Most active commenters
  • dang(8)
  • largehotcoffee(3)

←back to thread

142 points mzs | 20 comments | | HN request time: 0.001s | source | bottom
Show context
JohnJamesRambo ◴[] No.19401632[source]
I’m not so sure Hacker News is free of the same group. Post an article critical of China sometime and watch the comments. People genuinely posting opposing viewpoints is fine and normal but there is something very uncanny valley about most of them.

https://en.wikipedia.org/wiki/50_Cent_Party

replies(7): >>19401747 #>>19401877 #>>19401917 #>>19401961 #>>19401994 #>>19401998 #>>19402230 #
1. dang ◴[] No.19401961[source]
I appreciate your concern for HN quality, but this kind of comment is the reason why we have a site guideline asking people not to insinuate astroturfing without evidence. If you think you're seeing abuse, the guidelines ask you to email hn@ycombinator.com with specific links so we can look at specific data. We always look. Occasionally we find it, and when we do, we crack down on it hard. But it's rare, unless you count users getting their friends to upvote their startup or whatever, which is a different phenomenon. And the cases we've seen have basically all been of corporate abuse, not nationalistic.

Overwhelmingly the most common case is people accusing others of posting in bad faith merely because the other's view is so far from their own that they can't conceive of them having it for legit reasons. This is a reflexive reaction—a feeling that we all need to recognize and stop ourselves from expressing in raw form. When people vent it into comments, the result is either war between the two sides, or, if one side outnumbers the other, an ugly mob dynamic in which a few people are ganged up on for being different. Those few either leave, or they become resentful and break the site guidelines badly themselves, as a way of lashing back against unfair treatment. All these outcomes poison the community.

Not to pick on you personally—it happens because of how human nature reacts to the weird conditions of the internet, which we're not wired for. It is hard for all of us to grasp how large and diverse the community is, and how divided it is on divisive topics. Nationalistic themes are some of the most divisive ones, and unfortunately are growing more common these days.

https://news.ycombinator.com/newsguidelines.html

replies(4): >>19402129 #>>19402212 #>>19402694 #>>19403358 #
2. afpx ◴[] No.19402129[source]
Dang, are you able to report statistics on how many accounts have been banned or considered suspicious? I don’t doubt that HN takes this issue very seriously. But, it would be useful to some of us to see how often astroturfing actually happens here (and from which countries it seems to come from).
replies(1): >>19402408 #
3. gatherhunterer ◴[] No.19402212[source]
If I am understanding his/her comment correctly, its point is that this does happen sometimes, not that it is occurring in any specific thread or being done by any specific users. To that end your response seems to confirm that but asks us not to talk about it, even in the general sense.
replies(1): >>19402447 #
4. dang ◴[] No.19402408[source]
I don't have anything exact, but we've banned perhaps a dozen networks of accounts for corporate astroturfing over the years. Cases where the issue was nationalistic are much rarer. I recall only one, and it was years ago. (That of course is not to claim we aren't missing some.)

If we go by the data we actually see, the phenomenon itself is vanishingly rare. Why is there so much commentary then? I can think of two explanations: either the foreign spies are cleverer than we are, or there is something in how human nature meets the internet under current social conditions that is leading to mass projection. And of course it could be both—but how are any claims about the former falsifiable?

5. dang ◴[] No.19402447[source]
The problem is that the people who talk about this have no basis for the claim. It feels like it's happening, so they say it is. Other people also say it is, so there is social proof. Anyone who disagrees is reframed as a shill, puppet, useful idiot and the like, leaving no feeling of uncertainty internally.

As far as I can tell this dynamic has nothing to do with reality. Reality is that the HN community has millions of people, is diverse in many ways including internationally, and is divided on divisive topics. That is already enough to explain the comments that show up. But none of us is wired for dealing with that. We're wired for loyalty to our tribe, needing to feel safe, and suspecting outsiders.

