Tell HN: You can get ghostbanned for mentioning certain “unpersons.”

26 points by scaphist_302 ↗ HN
Pardon the throwaway account. I have a good reason.

About a week ago, I linked to an essay on AI written by someone who made a few people look bad, a long time ago, and got himself un-personed here a few years ago. The essay didn’t hit on any of his “touchy” topics and was inoffensive. I didn’t even name him. I just linked to it.

Anyway, I used an alt account for this and I’m glad I did because the account was “ghostbanned”. As in, it looked to me like it had been submitted successfully, but as soon as I logged out it disappeared.

You should be very cautious about what you post here. This is not neutral territory. Who knows what other bad-faith ghostbannings and algorithmic manipulations go on that we don’t even know about?

51 comments

[ 3.2 ms ] story [ 110 ms ] thread
You can see all shadow-banned submissions when you enable showdead=yes in your profile. I have it enabled so the "Who knows what other ..." argument is less scary, it can be seen.

You said you used an alt account. An easier explanation is that one of the anti-spam algorithms saw it as duplicate and auto-blocked the submission. That happens to brand new throw-away accounts all the time and it needs another existing HN user to click 'vouch'. Not a moderator, it's enough for another user to click it. Entirely possible simply nobody did. Can be time of day, can be the topic, and yes could be the "unperson" (your words).

Regarding the topic: I noticed AI-generated posts or posts of chats with AI don't go well (not enough users upvote) on HackerNews. Maybe it's deemed not relevant enough. Today there was a "ChatGPT Poetry: Trump vs. Biden" submission for example.

FWIW, OP says "essay on AI written by someone". I.e. not an essay written by an AI, but where the topic is AI.
Ah, ok. I missed that. I would help to have a link to that submission. Like I said it's not deleted, it can be made seen again and then (that needs a moderator though) the account status can be checked. The anti-spam algorithms are pretty good but not perfect.
Pardon my ignorance, but who is an unperson here?
[REDACTED]
That joke isn't helping. HN doesn't edit comments automatically. You wrote that.
The algorithm removed an adverb from the title string accompanying the url I submitted.
I only know Jamie Zawinski (one of the founders of Netscape and Mozilla.org). Not the person, not his opinions. But he hates HN - I think - and decided to show picture of a genitalia when people click on links to his blog. He doesn't want traffic from HN and doesn't want HN users to discuss his blog posts. So his domain is blocked.

[edit] I thought Foone, always good content, but they don't like HN either "I'm gonna have to email [HN] again and finally get them to just ban me from HN. this has happened too many times" https://twitter.com/Foone/status/1574140918478692352 but seems it's not https://news.ycombinator.com/item?id=32975173

Foone is definitely too good and sweet for this place anyway.
> decided to show picture of a genitalia when people click on links to his blog

Was this recently? I think that in modern browsers that has come to be locked down. I'm not fresh on the details, but the gist of it is that a given web page can publish a header which tells the browser what to do when users navigate away from that page. Whether to reveal that page as the origin or not.

I think HN can make it impossible for a site to know that a visitor "came from" HN.

My guess is Michael O. Church.
FWIW I don’t believe mentioning anyone gets you banned here, and that it’s more likely a low-karma alt account posting links tripped up some automated system.
They absolutely do shadowban users for having the wrong opinions here. Dang himself even admitted it to me When confronted.
That doesn't sound right. Where did this happen and what did I say?
Without a link, it's hard to evaluate this. I doubt you're getting "ghostbanned" for mentioning "unpersons". But this is posted under a throwaway account, so there should be no risk to posting both the link and the name of this supposed "unperson". Go ahead and clear things up for us, would you?

Somebody else on this thread seems to believe that HN will ban you for posting stories critical of Bobby Fischer. You can see why people are jaded about claims like these.

Moderators are human, with their own beliefs, biases, and values. People get ghostbanned for all sorts of reasons, but we should never underestimate the human element and ego desire to shape the narrative through unfair moderation.
Quite true, in terms of how it works. The way and how of ghostbanning combined with the biases and beliefs of moderators can be used to shape opinions and discussions, because "wrongthink" becomes an unstated defacto "restriction".

People can be given the impression that the majority thinks and believes a particular way about a subject, where in actuality, valid dissenting opinions were eliminated, banned, dissuaded, or afraid to comment.

