|
We tell people that we're banning them, and why: https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que... - but only once the account has an established history. We treat new accounts differently, for reasons I've explained before: https://news.ycombinator.com/item?id=21288858 https://news.ycombinator.com/item?id=20666742 https://news.ycombinator.com/item?id=20289994 It's a bimodal distribution. With established accounts, we warn them before banning, usually several times, and then tell them we're banning them and why. With new accounts, we don't say we're banning, because banned-new-accounts are overwhelmingly spammers and trolls—often serial trolls who have been banned many times and know exactly what they're doing. The cost-benefit of explaining why we're banning the account in those cases is completely different. To repeat from one of the comments I just linked to: it is a balance between transparency and defending the site against abuse. If we tried to give every banned account the same high-effort attention that we give established users, we'd do nothing else all day and still not get through them all. That would just be a new vector for people to DoS the moderators. A small number of abusive users can create a large number of disruptions. There is also a significant amount of spam, and if we told spammers we were banning them, they would spam us with emails demanding attention, asking why, and telling us how high-quality their articles really are. Actually they do this a lot already, and it's a pain. This does leave one class of accounts who unfortunately are the losers in this cat-and-mouse game: new accounts that get shadowbanned even though they were neither spammers nor trolls. It sounds like that is what happened in your case. Sometimes a new account shows up, and in its first post(s) behaves like a spammer or a troll would. That's when we use shadowbanning, but since we don't see the future, later it sometimes turns out to be a legit user who goes on to make good posts that get killed because we banned them. This happens and it sucks. We can't see into the account's future posts, nor can we ever know for sure that an account is spamming or trolling...it's all just pattern matching and sometimes a pattern matches on the early data points and diverges later on. The only solution I know of in such cases is to correct it later: either because we notice good comments that are [dead] and investigate, or because we hear from other users "hey, are you sure $username should be banned"? - and we look them up, see the mistake and unban them. It's particularly helpful to alert us to such cases, so if anyone notices an account that is banned and shouldn't be, we'd greatly appreciate hearing about it at hn@ycombinator.com. |