Comment by TheCoreh

Absolutely. We review the list of stories that set off that software penalty and restore the ones that are clearly not flamewars. No doubt we miss a few, and also - not everyone interprets these things the same way. But if you (or anyone) notice a case of a good thread plummeting off the front page, you can always get us to take a look by emailing hn@ycombinator.com.

User flags, once they've accumulated above a certain threshold, have a downranking effect. Pretty sure this is in the FAQ: https://news.ycombinator.com/newsfaq.html.

We try to, and often users help by, posting links to the previous discussions in the thread. But there isn't enough time to do that in all of them.

In this case, you can see from https://news.ycombinator.com/item?id=39204186, but this comment was flagkilled, probably because of the personal swipe in it. (You can still see it but only if you turn on 'showdead' in your profile.)

I could go on! because there's endless detail one can go into about these calls. But you "didn't necessarily want to dissect every little story" :)

"Transparent" means different things to people, but if you mean a full moderation log: I think most likely it would produce more questions and effort, for no clear gain. I've written about this over the years: https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...

Here's how I look at it: if trust is present, then we don't need to publish a full log, as long as we answer questions when people ask them. That degree of transparency has been available here for many years. If, on the other hand, trust isn't present, a moderation log won't create it. It will just generate more data for distrust to work with—and distrust always finds something.

Thus our focus is on building trust with the community and maintaining it. That happens through lots of individual and group interactions, answering questions whenever we get them, in the threads or by email. That's what I spend most of my time doing.

We're never going to take the community's trust for granted because it's what gives HN the only real value it has, and it would be all too easy to lose. But I would tentatively say that this approach has proven to work well for most of the community. If people learn they can always get a question answered, that's a powerful trust-building factor.

Equally clear is that it does not work for everybody; but that's always going to be the case no matter what we do. I don't mean that we dismiss such users' concerns—quite the contrary, I make extra efforts to answer them. I'm just not under any illusion that we can satisfy everybody. It's satisfying enough if a few people can occasionally be won over in this way—which does happen sometimes!

People will game it. We don't need a transparent algorithm when we have transparent results, e.g. enable `showdead`, or the OP's project.

[flagged]

yes! good point. Edit: I changed my GP comment to say "link" instead of "ID".

No problem! I see these threads as opportunities to explain things to the entire community so I try to make the explanations as thorough as possible, and to answer every question that I see. (though I'm sure I don't see them all - if anyone has (or sees) a question that didn't get answered, you can always let me know at hn@ycombinator.com)

The url for this page is https://news.ycombinator.com/item?id=39230513 so the id is 39230513

I changed my comment to say 'link' instead of 'ID' so everyone can follow the same links.

Thanks _kst_: https://news.ycombinator.com/item?id=39232594

Presumably its the id parameter in the URL?

Most likely in the URL, id=3923

That would create a bureaucratic nightmare for no significant gain.

Edit: see https://news.ycombinator.com/item?id=39232795, which makes a similar point.

That would create one more thing for people here to complain about. People here would just accuse the mods of faking the mod log to hide their "real agenda" whatever that is.

Just enable showdead if you want to see all of that. It's 99% botspam.

Wikipedia can and does vaporize edits.

[flagged]

By "the last item" you're referring to "Avoid Async Rust at All Cost", right? Personally I don't think that's abuse; I would have flagged that post if I'd seen it. That's despite the fact that I agree with a lot of what's in the post. The title is just too inflammatory. And there are more inflammatory bits in the post, such as saying the feature is "objectively bad", and saying that a community member's post "gracefully omits" some information (where the word "gracefully" sounds like an accusation that they were being disingenuous). Totally unnecessary. Chop off the inflammatory bits and you'd have a perfectly good blog post making an interesting point, but as-is that post was not going to lead to a productive discussion.

I assume you mean this one:

Avoid Async Rust at All Cost - https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...

[4] You Don't Have to Be a Jerk to Succeed - https://news.ycombinator.com/item?id=35734086 - April 2023 (0 comments)

dang - thanks for taking the time to respond in detail. You really go above and beyond. I imagine this whole discussion landing like a concrete block on your plate (but hey, maybe you dig this part of the job).