I got ghostbanned here on another account the day I submitted a link to the story about how Bobby Fisher (the World Chess Champion) got tortured in an American jail. When I confronted dang over it he claimed against all logic that the timing was merely coincidental and I was being retroactively punished for other, different instances of wrongthink posts that I posted several weeks prior to the ban. This is how they operate here. If you post a borderline wrongthink post here, dang and his goons will scrutinize your history going back weeks looking for a superficicial justification to eliminate you.
Not one of the complaints in this thread, including the OP, come with any links. Why not let readers see for themselves and make up their own minds?
I can't say about the others or all cases, as don't have the bird's eye view or ever have been a moderator, but do know that significant sections of threads can be removed (censored?). Recently I had made a comment in the "The bottom emoji breaks rust-analyzer" (https://news.ycombinator.com/item?id=34775549), that comment and others just vanished. My comment wasn't flagged. Just an entire section of comments (before and after) seem to have been removed. In those cases of heavy editing, for whatever reasons, there is nothing to link to for readers to see.

Additionally, when a thread is shadowbanned/ghostbanned, there is nothing for other readers to see as well. Only the creator, yourself, or possibly moderators can see. The public will not be able to see, so links to it serve no purpose. I'm just saying what I have personally observed, so don't know all the history or the behind the scenes. If something doesn't publicly exist anymore or is not accessible, there is nothing there for other readers to make up their minds about.

If you mean https://news.ycombinator.com/item?id=34788024, your comment was a reply to a comment that got flagged. Flagged subthreads are collapsed by default. They're still there, just in collapsed form.

I don't know what you mean by shadowbanned/ghostbanned but everything remains visible to users who have 'showdead' turned on in their profile. This is in the FAQ: https://news.ycombinator.com/newsfaq.html.

> there is nothing there for other readers to make up their minds about.

On the contrary - if people mention specific links, we can say exactly what happened to that link and this would allow readers to make up their own minds. Most "I got banned/censored" claims never supply the link because then they get to tell a one-sided, and usually entirely distorted, story.

The rebuttal makes some good points. However, many are not going to see the contents of flagged comments nor understand the decisions behind them being so. Not to say that flagging isn't justified at times, but scores of posts that are removed with it, are not guilty of the same violation or the posters below the removed comment in violation may not even know what happened.

Agree that the OP should be able to provide a link to the shadowbanned thread (if they're not having trouble doing so), because of course you will be able to see it. The context of how shadowban or ghostban is used, is debatably valid. Because the user can see the thread, but it's no longer obviously available to the public.

As for showdead, seems doubtful that many would have it turned on or even be looking. Once a thread drops out of sight (likely top 100), seems most won't check (even more so if no comments) unless they are searching for something specific (which could be anything).

Possibly a good example of shadowban/ghostban thread, is this recently posted one (below). It also doesn't show as "dead" either, though maybe there is some time element, and can't be seen by the public.

https://news.ycombinator.com/item?id=34838236 (Robot war dogs commanded by soldiers using brain computer interface [video])

I agree with most of your points here. The thing to understand is that these are all tradeoffs. For example, it's true that collapsing flagged subthreads by default is unfair to the users who post substantive comments in that subtree and haven't broken the site guidelines. (Sometimes we rectify this by manually snipping them, so they can go back to the top level, if we see them and they make just as much sense at the top level.) However, it's still a major improvement to thread quality overall to collapse flagged subthreads by default. Moderation is fraught with these tradeoffs, and when we find any major improvement to overall quality, we pretty much have to do it. In this sense we're basically all victims of statistics here.

Re https://news.ycombinator.com/item?id=34838236: that got hit by a simple spam filter. Clearly it was a false positive (another tradeoff—no spam filters without false positives). Sorry! I've restored it now*. The funny thing is, I was just looking at that particular filter this afternoon and realized it might need some work—I just haven't had time to review it yet, which involves looking at everything it has killed recently and evaluating whether it's still doing more good than harm.

* ("Robot war dogs" is linkbait though, so I've taken that bit out of the title, as the site guidelines ask - https://news.ycombinator.com/newsguidelines.html)

The very positive thing is that you fearlessly addressed each point and provided more clarity. This is a positive reflection, because of your willingness to take the time to do such. Users usually don't know the other side, from the administrative point of view, so sometimes this can resolve misunderstandings.
I'm glad! Thanks for saying so.

It's true, a lot of this stuff doesn't explain itself and HN can be particularly cryptic. We try to make up for that by answering questions whenever people have them.

