Comment by dang | Hacker News Reader

Comment by dang | Hacker News Reader

This article got a lot of user flags. That's bad—it's obviously a substantive article, squarely on topic for HN and not sensational, ideological, or predictable. Even a quick skim is enough to see that it doesn't break down along any of the obvious faultlines for this general topic. That not only means it shouldn't have been flagged, it makes it HN gold. If you've been wielding the flag club that indiscriminately...please stop! Accounts that do that lose their flagging rights.

It's not legit to flag an article just because you don't like the general category. There must also be something about the specific article itself that's demonstrably bad for HN (for example, maybe the article doesn't contain substantial new information, is gratuitously baity, or is mostly fuel for internet drama). None of that is the case here—quite the opposite. This is just the sort of article we should welcome, so we can have substantive discussions about relevant issues instead of hurling the same old nasty platitudes at each other yet again. (Can a platitude be nasty? or do I need a different word?)

A thousand thank-yous to the user who took the time to email about this while the thread was still fresh; I wasn't focused on the HN front page just then, but I got the heads-up. If you ever want to make a big difference to HN quality, emailing hn@ycombinator.com when you notice solid, substantive content not getting the attention it deserves (whether because of flags or for any other reason) is one of the highest-leverage contributions you can make!

hey dang, I'm the author. Thanks for cleaning everything up - I'm sometimes a bit nervous when my work gets posted on HN for what happened here. I really appreciate you & the HN team being so proactive in fixing it!

Absolute gold.

Especially for those of us where top software can be very critical, but who are not in the software business and are committed to efforts in other areas of technology.

I can only imagine the challenges you have been facing are on the increase.

I don't understand this comment. Below we see plenty of dead comments posted at an earlier time, which seems to illustrate some existence of fuel for internet drama.

At this time the first 50% of the comment page does not discuss gender so it clearly turned out right, but they were posted after the dead comments. If a fresh article has the majority of comments being people warring about gender then the option given to users is either to ignore the war, flag it, or engage in the discussion. Having seen that multiple times it is often easier and uses less time and energy to simply flag the article regardless of how great the article actually is. The flag would not be a comment about the quality of the article, but rather the climate and culture for which it was posted in.

Dang, is this a problem?

There are separate flags for comments and articles. At the time of erroneous article flagging, there were worthwhile comments on this article.

The important question is when the flagging occurred by the users. If the flagging occured when most comments were culture warring, and the moderators find that to be inappropriate, then it would be good to know for future reference. If most users pressed the flag button long after the discussion switched to be more productive then I can understand what dang is talking about, but from his post I can't determine that.

Is there a documented HN policy that high-quality articles can be censored on HN because of low-quality comments? Why not allow time for low-quality comments to be downvoted/flagged and high-quality comments to arrive? If good articles could be rapidly flagged/censored, how could discussions ever "switch to be more productive"?

Yes, because if an article is particularly good, it's not fair for disgruntled users to be able to sink it by posting shitty comments. That would be a big loophole.

If the article were garden-variety flamebait then it would be a different story (no pun intended) – but this was such a clear case of the opposite.

The best thing is to give us a heads-up at hn@ycombinator.com. That could be either "hey, there's a really good article getting unfairly flagged" (a user did that last night and it was extremely helpful) or it could be "hey, there's a really good article getting a lot of flamewar comments". Actually it could also simply be "hey, there's a really good article that hasn't gotten any attention yet". (People often send those for their own articles, but that's a little less noble and not what I'm talking about here.)

You can see the common denominator there: "a really good article". There aren't very many of those. Just as the most important thing for startup investors is not to miss the best startups, the most important thing for HN is not to miss the best articles.

Thanks for answering. Not being able to use the flag feature when a article is being used as fuel for culture war is an answer to the question I had. Emailing sounds like a useful feature through I doubt I will be using it myself. What I find as being an really good article getting unfairly flagged might be very different from others.

A standard version that I often see is when articles get sunk by shitty comments are when the author is controversial. The negative comments are focusing on the author, and often that dominated the article comments until the article get flagged to death. Fair? No. It is however a pattern that keep repeating itself.

Yes, that's a problem. It requires human intervention to fix it, and that inevitably includes interpretation and judgment calls and no doubt bias too. We try to be as unprejudiced as we can and stick to HN's first principles, but people have such vastly different perspectives that satisfying everyone is out of the question. Still, it's not entirely subjective, and one's ability to stick to first principles gets better with practice.

I don't recognize this bit though: "Not being able to use the flag feature when a article is being used as fuel for culture war". That doesn't sound like what I said and certainly isn't what I meant. Edit: oh, I see what you mean - "being used as fuel for culture war" is your rephrasing of what I described as "shitty comments". Don't take this absolutely, though. The median case is that both the article and the comments are garden-variety culture war; those are legit to flag. I'm just saying that you shouldn't flag indiscriminately. We want some discernment when an article clears, let's say, the 95th percentile of thoughtfulness.

> Can a platitude be nasty? or do I need a different word?

It works. Nasty banalities could be used too. Slightly different meaning but both reflect what is happening.

How can you tell how many flags a submission has received?

That information isn't public. I'm a mod here (in case that wasn't clear), so I looked at the private data.

Edit: if you meant what you can tell from public information: (1) sudden drops in rank, as walterbell points out downthread—(edit: but it's important to know that sudden drops in rank can also be because of the flamewar detector, which is unrelated to flags), and/or (2) the [flagged] annotation, which appears when flags outweigh upvotes by a certain threshold.

I assume they were asking how does a non-mod tell if it's flagged so they can report it to you? (aka when it says [flagged] next to it or something, or drops off the front page rather quickly, etc)

Possibly from abrupt drop in rank?

http://hnrankings.info/27307977/

Thanks! Added above.

Oh I see. It seems that it would be difficult to report a big drop in rank unless you’re frequenting tho.

I guess I should turn on the show dead to see flagged posts

[flagged] and [dead] aren't the same, but if enough flags occur relative to upvotes, the thread will eventually also become [dead] - however, not if there's an active discussion already. So 'showdead' might not do much for you in this specific situation.

Reply on news.ycombinator.com