Hacker News Reader

Comment by dang | Hacker News Reader

For the obvious reason: the community is divided on whether it's on topic, so some users keep submitting the stories and other users keep flagging them.

I don't know that there's much here that's intellectually interesting. This is probably one of those moments where HN can distinguish itself by not paying attention to a story.

Regarding priors, the argument "$X1 got attention so why shouldn't $X2" doesn't work on HN, because there's a power-law dropoff in interestingness along any predictable sequence. So if Jeff and MacKenzie got a lot of attention, that's actually a reason why Bill and Melinda wouldn't.

I'm not really sure that was the case to begin with though—these threads look pretty uninteresting: https://news.ycombinator.com/item?id=18868713, https://news.ycombinator.com/item?id=18865291

View on HN

I appreciate that this post was unflagged, after having been flagged. But it appears not to be listed on the front page anymore, or even on /ask. Is that on purpose? If something is not flagged and is popular, what would the reason be for hiding it?

People are not discussing the underlying article here — it is just a meta-discussion about what is allowed to be discussed on HN. It's a little surprising that this is being suppressed (either by community members or mods).

The [flagged] marker shows up on a post when there are enough flags to kill an item. Whether it's actually killed or not depends on that and a few other things, such as how active the discussion is. That's why https://news.ycombinator.com/item?id=27030269 shows [flagged] but not [dead].

But flags affect the ranking of a story long before they reach that fatal threshold. That's the case here. Moderators haven't done anything.

Thanks for the explanation. Is it only possible for people to [vouch] a submission once it is dead? If not, then it seems like posts can get caught in purgatory — not flagged enough to be dead (and therefore vouchable), but not alive enough to be seen anywhere.

In this case, it seems a reasonable compromise would be to keep the post on /ask, but let it fall off of the main front page. This is a respectful discussion about HN/moderation, and that seems appropriate for Ask.

That's accurate, except that upvotes also counteract flags, so posts don't necessarily get caught in purgatory.

Ok, it's on /ask now.

That explains why the story on Michael Lewis’ new book got sent to page 3. Seems to be controversial.

https://news.ycombinator.com/item?id=27030283

That one set off the flamewar detector.

Thank you for the excellent job you do here, Dan, and for taking the time to explain decisions. I know from moderating another, much smaller community, how difficult and thankless of a task it can often be. But thank you.

Is there somewhere that the inner workings of HN upvoting, flagging and moderation are detailed so that people can understand how it all (or most of it) works?

I understand that some things you'd want to keep mum about to avoid vote manipulation.

In any case, I hope you have a great week.

Using HN Search on comments with "by:dang" and relevant words will bring up lots of past explanations. There's no document that attempts to describe everything comprehensively because I don't know that that's really possible, and it would eat my soul to try.

Ahhh, I forgot about [dead]

I dunno, it seems like those that deem it purely personal and therefore uninteresting fail to appreciate the broader impact this can have. This kind of announcement isn’t for individual nobodies interested in gossip, it was an even-handed announcement designed to calm global markets, healthcare companies, and developing countries. The Gates Foundation has issued over 55 billion dollars in grants itself, and that doesn’t even cover all the other knock on effects. Not to mention the countless lives that have been saved by their direct help. This isn’t just gossip. But I realize that my thoughts have more to do with the substance/impact of the story, and you are doing a great job of explaining why the community has repeatedly flagged it. Thanks for the explanation.

While I agree it's probably uninteresting intellectually, it does have potential to be interesting emotionally. Not a lot of opportunities for HN to discuss topics such as divorce.

You're right that there are many kinds of interesting, but we mostly use that distinction to figure out what things aren't on topic. For example, celebrity gossip gratifies social curiosity (I say that without judgment—gossip is an essential human activity) but not on topic here.

HN gets plenty of opportunities to discuss universal topics like marriage/divorce, parenting, and so on. But because those topics are so generic and so emotional, as you say, they tend to slide quickly into the unsubstantive-sensational sort of discussion rather than the thoughtful-curious kind that we want.

We've learned over the years that discussion on such universal themes tends to be better when the initial submission has a lot of information in it, and also is alloyed with something unexpected rather than entirely generic. I don't think the current story has those qualities. In the first place, there's literally only one bit of information—already fully communicated by the title—so there's almost nothing to discuss; and in the second place, the story type is billionaire-blockbuster, which is about as sensational as these things get.

Previous comments on this point, if anyone cares:

https://news.ycombinator.com/item?id=8348372

https://news.ycombinator.com/item?id=23087737

There's also the old pg phrase "intensely but shallowly interesting", which I think probably describes this story, no?

The worst thing to post or upvote is something that's intensely but shallowly interesting: gossip about famous people, funny or cute pictures or videos, partisan political articles, etc. If you let that sort of thing onto a news site, it will push aside the deeply interesting stuff, which tends to be quieter.

https://news.ycombinator.com/newswelcome.html

I concur. Analyses might come along that will make this story a better fodder, but for now it's probably not substantiative enough.

I get wanting to keep HN from old-fashioned vulgarity and gossip, but I can't imagine finding it less-than-intellectually-interesting.