The evidence where you pretty much admitted that you ghostbanned my main after mass reports (which could have 99% been in bad faith and/or bots) flagged me and then you judged Me to be guilty of wrongthought and taken my away link posting ability has been posted above.
You're going to have to be more specific if you want a response. I really don't know what you're talking about. We're dealing with far too many users and posts to remember anything for very long.

I don't mind answering questions (if you have one?) but I need links. Also, links let any interested reader check what happened for themselves and make up their own mind. That's important, since the community is the jury here.

Edit: ah ok, you mean https://news.ycombinator.com/item?id=34853852. I've answered there.

Here’s the evidence for you. Fact: You ghostbanned my account “unknownaccount” after posting the Fischer article - allegedly for making “unsubstantiated” submissions despite not posting any submissions for a week prior. You admitted in your own comments that I was indeed linkbanned and it was just a coincidence about the Fischer link and that I was actually retroactively banned for posting unspecified “unsubstantiated content” Like what does that even mean? All my submissions were legitimate, thought provoking logical and relevant including some well-received whistleblow criticisms against Megacorps and including the likes of Microsoft and Cloudflare-which hundreds of users upvoted mind you. It’s understandable that a myriad of corporate elements wanted me silenced, but I think you fell into their play too much. You should really give the community the FREEDOM to see what it wants, stop taking down or editing posts which PEOPLE want to see. Not just what (your handlers, their handlers, other moderators, downvote&reporting botfarm ops, gov & political agencies etc) permit the sheeple to see. Please for the sake of this community tone the oppressive censorship down a little, Get rid of your dystopian mechanisms of [deleting] wrongthink, of shadowbanning and ghostbanning. Give less credence to false flag and bad faith reports which lead to silencing as what happened to my account. The right thing to do would be to reverse the ban as there was and is infact substance to all of my claims, which many reasonable people were able to see hence why over many people agreed with my posts yet so many of them got taken down anyways. Funny how that works.
I explained this to you some time ago (https://news.ycombinator.com/item?id=33445000): you were posting a plethora of unsubstantive Ask HNs that users were not only flagging but emailing us to complain about. That's a strong signal of community rejection, and after reviewing the posts as a moderator it was clear those users were correct.

Anyone who wants to check https://news.ycombinator.com/submitted?id=unknownaccount with 'showdead' turned on in their profile can see this for themselves.

I don't know what you mean by "the Fischer link" but the way this works is that we take away submission privileges from an account when they've been submitting too many off-topic or bad (for HN) things, and that causes the software to kill future submissions. So of course there was a time lag between the time that we did this and the time you noticed it—there always is.

Btw, I didn't say "unsubstantiated" but "unsubstantive", meaning that the posts you were submitting didn't contain much information (https://en.wiktionary.org/wiki/unsubstantive).

There’s a lot wrong with this. First off: The number of times a post is flagged or reported should absolutely have no bearing on whether it stays or gets deleted by staff. Why even bring this metric up then? In a community like this especially where every other user possesses the ability to spin up their own personal HN flagger-bot army with a residential proxylist / hacked email list. The fact you admit that you use variables like flag count and email report count to base your decisions is really disappointing. As it stands now the mob is running roughshod on your website- abusing you - and the algorithms - to take down valid posts & silence entire accounts that expose abuses committed by corporations or persons. I still disagree with the premise that my posts were “unsubstantive”. If anything, I was forced to make my posts increasingly more and more syntacticly succinct BECAUSE I was forced to by your mechanism which only provides a short window of time-a couple minutes-to blow the whistle and reach the attention of as many real people as possible before the bad faith takedown mechanism kicks in( bad faith humans & bots reporting the post & emailing complaints en-masse. My original posts about any given topic were on-topic, truthful, relevant, and that’s why they were gaining momentum with positive upvote counts within the short windows before takedown. None of my posts should ever of been taken down in the first place because they were all published factually correct statements that were both on-topic for this website and pertinent to its users. It’s your own fault that I was forced(as a person with a conscious) to repost the same ideas in an increasingly succinct and consolidated manner. There is a factual basis behind all of my posts. You were manipulated / social engineered into silencing valid information, with tragic consequences for the world at large. Taking into account our previous conversations I think you tremendously underestimate the impact a post can have on the world. This site is THE forefront of tech thought. You, dang, are the gatekeeper of the tech world’s information access. And, under your leadership your website is being used to facilitate the coverups of horrific abuses in the world of tech. Case in point: Remember that time you silenced my submission which blew the whistle against Twitter for blatantly allowing a public figure to distribute controlled substances(DIY sex change drugs) to minors? Yeah, good job coveringup that one. If you hadn’t covered it up it would of gone to the front page and gotten enough media attention that something actually would of been done in order to stop it. But no, you let the kids down. I wonder how many other Sexual Abuse conspiracies involving major Tech Companies have you been instrumental in the censorship of? Statistically likely many. I don’t know how you sleep at night having the knowledge that you and your website are openly being abused by corporations and private persons to takedown evidence of their wrongdoings. I implore you to do the right thing, re-evaluate your entire strategy for content takedown: stop taking down factually true, relevant, pertinent information just because it’s controversial. Your current mechanism and negligence is depriving the world of justice in a massive way.
When I recently posted replies to an HN thread that the Internet had been originally created by DARPA, and that the current controllers had no interest in free speech, e.g. Twitter and Facebook, my posts were immediately flagged. Since both facts are obvious and freely available, my posts were flagged because they contained ideas that either the HN moderators did not like, or did not want anyone to consider, or both.

