and then promoted/upvoted without being taken down. And debunking something popular with facts will lead to it being parodied, downplayed, or ignored in favor of keeping the false narrative.
No wonder a lie travels halfway around the world before the truth puts its shoes on.
I've been downvoted several times for asking simple questions. No shouting or accusations needed. It's enough that you are seen as political, and your post will be downvoted. Sadly this is turning HN into a stigmatizing and self-censoring echo chamber. I'm glad that HN has posted this retraction, though.
I don't think this allows you to see flagged comments. There is an elite group on HN which are allowed to flag comments and these really will be hidden from everyone.
It's not really an elite group since I'm in it :-P You can flag after you get enough points. I can't say I've ever seen flagged things disappear. They may be marked "[flagged][dead]" but the content is still there.
It seems if enough users flag it gets killed instantly.
If not enough voters flag it might still catch Dans attention and he might manually kill it.
If a comment is dead we (users with over 500 points I think) might still revive it if enough of us vouch for it.
I'm not sure if we can revive a comment after Dan has used the banhammer.
Also I can confirm that unlike what someone thinks we who flag and vouch are not an elite group.
After 500 you get downvote and I guess vouch as well. You don't have to apply it anything, at least I didn't have to with the accounts that had those privileges, and I don't know anyone important around here, I'm just a bored sw engineer currently located in Northern Europe.
It's even less ‘elite’ — I have < 500 points and see a ‘vouch’ link on dead comments. The FAQ doesn't state the exact threshold; I'd guess it's 100 (at most).
It's not just one person normally. Not only you can click "flag" and find that no change was made, you can also vouch for flagged posts to make them visible again.
It’s not an elite group. I used to believe as you do, to a lesser degree. But even if that were true, you aren’t meaningfully disadvantaged by not being able to flag things. It’s like a super-downvote. You don’t even have downvote at 0 karma to my knowledge. There are karma tiers for feature unlocks, but they are pretty low, like 31 or something for vouch? I know there’s a link on HN with this info.
The lack of negative interactivity for new and lower karma accounts is meant to be a kind of training wheels period, where if you get karma, you’ll implicitly learn what is acceptable at HN, and until you do, your ability to embody and enforce the ideals of HN is limited to positive or neutral actions of upvoting and posting.
I think HN works pretty well. Yes, some wrong material may get upvoted to the top of HN, but then it usually gets corrected quickly. In fact, the correction is staying on top of HN longer than original and the comments correcting the error in the original thread have been upvoted, too.
That's a pretty efficient way to handle errors. I wish mass media worked the same way. My 2c.
It only takes one net downvote to get there. Even one fat finger could flip that, and I fat finger on mobile a lot.
Anyway, discussions on downvotes are boring and guidelines-breaking, and even worse when it's misinformation, as a single comment could only contribute -4.
I've emailed the mods to detach this thread.
Edit: And... this comment has promptly contributed -4 within ten minutes. I don't really give a crap since I've got enough points to withstand a scratch like this. But the irony of being heavily downvoted for correcting misinformation on a thread supposedly correcting misinformation...
I have learned, as most of us have, that communication technique has everything to do with how well information is received.
But IMO we need to stop acting like that’s a good thing. It’s an unfortunate reality.
Niceness in not a virtue. We should be careful not to elevate it to such.
Truth is more important that how something makes you feel. Honestly is more primary than politeness.
The problem with believing people based on how well they communicate should be obvious: if someone can communicate well, it might alter the chances towards more expertise, but it is not a guarantee.
At a certain point we start fetishizing polite communication.
This isn’t wholly different from liking people more because they are good looking.
It’s inevitable that people act that way, but we shouldn’t celebrate it, should try to resist it.
I've struggled with this myself and I've also found that stating truth and using civil words are not mutually exclusive.
Like you wrote, the words used can promote truth to a greater or lesser extent. Almost always a polite but confident phrasing increases the transmission and replication of truth.
It does take more work. It does take increased linguistic and emotional skill (and work). It gets easier with practice.
I mean they were factually correct but that doesn't take away from the extremely toxic way they interact with the community. I think the comment deserves to be downvoted (although that doesn't change the overall point that accurate information wasn't bubbling up).
This is their bio:
"HN was created as an example of how to foment Dunning-Kruger among a community. Nowhere else will you find such a wonderful collection of imaginary "experts".
Today's outrage -- someone who is clearly wrong about how refunds on the App Store work, yet the overwhelming majority of HN are too profoundly stupid/ignorant/myopic to see the error and just run with it.
