Nine times out of ten, the cabal makes good calls about hellbanning people, and the bans improve the site.
But when they get it wrong, boy, do they ever get it wrong. For instance, Paul Graham personally hellbanned Maciej Ceglowski for calling Sebastian Marshall a "prolix douchebag". Maciej's comment was not a high point of Hacker News decorum and in that instance probably degraded the site slightly, but no reasonable person could have looked at 'idlewords comment record and not come to the conclusion that Maciej is someone we want on the site. Sometimes, you take the bad with the good.
There have also been times where it's apparent that there is, if not "automatic", than at least "mechanical" banning for reasons other than nasty comments. I don't know or care what the abuse mechanisms being tripped are --- I buy into the idea that in this case, obscurity is making some baseline level of security cost-effective for Graham and his team. But it's important to know that whatever mechanical process is banning people for abuse, it's imperfect.
Ultimately, I think Graham is in a bind here. A more typical forum site would enlist more community support to help police the site. Graham can't do that, because HN also runs important parts of YC's (very sensitive) applications and communications process.† I have a hard time holding this against him. Managing a sprawling forum site is a hard problem. Doing that while having the site do double-duty as the front-end for your startup seed fund system must be a nightmare.
So I'd generally say: assume good faith; don't make a production of the mistakes; but, know that the mistakes do happen, are pretty annoying, and are worth at least remarking about.
† (It's worth knowing that YC members also have some silent perks on HN, beyond the job ads).
Thanks for correcting that. He did have comments killed; that's how we noticed he was hellbanned. And, like many reasonable people would do, once he found out he'd been hellbanned, he left the site, which is my point: his leaving was more our loss than his.
Mistakes happen. I'm just pointing this out as evidence that this particular form of site policing is error prone.
I'm definitely not inclined to rant about the unfairness of it all, because I do not believe I could do a better job running a site like this.
<assuming notion="your edit is in regards to tptacek's response">No, what tptacek actually said was: "he did have comments killed; that's how we noticed he was hellbanned".</assuming>
I personally remember seeing a few that were dead after the "prolix douchebag" one when I was looking into it. (I am known in my friend circle to maintain a database of users that are hell-banned; so, when Maciej, a long-friend-of-a-close-friend, seemed to be hit, I was notified almost immediately.)
I also remember those comments mysteriously being un-killed when his account became un-banned. (So, if you are using the current status, that isn't accurate anymore; or, if you are relying on us to not have seen the temporary status, we did.)
I only banned him while I was waiting for a reply to the email I sent him. In fact I thought I'd already unbanned him, though I noticed just now that he was still banned. What a tempest in a teapot.
Incidentally, this post and its predecessors illustrate perfectly why the site guidelines ask users to contact us about problems with moderation rather than posting on the site about them. Though nearly empty of content, they get upvoted faster than anything else.
I would assume they are typically upvoted for a reason. Obviously you have access to data I'm not privvy to, but 400+ upvotes in that little time would, I imagine, represent a significant portion of people who read the post. Obviously the site is no democracy, and I think we all understand that. But I would also guess that user feedback is appreciated (hence the request to send feedback via email). I would also think that many users have the same questions, but do not bother with an email because it is not really worth the hassle, and upvote a thread like that just to voice their agreement [else they would not upvote, unless they were simply trying to troll].
The whole scenario is fairly minor, but it seems to raise at least a small concern among what I would guess is a respectable portion of the userbase.
I can relate. I posted a story about Apple's disruption at the POS talking of how the new retail App allows for purchases in store without ever having to contract a store rep. I posted that alongside why I believe this is important considering the missed opportunity that was GoogleWallet. The post got front page before getting sunk by mods. I dont know why, but there was nothing sensationalistic about the title or the content. And this is a space that I have considerable capabilities in. And I was not trolling.
Since then none of the stories I post under this handle ever see the light of day. They cannot be voted up and they just face a slow death. I stopped spending any time posting on other threads since then because I cannot see the value in contributing to stories if my stories are not allowed to play fairly.
During my hellban about two years ago (I posted an article which had been silently banned, and was flagged for spam, I guess?), I was anonymously advised that the HN moderators were "capricious and tone-deaf", and that my recourse was to contact PG directly.
After a curt and unpleasant email exchange, I was back on my feet. Pretty stupid that I had to bother PG at all, and even stupider that my one interaction with him was so needlessly distasteful.
Not this time I guess, it was pg personally who hellbanned him. I hope he listens and drops it.
