I think shadowbanning is unjust, and if is someone is violating the rules nonstop they should be made aware of it through notice and an actual ban.
I think shadowbanning is unjust, and if is someone is violating the rules nonstop they should be made aware of it through notice and an actual ban.
Shadowbanning is reserved for spammers and for new accounts that show signs of abusing the site, e.g. by trolling or egregiously breaking the site guidelines, or if there's evidence that they're related to past accounts that have been abusing the site. It's not a good idea to use explicit banning in all of these cases (for different reasons).
Here are some longer past explanations:
https://news.ycombinator.com/item?id=23686672
https://news.ycombinator.com/item?id=21288858
https://news.ycombinator.com/item?id=20666742
https://news.ycombinator.com/item?id=20289994
This system has drawbacks—banning new accounts means guessing—by definition, there isn't a lot of information to go on—and we guess wrong sometimes. But we try to correct those errors later. Btw, if anyone notices a banned account that shouldn't be banned, please let us know at hn@ycombinator.com—we've unbanned many accounts that way.
In fact I'm just working on a ban review system that will make it much easier for mods to track banned accounts that are still posting and unban the ones who are making good contributions (whether because we guessed wrong when banning them, or because they've reformed since having been banned).
On rare occasion I've seen [banned] accounts as well with showdead on.
Weird thing I've noticed is [dead] accounts that aren green tend to be typical troll, spam, or people who can't calm down during certain discussions. They may even have some comments still love.
[banned] accounts on the other hand tend to be completely deragned. They seem almost obsessed. Despite being banned they continue to make essay posts about whatever it is that got them hanne over large periods of time. Could be bots, but the complexity and variety of the language in the posts tends to be moderately complex.
I've definitely seen "[dead]" on accounts which, if you go into their post history, were told by dang that they were being banned.
AFAIK shadowbanning on HN means your comments default to "dead", as if they had been flagged. You can see them if you enable "showdead" in your profile, and if people vouch for them, they can become visible again. See Terry Davis' comment history for examples (this is the only shadowbanned account I know of): https://news.ycombinator.com/threads?id=TerryADavis - if you look a few pages in, you'll see some live comments of him talking about the early days of Ticketmaster and such.
To me this seems like a better solution than simply kicking people out who are unable to uphold the minimum guidelines necessary to engage in reasonably civil discourse. What would you suggest as an alternative?
Shadowbanning might be effective against actual spammers but preventing the user from seeing that they were banned removes even more accountability from the moderation than other tools which makes shadowbanning particularly prone to abuse as well as mistakes that won't get fixed. I think it's pretty dishonest to punish someone without telling them they are being punished.
It's also worth pointing out that shadowbanning is more effective against good faith users than malicious users since all it takes is to check using a disconnected session to uncover the ruse - something that will hardly be difficult for a motivated malicious user.
I would disagree with the form of banning you are describing, however.