Top of the front page, thousands of comments...original guy retracts it. Hours later the story is still here and still getting comments from the idiot base.
I'd be embarrassed to be herding this collection of dumbshits."
That sort of approach to talking with HN is going to be downvoted. And I would argue deservedly so.
It could be the other way around. The person came to HN in a good faith but faced unfair downvoting of otherwise reasonable comments. Which eventually made the person write in their profile what's currently written.
Sometimes people become toxic because they weren't treated fairly.
This thread shouldn't be about how the guy was a meany poo-poo pants, but about what is it with today's world that makes people fall for the most unbelievable things based on the flimsiest of evidence. People with harsh words speaking the truth are far less worrisome than mobs being suckered by obvious falsehoods based on 10 second video clips, a random tweet, or some random post on HN's.
Counterpoint: downvoting is how Hacker News has maintained a comparatively high quality focused discussion while having only one full time moderator. It's not without its disadvantages - certain things will get instaflagged off the front page, especially "sexism in tech" stories. But it's like a firewall. If you don't have one, you get pwned eventually.
Occasionally you get "Boston Bomber" moments, and something true gets downvoted. That is on the poster to figure out why and how to present it in a way that will be accepted.
(I have 60k karma, so I know how to play ranked competitive HN, but even I still get downvoted occasionally if I'm lazy about sourcing or needlessly confrontational)
(edit: first downvote received in less than thirty seconds. Walked into that one, I suppose. Now leaving the edit window at +3. Sorry, +8. You're not really supposed to threadsit your own comments like this, but since we're talking about the voting system I thought I'd mention it)
If you go against the grain of HN on Elon Musk, Tesla, India or the US (and many other topics) then your comment, like this one, is tanking. Which also shows as many regulars from 5 years ago have left, I seldom see someones comments I recognize.
So I do object your argument.
Comments are good to get some opinions on a topic if you dont know about the topic of the linked post. I still read topics therefor.
I was not around for the early days of HN, and I recognize certain posters by their usernames. Here's a few off the top of my head (I am able to give a short description of each, but that's more work than I want to put into this comment). Also note I may misspell since this is from memory: Bytecode-dev, DoreenMichelle, ChrisMarshallNyc, Arathorn. This tends to happen when the person focuses on one topic in their posts, like bytecode-dev and python (and I hope at some point, myself and funding FLOSS, but maybe I comment too widely for that). There's more people; the point is not necessarily that I can regurgitate their names, but that I'd recognize them if I saw them (for example, I'm blanking right now on Drew from SourceHut's username, but I'd absolutely recognize it on sight.
Every now and again I wonder whether something like "reddit enhancement suite" would be useful for HN. It probably already exists. The ability to tag usernames with your personal opinion of them and reminders of their place in the community.
I recommend validating your perceived bias. Look through the archive for posts about Elon for example. You'll find people arguing HN is biased both ways. It's even more common for articles about Apple. (And it's quite funny to find in different comments on the same story)
The true comment "This is not contradicting what I have said." making an argument to a comment about discussions is voted down. What more do I need to say?
> If you go against the grain of HN on Elon Musk, Tesla, India or the US (and many other topics)
I'm not even sure which direction the grain is on India. I've even seen Indians arguing vigourously among themselves on this, which is not surprising given how contentious Indian politics is.
I think downvoting should remain available but more strongly dis-incentivized, maybe for instance by rationing the number of downvotes that one user can issue in a day.
But in the end I don't know if it'll make much of a difference, a bad community will always manage to push bad content to the top, a good one will manage to reward insightful content. Technical solutions can only go so far to fix what is essentially a social problem.
Depends on how many points. It should scale with the amount of points the user has. A percentage based scheme seems fine to me. 0.1% your points for every downvote.
I'm not sure I'd like to spend six hundred points dealing with a single idiot; I'd have to develop a better battery of cut-and-paste denouncements instead. I'm not sure this would improve the overall signal-to-noise ratio.
Downvotes are already rate limited. I know because I run out of them within 10 minutes of browsing the site nowadays.
I couldn’t disagree with you more, by the way. The amount of throwaway comments, puns, etc. are out of control. We need stronger downvoting, or weakened upvoting for new users.
I prefer upvoting good comments to downvoting mediocre ones. I keep my downvotes for trolls/insults and reddit-tier "meme" comments for which I apparently have a higher tolerance threshold than you because I don't find them all that common. Maybe we just tend to read different stories though.