I think it's massively disrespectful to users to hellban and delete from view strongly upvoted posts, or to silently mass-downvote them. Pg and mods, respect your users' opinions a bit more than that, please...
PG has the right to curate HN posts as he sees fit. I just think a little more transparency would go a long way. Hellbanning is, by design, sneaky. The fact that it's the predominant ban method is extremely frustrating for a non-malicious user.
Violent moderation leads to people being annoyed, some questions, more moderation; a minor outrage, bans;amajor outrage spilling to social media, nuclear wipes. Damaged community, sometimes even dead community.
If you see a flamebait post, you delete it.
But if you see a question about where the original post go, and you delete it: then you just started a toxic flame war with your own hands.
And "guidelines" never help. When people are outraged they would not observe same rules they supported a day before. Sorry, not going to happen, so don't blame flame war on public.
I consider hellbanning a rude and deceitful tactic that should only be used against users who immaturely refuse to accept a regular ban. Having hellbans as the only type of ban is rather odd.
It's hard to detect whether someone refuses a regular ban. People use proxies. It's more trouble than it's worth. A account hellban is the simplest ban that works.
Whenever you give some people more power than others, in any form, abuses of that power will eventually happen. Police, prison guards, bosses, government officials, parents, website moderators, etc, etc, etc. The system becomes dysfunctional.
Power just amplifies people's egos, personal insecurities, and prejudices. We can't deal with it.
Though the most successful founders are usually good people, they tend to have a piratical gleam in their eye. They're not Goody Two-Shoes type good. Morally, they care about getting the big questions right, but not about observing proprieties.
If neglecting to observe proprieties is a characteristic of the most successful founders, surely we can shrug off a minor violation of the site's rules that led to what many thought was an interesting discussion.
Basically you can't use HN to discuss itself. Although I'm not a believer in the "wisdom of crowds", I think many-to-many discussions about HN could be more productive than many-to-pg private emails. Maybe there are some lessons to be learned from MetaTalk here.
Is it just me or did this get post just get banned for a few minutes and then return?! Admins can't make up their minds now? Regardless, glad to see it back.
Update: Yup, had the dead one in another tab still and now it's back. For now...
Update 2: It's now dead again. It looks like it only came back for about five minutes.
Update 3: And it's back again. Really wish I had put times on my updates now. Oh well. I wonder how long it will take until it gets pulled yet again.
I've had plenty of time where I was hell-banned through IP ban or something to that effect, never knowing what the hell the reason was. Often, I could just browse it with Tor, but obviously few, if any, sites are worth doing that.
Is this some kind of weird flood control or a opaque and stupid moderation policy I am not aware of? I am not the kind of person to call people names, and I have often received the supposed bans after I have written some of my most popular posts or threads.
Moderators may make mistakes here and there; they are human so just deal with it. It's what makes this place great. The hand of god's helpers has always been here, most of the time you don't know it.
Can you imagine how snarky this place would be without it? I don't think (reddit * intellectuals * anonymous) would sustain for very long.
Who are you talking about? The linked site is down, so I can't even tell who this is about. Assuming the user's name is the same as the domain name (jcs), then that user hasn't posted much of anything as far as I can tell... certainly not enough to be considered a top 50 user.
But when they get it wrong, boy, do they ever get it wrong. For instance, Paul Graham personally hellbanned Maciej Ceglowski for calling Sebastian Marshall a "prolix douchebag". Maciej's comment was not a high point of Hacker News decorum and in that instance probably degraded the site slightly, but no reasonable person could have looked at 'idlewords comment record and not come to the conclusion that Maciej is someone we want on the site. Sometimes, you take the bad with the good.
There have also been times where it's apparent that there is, if not "automatic", than at least "mechanical" banning for reasons other than nasty comments. I don't know or care what the abuse mechanisms being tripped are --- I buy into the idea that in this case, obscurity is making some baseline level of security cost-effective for Graham and his team. But it's important to know that whatever mechanical process is banning people for abuse, it's imperfect.
Ultimately, I think Graham is in a bind here. A more typical forum site would enlist more community support to help police the site. Graham can't do that, because HN also runs important parts of YC's (very sensitive) applications and communications process.† I have a hard time holding this against him. Managing a sprawling forum site is a hard problem. Doing that while having the site do double-duty as the front-end for your startup seed fund system must be a nightmare.
So I'd generally say: assume good faith; don't make a production of the mistakes; but, know that the mistakes do happen, are pretty annoying, and are worth at least remarking about.
† (It's worth knowing that YC members also have some silent perks on HN, beyond the job ads).
e45f8d80989cd423c9bfa66738395f8d71c89d8f