What does HN do about these fake accounts playing some kind of SEO spam game?

(For the benefit of other readers: these people are talking about invisible spam comments in this thread, which are hidden unless you are an HN user and have opted-in to seeing moderated content on HN).

And where do I opt in? I have no idea what these comments talk about and why under this post

You can just click on your username in the top-right corner, and choose showdead="yes" in the preferences section.

Spoiler, most dead comments are usually just flame bait and utter drivel. HN does groupthink down vote opinions that are different from what's generally accepted here, but for a comment to be actually dead, it's almost always just complete trash.

If I had to guess, it successfully blocks them some 99%+ of the time, but relies on a anti-spam filter that can fail independently from the rest of the site.

And there are probably countless bot accounts with some token maturity, like not-green one here, that intermittently test the status and then quickly unleash spam floods if a gap is detected.

It's the only way to explain how little we see spam here, on a extremely high-traffic, high-value site with trivial signup and few apparent safeguards.

Nothing. They get flagged pretty quickly. Best not to comment or refer to them at all.

Oops.

Wow I've had see dead turned on the entire time I've been here and I've never seen that happen before.

What's the purpose of that and why all at once now?

I have a feeling that some places give more weight to outgoing links from highly reputable sites like HN. Thus downvoting them into dead is not enough.

Except you can't see dead posts unless logged in, and also have the option on.

So no SEO for those links.

Interestingly, they apparently are counted when showing total number of discussion comments:

> 20 comments

But there's only 20 if counting the dead spam comments too.

So an idea: One can use them, to make one's own submissions look more interesting (more people discussing).

Does that impact page ranking on HN? And what purpose would that serve for OP or the study about rat social structures?

That being said, the poster of this study does submit quite a bit. So maybe there's something to that.

Oh, it's just that I went to this discussion because there were (seemed to be) many comments, so I thought it was something interesting. Social proof. And I suppose others a bit look at the comment count too. But this time, turns out it's all just comments about comments :-)