People are bizarrely attached to their magical internet points and such consequences would perhaps have most think twice before flippantly greying out a comment they only read the first sentence of.
With the rampant vote inflation I think the limits should be at least quadrupled.
That is how you kill a site. It should be easier to remove bad content than to create more crap. The table is already tilted towards those creating misinformation and spam, making it more costly to prune this is an incredibly bad idea.
Side A: 1000+ voted story, with Apple’s recent App Store sentiments, emotionally engaging. This is the mob.
Side B: A lone fact checker. The comment posted by this fact checker was not ambiguous. It literally said that this whole thread is false. That kind of gravity in a comment is different from your usual discourse.
Side B is overrun by Side A. I am seeing this in our society, not just on this forum.
I feel like it’s different than you and I discussing whether tabs or spaces are better. It’s being overrun by emotions. We need to allow criticisms to surface and counterpoints to be heard when they’re devoid of attacks/trolling. Downvoting is creating a rather reinforcing mechanism for the mob the feel good about their stance. Further fueling it.
I am spooked by the current atmosphere on the internet. And I’m spooked by many instances of mob mentality in our society. Note: I am not talking about US protests.
What makes side B a "fact checker"? How can you identify them as such?
The real problem is that for pretty much any claim I can find you a tweet saying it was false. That doesn't help. We're into relative assessments of credibility by origin here. It looks like the original tweet author has retracted, which is pretty convincing.
> Apple takes back from the developer exactly what they gave the developer. This has been verified by a number of people, and this submission is just farcically wrong.
How is this nonsense front page on HN? Is this community this clueless?
This is a pretty harsh opposition. Not a small thing to gloss over. Why was it downvoted?
Here is what I fundamentally don’t understand. You have one guy make an unfounded, yet emotionally strong claims (every developer hates the appstore) and now his word in written in stone.
To refute him, you need multiple people to show that he’s wrong, a written statement from Apple and consignment from the Queen of England. This seems broken
When an account calling itself "hn_check" posts a series of comments like these they're going to get downvotes across most of their comments:
"Every clown who repeated this and argued the unfairness of it, like rattray here -- delete your accounts."
"rattray is intentionally spreading hilariously dumb misinformation to make HN look like a bunch of stupid assholes."
"This place is a den of absolute idiots, time and time again. And when it is realized it goes to auto-dead time. Absolutely fucking embarrassing for anyone who is proud of their association with this shithole of stupidity."
"Don't ever change your adorable ignorance, HN. It's such a laugh.
And this will be auto-dead. I was spot on and completely right and got pummeled down by the imbeciles of this cesspool. Hilarious. "
Haha, so would I. I think I'd instantly suspect I've fucked up in some dramatic way. But HN is usually civility > truth. The community values politeness over correctness.
Useful to know that characteristic. I use a personal user-moderation tool built by a friend to remove these people from comments. I actually really like these threads since it lets me identify people with lower thresholds for truth and lower thresholds for what they'll amplify as truth.
Looking at his comments I see insults coming 3 hours after his original comments were voted down. Apparently, being civil did not worked out for him either.
That you are being downvoted to oblivion is another indication that something is wrong.
Slashdot had a better moderation system ca. 20 years ago - you needed a reason along with the downvote and people would think twice about downvoting for "comment says something that may be factually right but somehow hurts my feelings or aligns with orange man". Meta-moderation also helped.
The moderator here is doing a good job, but perhaps he's being overwhelmed lately by the polarization and extreme groupthink on many issues.
I would love to see some academic research on the topic of forum moderation. The problem is probably as difficult as fixing democracy, but it would be nice to see what progress we have made in this area.
Community moderation has been a problem for a long time. There's a paper on Lucasarts "Habitat" MMO from the mid-80s! (can't find it at the moment). There's also JC Hertz's somewhat sensationalised anthropology of online culture from the 90s, https://www.goodreads.com/book/show/2284572.Surfing_on_the_I...
A number of experiments were made and abandoned; Slashdot's different types of up/down vote that users could weight themselves, Advogato's trust network, and so on.
But I don't like the word "reputation" as it goes against the idea that people who are new might have good input too. Using reputation as a discriminator is very much like committing an ad-hominem fallacy. But perhaps that's the best we can do.
A big problem is that an apparent new contributor may not actually be either new, or even a person. Only then can you determine that they're not somebody you've previously assigned a negative reputation value to.
(mind you, this is not at all sufficient; a major problem of the present day is people tweeting abuse from positions of power under their real names)
> Downvoting has 13 years of work and stress testing by 10,000s of people, you don't just turn it off.
