[tptacek]

Only for actual bad actors --- spammers, overt griefers, and people evading bans. A lot of HN's shadowban rep comes from Paul Graham's stewardship of the site (this whole site was a side-hustle of a side-hustle for him) and ignores over a decade of Dan's work professionalizing it.

Almost everyone banned on HN is banned publicly, with a public message explaining why.

[lagniappe]

>Almost everyone banned on HN is banned publicly, with a public message explaining why.

I would love for this to be the case, however I quite extensively investigate this phenomenon and this does not match what I've seen. I'd like for us to be better than shadowbans. In some cases, I don't even get to vouch, it's just a comment that is banned-banned. It feels the worst when they're saying something substantive to the conversation and we have no means to surface the comment.

Some type of annual amnesty consideration or something of that nature is in order, or soon we'll recreate other echo chambers that are slowly fading out.

[tptacek]

Every time I've looked into it, when you see suddenly and without reason ban-banned after a string of real comments, the backstory has been that it's someone with a track record under other usernames.

At some point, no matter what HN does, being comfortable with its moderation requires you to take Dan's word for things. I take his word for it on shadowbans.

Ironically, I'm irritated with moderation in the other direction: ten years of "if you keep breaking the guidelines under alternate accounts, we'll ban your real account" sort of makes my blood boil (people having long-running alts does that too), but I roll with it, because I couldn't do the job better than Dan and Tom do.

[6510]

I think the tech is advanced enough to recognize a user by their choice of words. Not to downplay the problem but currently a new user might start out with a few comments on the poor side then talk into the shadow realm indefinitely. In my not very humble opinion the asshole level of the poor comments can easily be lower than the asshole level of the response.

I wrote a fun solution one time where the document comes with a token that needs to mature for a duration depending on the user. Then, when your [say] 30 seconds are expired the input area is displayed but the submit button only appears if you input enough characters - where enough again depends on the user. If you are likely to make low effort postings I want at least 500 or 1000 characters worth of low effort. In even worse cases ill also hold the comment for moderation - until I get to it. (which might be a long time)

[lagniappe]

>the backstory has been that it's someone with a track record under other usernames.

This has gaps, as you know, and doesn't wash. Let someone turn a new leaf. Amnesty puts a stop to this.

[tptacek]

I don't think it does, no. I've seen people raise innuendo about this kind of thing for over 10 years and have never seen someone vindicated. Maybe you have an example you can share.

[basilikum]

I've had submissions and comments of mine on this very account shadowbanned before I got some karma. It's just how the spam detection system works. So no, it's not just for bad actors, but for anyone an automated system suspects to be spammy enough. It's also easy to see whether your submission was shadowbanned by just looking for it while being logged out; only works if you know you might be shadowbanned though. So unsuspecting people who just got falsely flagged might never find out they are not being displayed to others.

[lossolo]

They're also shadow banning/silently disabling your votes, and they will not inform you about this. You think you're voting on stories or comments, but you aren't if they perceive your behavior as "upvote too many flamewar comments, culture-war/ideological battle comments, or otherwise low-quality comments for HN" and "if a user has a track record of upvoting comments that break the guidelines and/or downvoting good comments, or voting in ways that seem unfair – e.g., voting based on political side or personal acrimony, rather than on the objective merits of the comment itself".

[tptacek]

This seems like an especially silly complaint on a site that is clear on the label about votes being just one of many signals deciding placement on pages and threads. We've known since 2008 that the HN experiment doesn't work if it runs off raw votes; you just get a front page full of memes.

[lossolo]

If this were clearly public (like written in the rules) then maybe it wouldn't be worth mentioning. But if it isn't, it's good for people to know, so they understand how their voting habits can affect whether their votes count, right? That's why I mentioned it.

[tptacek]

Perhaps you should acknowledge that your claim of them refusing to tell people about this is false.

[lossolo]

That reply feels needlessly adversarial. I'm not claiming they "refuse to tell people", my point is that this isn't clearly documented in the public rules and, as far as I can tell, users aren't notified when it happens (nor is it something staff states proactively).

I only learned about it after I asked via a non-public channel, with evidence. Otherwise I wouldn't have known, and I suspect most users are unaware. What I cited in previous comment is also from a non-public conversations.

If I'm wrong and it's documented publicly in rules or users are notified when it happens to them, I'm happy to be corrected, link?

[6510]

I don't have a down vote button with the comments. If my upvotes are irrelevant that button should also be removed.

[6510]

Maybe it is possible to create such a [hidden] page.