The other side is that internet users massively project "astroturfing", "shill", "spy", "foreign agent" and so on, onto other commenters and opinions they happen not to like. That's actually the larger problem; in fact, based on countless hours I've spent investigating these phenomena on HN, it is overwhelmingly the larger problem. It's not a close call.

Users taking such cheap shots at each other degrades discussion in really bad ways, so we don't allow it unless they have some basis for what they say. If you look closely at how often it comes up, you'll see that people almost never provide any basis—not a glimmer of anything objective—beyond just disliking what someone else said. Someone posting an opposing opinion on the internet is evidence of nothing but that people have differing views. That is no basis for denouncing someone as a traitor or a criminal, least of all on a site that asks users to be kind and thoughtful and make substantive comments.

So HN's rule is that it's not ok to lob such accusations, or insinuate them, unless there's something more than pure imagination to go on. If people are worried, they are welcome to contact us; we investigate, and if the data gives some reason to conclude that abuse is happening, we crack down on it. But if we look and find no evidence of manipulation, we don't. That seems obviously the right policy to me. The alternative would be to punish users without evidence. That wouldn't be right.

Does that mean we're vulnerable to super sneaks who manipulate the site so cleverly as to leave no traces in the data? Yes, any internet community is vulnerable to that. The best we can do as a community to defend against that is to do what we should be doing anyway: refute bad arguments.

There is a ton of previous explanation of why we moderate HN this way in the archives: https://hn.algolia.com/?sort=byDate&dateRange=all&type=comme....