It's an improbably clean example of the conflict between modern and traditional religious views of marriage and money, and whether wanting divorce to be less traumatic can actually make it less traumatic. The symbolism is off the charts.

[Churches] all regard divorce as something like cutting up a body, as a kind of surgical operation. Some of them think the operation so violent that it cannot be done at all; others admit it as a desperate remedy in extreme cases. They are all agreed that it is more like having both your legs cut off than it is like dissolving a business partnership or even deserting a regiment.

What they all disagree with is the modern view that it is a simple readjustment of partners, to be made whenever people feel they are no longer in love with one another, or when either of them falls in love with someone else.

-- CS Lewis, Mere Christianity

You're making general points about divisive topics (marriage and religion in this case). The odds of interesting conversation on the open internet are low when the topics are intense like that. HN threads feel like an intimate conversation (something we all like about it), but they're really broadcasts to a large audience. On polarized topics, the odds that some message will land explosively in some portion of the statistical cloud are high. Some of those readers will rush into the comments to heatedly answer some provocation, and off we are to the flames.

We don't try to avoid divisive topics altogether, but we do to try to wait for solid opportunities - the ones where the odds are less skewed to flamewar. Threads are sensitive to initial conditions, the most important of which are the title and the article (in that order). If we start off with something substantive, where there's enough information for the mind to sink its teeth into [1], we're more likely to get reflective responses rather than reflexive ones [2]. In other words, it's best to play the good hands and fold the bad ones.

A blockbuster billionaire story with a sensational title and no additional information will probably produce discussion that burns hot, fast, and shallow. People will react reflexively with their pre-existing associations about the personalities involved and the generic topic (divorce, in this case). That is the sort of discussion that HN is not for. We want the kind that burns slower, goes deeper, and wends through unexpected places.

[1] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...

[2] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...

I think leaving one thread up, such as this one, would help the community discuss this story for a short time and move on. Instead half the new stories are this story at the moment.

Given the Gateses' roles in the tech world, venture funding, philanthropy, and Big Problems (scroll through BG's Twitter feed for numerous recent examples), including malaria, fresh water, sanitation, COVID-19, nuclear power, carbon emissions, global warming, disinformation, and more, the adjacency to numerous topics highly relevant to HN are high.

I say this as a very long-time critic of Microsoft and Gates, including on recent issues (patents as applied to vaccines).

Reader flags are a useful indicator, but can still reflect a minority quashing of a discussion that seems to badly want to happen.

Flagging also prevents tools such as Algolia from returning results of those discussions. That's sometimes merited, but not, I'd argue, in this case.

Sure, but what's there really to say about those things on the occasion of them announcing their divorce?

One obvious question would be whether high-profile marital status changes have been consequential, and how.

The case of Warrent Buffett. It was the death of his wife, and a sense of his own mortality, which inspired his own philanthropic bent.

There's the question of whether or not oligarchic philanthropy itself is a net good. There's much criticism of this, not the least of which comes from within the NGO sector. (See the Tiny Sparks podcast for significant discussion.)

Events are a foil. Life transitions become an opportunity to focus on what has been and what may be. Divorce isn't obituary, though it functions similarly, with the notable distinction that the principles aren't dead yet, and may yet do more.

(Uninformed speculation on why and wherefore does tend to be tedious, I'll grant that. I'm tugging out of that particular wallow, if possible.)

> there's a power-law dropoff in interestingness along any predictable sequence

brilliant quote. is there a name for this, or did you just make it up on the spot?

It's a point I've been trying to make for a while (https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...) but that particular wording just came to me.

Related concepts, if anyone's interested:

https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...

https://hn.algolia.com/?dateRange=all&page=0&prefix=false&so...

First application that comes to mind when trying to internalize this concept is drug usage.

Wouldn't marriage and estate planning be a serious consideration for tech founders?

The odds of a serious discussion of marriage and estate planning under these initial conditions are low. And the vast majority of HN users aren't tech founders.

More explanation here: https://news.ycombinator.com/item?id=27031522

I've seen you in threads defending the relevance of submissions. You even recently replied to me saying that: "That pattern still happens. At least I hope it does; we want HN to have weird articles." https://news.ycombinator.com/item?id=26814566

Maybe now is a good time to give the community the benefit of the doubt?

This story is the opposite of what I'd call weird. It's a major celebrity human-interest story, which must be all over the news.

And it's something I'd like to discuss with my fellow hacker news readers

Fine. Gates bad. Bezos good. I get it.

according to that link, the initial announcement is also [flagged] and has less points and comments than todays main thread about the Gates divorce announcement (which could comment in instead of complaing here). And the one post with >100 comments is as much about Bezos being even more rich as about the divorce.

Bezos is an active leader of a major company. Gates is a shareholder of a major company who has little if anything to do with it anymore. It’s not really comparable.

That doesn't follow!

Could you please stop posting flamewar comments? Between this and https://news.ycombinator.com/item?id=26939564 and https://news.ycombinator.com/item?id=26849805, you've been doing it way too much lately.

Reply on news.ycombinator.com