It's painful to encounter a sharp opposing view, and people are angrily sharp on divisive topics. Reframing the other as a foreign spy or whatever insulates you from that. It relieves you from considering what truth there might be in that view, and reinforces loyalty to your own—at the cost of feeling surrounded by infiltrators and enemies. This is poisonous to thoughtful discussion, which depends on people being willing to open to differences and truths they may not yet see.

Since the actual phenomenon is vanishingly rare compared to the insinuations people make about it, we have a rule that users not post such insinuations without evidence. A feeling is not evidence. Even the sense "there are a lot of green accounts saying things I disagree with" is usually just a feeling, because we notice things we dislike more vividly than we notice anything else [1]. People on the opposite side have the opposite perceptions, but the identical feelings.

Notice how when people post claims about astroturfing, trolls, and spies on HN, they don't include links; just as when they make claims about "threads about $topic", they don't include links. Why is this? If the perception were of reality rather than a feeling, specific examples would always be available, yet in practice they almost never are. The discussion falls apart when specific cases are mentioned because (a) people never agree about those, and (b) such data as there is never supports the claim. The discussion always stays in a mist of generality, breeding bacteria of suspicion.

[1] Does anybody know or have a name for this bias? It's a huge factor in these discussions and it needs a good name.

replies(2): >>19403123 #>>19405357 #
6. Balgair ◴[] No.19402694[source]
Oh wow! I never knew that existed! Thanks for the email address. Do you all take in temporary email addresses or are those sent straight to spam? I'd love to send some in anonymously.

Also, thanks for all the hard work here on HN. I know it's not a glamorous job, but yall are doing a great job handling all this. You work hard and it shows.

replies(1): >>19402756 #
7. dang ◴[] No.19402756[source]
We look at all emails and people can email anonymously if they want to. Some go to spam, but I personally check the spam bin and fish out every genuine email that I can. I'm sure we miss a few, but not for want of effort.
8. cwkoss ◴[] No.19403123{3}[source]
How are you able to use data to determine whether someone is part of a coordinated disinformation campaign?

I agree that these are probably exceedingly rare on HN, but it seems technically implausible that you'd be able to accurately identify them.

replies(1): >>19403438 #
9. largehotcoffee ◴[] No.19403358[source]
You say that, but here's a perfect example of what appears to be an obvious Chinese account that is currently unbanned.

https://news.ycombinator.com/threads?id=thetechlead

replies(2): >>19403578 #>>19404162 #
10. dang ◴[] No.19403438{4}[source]
That's in the class of things we can't explain without ruining them, but I can at least offer a couple things. First, we're careful not to say anything beyond what we see in the data we have. For sure a sufficiently smart campaign is going to exceed our ability to detect it. But what can one meaningfully say about that? We have to stick to what our flashlight can show, and trust that it is at least more reliable than no flashlight. I can tell you that when it does light on an accusation of astroturfing, it nearly always tends against the accusation.

Second, some of the analysis can be done by anyone who wants to. When you encounter specific claims of astroturfing or shillage, look at the history of the commenter being accused. Most of the time their track record makes it implausible. If someone has been posting to HN for five years including about, say, garbage collection in Julia, what are the odds that they're secretly a foreign agent? Far lower than that the other user tossed off an accusation without pausing to look. It's usually not even a serious question.

Now consider that these demonstrably low-probability cases are exactly like the rest of the accusations people post here, and one has evidence for a common mechanism underlying the entire class. I don't assert (how could I) that there are no cases of genuine manipulation that fall outside it. But after looking at thousands of such claims, I believe that this and similar tests account for nearly all of them.