Looks like HN is tending toward Slashdot's heavy censorship program. The golden days of CmdrTaco are over. That program has ruined open expression at Slashdot and it now reads like bald PR and advertising, which is why I don't bother with it anymore.

Too bad for Slashdot, and too bad for HN.

My post above has lasted for five minutes without being flagged. But is anyone else seeing it?
See for yourself using a private window. "HN moderators" are just HN users with more than 500 points of karma BTW.
Which post, the one about hot grits? ;-)

showdead=yes (from elsewhere in this thread) is a nice tip. Your post was RSS syndicated to brutalist-dot-report, which led me to this still-alive comment page.

Hey, didja know this site was created when a bunch of Internet trolls made fun of a new version of Lisp?

I stopped posting to or reading slashdot twenty years ago in part because the comments were so lousy.
There's always chaff in every message board. Slashdot comments used to bring occasional kernels of truth, which were worth it to me. My attention waned when Anonymous Cowards were no longer allowed. Email sign-in has a chilling effect on "free speech" and that, along with getting flagged for posting facts, is always the signal for me to stop wasting time with anything even slightly controversial on a message board.
Anonymous (not to be confused with pseudonymous) posting is a great way to attract all sorts of weirdos of the variety that drive out all the users one typically wants to keep.
(comment deleted)
And there's a big difference between pseudonymous and an email address. The first permits anonymity which promotes free expression, while the second tends to inhibit, even if the email is not visible to other users. Enforced verification via email provides a heavy trackable target for those with access to these things. Real journalists know this well.

You can always start your own message board and control all users and comments yourself. That's the only way to try to ensure that you won't be annoyed, but even then some old friend can come along and mess up your day.

I see it without flag.
Thanks
Btw. I am banned from posting links for change, in last 12 hours I posted two comments and zero links and when I try post link to ordinary news website I will get message that I'm posting too fast, what a joke, of course at same time this comment will pass, since I am just banned from posting links.

This is the controversial stuff I tried to post:

https://www.channelnewsasia.com/commentary/public-speaking-c...

Heavy moderation on slashdot? It’s news to me.

Every time I look at slashdot comment I find them inundated with ascii swastikas, all kinds of slurs and s*t slinging. I would have guesse they lost the battle against moderation.

I see your post in the RSS feed.

Never saw a swastika there, but if that's not being moderated, then the "moderators," if any, clearly don't care to have anyone reading nor posting legitimate comments. The site owners just want PR space for rah-rah about new tech. I don't care to bother with researching the history, but maybe someone can confirm whether Slashdot was "purchased" or something.
An established user posted that same link a week after you did and it got marked as '[dead]', there were three submissions from that site last year and two the one before, all are '[dead]' with one point and no comments.
> An established user posted that same link a week after you did and it got marked as '[dead]', there were three submissions from that site last year and two the one before, all are '[dead]' with one point and no comments.

What was "that same link"?

There's enough info elsewhere in these comments to figure it out. (FWIW I read the first bit up until the part about Barthes and skimmed the rest; you're not missing much)
A submitted link often gets removed automatically if it's a resubmission. Not sure what the definition of a resubmission is, maybe that the title or the link matches a previous (recent) submission.