This may be an argument for its correct functioning in the technical sense but is blatantly ignorant on changes in society and communication. If a mechanism becomes misused and exploited after a community grows and welcomes people from other communities, it has to be reconsidered. Social interaction definitely doesn't scale smoothly.
I think this guy downvoted for being rude, not necessarily for being wrong, although probably everyone who downvoted him thought he was wrong as well, when he was correct.
This whole "cancel culture" thing is a 2020 thing, while HN has operated like this since when? 2005? HN has been "cancelling" comments since forever, and it's suddenly a problem? It's part of the rules of the game here :-) You play or you don't.
ok when I said he was correct I meant his main point was correct, which I did not see as the cancel culture thing but rather that Apple does not keep the 30% commission. The cancel culture thing I saw as more of the rudeness.
I've been downvoted for stating a simple sourced fact last week. It happens all the time, when there are echo chamber or when the truth hurt some fanatics feelings they downvote in order to hide the truth, truth which hurt their ridiculously high sensitivity.
https://news.ycombinator.com/reply?id=23977461&goto=item%3Fi...
Such people should be accountable and loose their ability to downvote (and to a lesser extent to upvote)
As they actively hurt the epistomological quality of HN comments
So here's the idea: we should be able to call a moderator/contributor, explain to him why you think the downvoting is unfair and he will give retroactions.
I don't see how that would be irrelevant.
I talked about async stack trace and the original blog post talked about it.
BTW a guy thanked me because he found my comment useful.
I disagree with this (tho I acknowledge the dissent points and I'm one of those who moans occasionally - with the corresponding downvote hit about unreasonable downvoting). I've also never (yet) downvoted anybody despite severe provocation :)
I think some limits on downvoting might be useful (some of which might already be in place, I can't remember).
e.g.
You can't downvote replies to your comment. (Already in place).
A limit to downvotes to a particular comment. (I thought this was in place but can't find a reference)
Time limit on downvotes. (I think this is already in place).
X downvotes use up Y karma.
I would also like a guideline that you should only upvote or downvote based on the quality of the comment rather than whether you don't like what you think of the commenter...
The point is a serious one and really shouldn't be downvoted.
On balance, I disagree, but social bubbles are a systemic risk in our online communities, and the way downvote buttons are often used can promote a groupthink tendency,
That a profile where half the comments are rude gets downvoted? If I want to get called an asshole and read rude, barely intelligible comments I'll go to 4chan. You can disagree and present well-reasoned arguments without devolving to "LOL WAT? U IDIOTS"
> We need to remove downvoting temporarily for next 6 months.
"Need" is an overstatement if I've ever heard one, wow. Sort of an arbitrary timeline, too.
> Cancel culture.
Equating downvoting and community moderation with "cancel culture" is pretty far off base.
> Dissent and attack on free speech.
You have absolutely no right to free speech on HN, nor should you, nor did you ever, nor is that what free speech actually means.
The number of people who equate "the government shall make no law abridging the freedom of speech" with "I can be an asshole in a private forum and you're not allowed to call me an asshole" is astounding, and that is the real problem here.
> Smearing of truth.
By downvoting rude comments?
> People are afraid to speak up.
They're demonstrably not.
> Chilling effect.
Word salad.
> @dang - this is urgently needed.
It's not.
> It’s raising alarm bells like no other to me.
By most news accounts the US is careening toward a totalitarian apocalypse, China is literally sending train loads of Uighur muslims to re-education camps, and we're in the middle (or start, in the US) of a global pandemic, but HN's community moderation standards and practices are what's got your spidey sense tingling?
Not only upvoting the story but also downvoting posters to minus-12 who tried to counteract it with correct information:
https://news.ycombinator.com/threads?id=hn_check
[ footnote: his profile was -12 at the time I wrote the comment: https://news.ycombinator.com/user?id=hn_check ]
[ another edit:]
Sorry for confusion. I wasn't trying to say that _one_ comment got downvoted to -12. His 2 top-level comments ("and this submission is just farcically wrong" and "This isn't correct at all.") ... were at various levels of grey so those 2 comments contributed -4 or -5 or more to the -12. I do agree his tone (especially the [dead] comments) contributed to negative scores but it can simultaneously be true that downvoters thought he was incorrect.
This was how his downvotes looked early this morning: https://imgur.com/a/GuTffzP
The top-level comments have since been upvoted so they are no longer greyed out.