(Edit: it's different in cases of startups or projects trying to game HN to promote themselves; that's super common. And more sophisticated corporate astroturfing is something we've occasionally run into. But on these political, national, and ideological issues: zilch.)

replies(1): >>19403990 #
11. yorwba ◴[] No.19403578[source]
That person is 100% not a shill. I think your comment is a prime example of what dang meant by

> Much more common, however—by far the typical case—is people suspecting someone else of posting in bad faith merely because that other person's view is so far from their own that they can't conceive of anyone having it for legit reasons.

replies(1): >>19403751 #
12. largehotcoffee ◴[] No.19403751{3}[source]
> And now I'm into the 2nd year of my startup. As contrary to popular belief, the Chinese are true entrepreneurs and the society is generally very supportive for changes. And government's stimulus is insane. For example, my company got six million yuan (almost $1 million) fund from the municipal government when it's just established, with almost no requirement and absolutely no string attached. Free money and that's all. Probably the best place in the world to start a company and I'm greatly thankful.
replies(1): >>19403887 #
13. yorwba ◴[] No.19403887{4}[source]
By that measure, founders of YC companies are shills when they comment on their experience with the application process. Being happy about their loose investment requirements is not the same thing as being paid to spread misinformation.
14. cwkoss ◴[] No.19403990{5}[source]
Interesting, thanks for the response and the work you do.

"Oh no, now all the coordinated disinformation campaigns will start posting five years of high-quality good-faith technical discussion on a variety of topics to evade your heuristic" /s

15. dang ◴[] No.19404162[source]
Obviously we don't ban people for being Chinese. Presumably you mean it's "obvious" that they're a Chinese government agent. If so, your post is illustrating the very dynamic I was writing about. Your intention is positive, to protect the community, but when you express it this way, the effect is to poison the community you mean to protect.

I'm familiar with that account. Their posts, and what private data we have, are completely consistent with who they say they are: a former Google employee and startup founder who has lived in both China and the U.S., has a Kubernetes war story, opinions about Python, Go, PHP, software deployment and so on, and who is frustrated by comments about China here because they feel many commenters don't know what they're talking about. It's natural that someone who lived many years in both countries would feel that way. Some of their comments have broken the site guidelines, but that's a separate issue—and who of us wouldn't, having our integrity attacked outright like they have?

This is clearly a case of somebody being singled out for suspicion because they have different views, formed by different experiences, than others here. When users do that, it puts us in toxic territory. Is it ok to accuse people of being government agents, shills, spies, or astroturfing, just because they have a different view on some geopolitical or economic question? Obviously we need to not go there.

It's fine if you're not persuaded—I don't expect that—but please consider the downside of being wrong. What if this person is as innocent as you are, motivated by the same things as you? Can you imagine what it would be like to show up here and see it debated whether you're a spy and a liar? Even a single case of someone being unfairly subjected to that is unacceptable. If the community is to avoid "sinking its teeth into itself without realizing it" (Schopenhauer's memorable phrase) and falling into a poisonous swamp, we need a presumption of innocence. And so we do: the guidelines say Assume good faith. If you stand on the dry ground of that assumption, I see no path that gets you to that user being a bad-faith actor any more than you or I are.

I feel bad about holding up an individual user to some sort of public trial like this (another reason why the guidelines ask people to email concerns to us rather than posting them here)—can you imagine what that must feel like? But since the issue is the integrity of the community and its moderation I feel like I'd better say something.

replies(2): >>19404425 #>>19405057 #
16. largehotcoffee ◴[] No.19404425{3}[source]
Thank you for the well thought out response, in that case I am happy to concede I was wrong. Let my post serve as an example of what you were pointing out.
replies(1): >>19404837 #
17. dang ◴[] No.19404837{4}[source]
That's an incredibly generous response. Thank you!
18. meruru ◴[] No.19405057{3}[source]
You're fantastic dang. Thank you for helping keep HN a great place.
19. meruru ◴[] No.19405357{3}[source]
>Does anybody know or have a name for this bias?

Uh, confirmation bias? Your description is a bit off, it's not that "we notice things we disagree with", but that we notice things that confirms our previous beliefs, in this case that green accounts post opinions that you think come from spies or astroturfers.

replies(1): >>19405719 #
20. dang ◴[] No.19405719{4}[source]
IMO these are distinct phenomena. This one comes more from the fact that pain makes a stronger impression than pleasure. Running into an opposing opinion online is painful.