I have/had no objection to the moderation on these posts. In fact, if I were monarch of HN and the Internet, I'd want an order of magnitude higher standard for the quality of posts, comments, and conduct. I want to spend my time and on the actual very best intellectual content and discussion possible - it would probably be mostly the very best books and papers from journals if I had my way. (Not that I think HN should serve my personal preference, I'm just demonstrating that I am far away from criticizing the moderation.)

My GP comment and my other one that you responded to [0] were trying to recenter at least part of the discusson on a factual basis, which I find much more interesting than the (completely unintersting) conspiracy theory aspect. That is, if we explore it factually, objectively, intelligently, how does it work? how does it work out? For example, I imagine there are some interesting emergent properties which would tell us about the HN population, emergent properties of algorithms, and the interaction between them.

[0] https://news.ycombinator.com/item?id=39231055

I think generally it works well- when there are actual major events like early COVID or Ukraine - HN managed to inform we way ahead of mass media with various interesting sources. But I’m happy to have a “news” thing pop up only a few times a year. You’re gonna have someone be mad about every instance when you moderate

>LLM stories

Does that mean stories about LLMs or by LLMs?

Serious question.

I am one of the (few? many?) people (devs) who haven't look into LLMs or even tried out ChatGPT yet :), except to make jokes about it here once in a while.

There ought to be a time-based flagging limit, so that people don't abuse the system. I've already raised this earlier.

If Company A makes a killer product announcement, rival Company B could simply get its employees to spam down votes on and flag that post. Company A gets less visibility, and dang won't be able to come on time to stop it.

This is an easily plausible hypothetical, which may already be happening.

I think you're probably generally correct, but "blaming the algorithm" sure smells to me like a whole lot of camouflage for censorship, which we ought to know by now has as much to do with 'quality' as it does 'shaping the narrative'

Generally speaking HN is a good site and a case study in successful community moderation, but you have to wonder 'who's watching the watchers' these days as the Overton window on free speech continues to be narrowed, almost entirely at the behest of big tech.

The second story is evidence of a new phenomenon: The far right political movements thinking about an anti-constitutional policy, a new step on the ladder of escalation.

There's a reason it's a big deal in German politics and already had some fallout (and thankfully multiple dozens of counter-demonstration of ten of thousands of people all over Germany.)

Users flagged them. We can only guess why users flag things, but in this case it might have been the desire to avoid gender flamewar hell, which is mostly always the same and which HN has had more than enough of. Also, one of the submissions was paywalled (https://journals.sagepub.com/doi/10.1177/00169862231175831, the paper, has been submitted yet. That one might work, if you or someone wants to try submitting it.

Comments on frontpage posts that go up quickly but then suddenly don't receive any votes in any direction makes me think you do freeze vote counts.

Yes, HN is a moderated/curated site and always has been. Here's 10 years' worth of me explaining that: https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...

What people maybe don't realize is how many constraints there are on HN's system. There aren't many degrees of freedom for us to change things that wouldn't lead to a massively different site, and most of those outcomes would be worse, because most of them would be closer to internet default.

It's easy imagine "HN, but without the things that I personally find annoying". But try to generalize that for a moment and the problem quickly becomes intractable.

The "force" is actually one or two people. It's hard to prove and impossible to change. No one will believe you, either.

I’m sure the HN codebase has some secret creed to make lisp more popular

Dupes generally drop off the front page, whether or not someone links up the previous stories in the thread. The whole point is not to let the duplicate story crowd out other stories on the front page. Redirecting would defeat the purpose.

We do redirect to earlier posts that had the exact same URL. But most dupes, in the sense of repeated posts about the same story, don't have the exact same URL. What do you do then? That's a hard problem. It would be great to have an automated solution to that but I don't know of one.

moderators are not omnipresent, and some times the users are faster to react than the mods.

Is that possible?

I like Sam, but "$celebrity says $thing about $common-topic" is almost never a good basis for a frontpage thread on HN.

It's vital to HN that user flags and/or software like the flamewar detector clear most such submissions off the front page. They tend to attract a lot of upvotes because that's what sensational (and especially indignant-sensational) stories do. Without countervailing mechanisms, HN would be completely taken over by those stories.

s/mods/users? everyone can flag stories.

[flagged]

> Mods can downrank a post so that it doesn't change anything for users, aside from the fact it won't be on the front page.

That's a big change?