Every reply so far has missed the point of this sentence, so here it is:

Facebook reporting accurately is sane, but Facebook is in charge of what is being reported, which is not sane.

Like, imagine if your friend was slowly poisoning himself by flavoring his food with lead for a month. Even if he is honest with you about what he’s been doing, it’s still not a sane situation. If anything it’s less sane to be seemingly lucid and forthright about a month of self-destructive behavior, but not actually do anything to change it.

The question here is: why did Facebook allow the situation to get so bad that this page was the top for a whole month? Why didn’t someone at Facebook notice this stupid pointless scam page was surging and do even a little of research and tweaking to change that?

They have seemingly abandoning any responsibility for the behavior of the machine they built and operate. Personally I don’t give them many points for just reporting that.

Ex-FB here. My guess is that it's primarily a combination of two factors:

1) Someone / some org was slow to notice that this was happening. 2) Bureaucratic muck slowed resolution down. Even once escalated, there would be a lot of red tape to go through to make the necessary changes to how things are ranked. The people noticing the problem most likely aren't the people able to make changes, the changes would need to be tested at scale and their effects evaluated, and the people making the changes would have a lot of deliberating, politicking, explaining, and appeasing to do. Add onto that that everyone is stressed out of their minds, worried about performance reviews, and having to balance changes they want to make against how they and everyone else are evaluated at perf time...

Ok, but all that must also be true for the change that originally created this situation. There's all this red tape and large-scale testing and then... nobody actually looks at what hits the #1 spot?

Presumably there are people who are in charge of designing the ranking algorithm, so they would look at the #1 spot, but they would also look have to look at numbers 2 through 1 billion, and evaluate over long timescales and have a lot of reports, and compare how one algorithm performs vs another one.

Maybe they've already evaluated and decided that they're ok with #1 being off, if the majority of everything else performs better. Doesn't really make any difference per se what is #1. And then even if you want to stop X becoming #1, you'll never truly understand the myriad of butterfly chain events that caused us to get here, and how do you know even if you've solved this one issues, you haven't just ruined everything else.

Remember spam is an adversarial relationship. Almost certainly when whatever ranking change was exploited here was deployed the top ranked items were great, then some actor started realizing what was being selected for and exploited it

This, and the comment above about 2 through a billion, seem pretty spot on from what I know from talking to people still working there.

Good lord. I'll never invest a single dollar in Facebook after reading this. Talk about dysfunction

> Even once escalated, there would be a lot of red tape to go through to make the necessary changes to how things are ranked.

When I write a program that accidentally gradually fills a disk with garbage, I don't wait until I can redesign the program before I delete the garbage.

Yea, but does the garbage earn you money on ad views?

It's more complicated than that. Facebook is a company. People in different parts of the company do different things, have different motivations, etc. One cannot look at a large organization as an individual. I tried to come up with a more apt analogy... but analogies are awful here.

So, probably lots of people actually did notice it, and either (a) shrugged and thought "not my job", or (b) notified someone else about it, who then did (a)?

There's also likely a (c), which is people who don't think that Facebook should do anything about it. I guess that could be considered part of a set with (a), but I thought it should be mentioned separately as it's a different motivation.

Might I suggest (d): analysts, disgusted by the fact this crap is on top, tweaked The Feed in an attempt to disfavor that type of content then ran a simulation but saw their simulated ad revenue become lower than allowed minimum, then did (a) and went out to have lunch.

“It is difficult to get a man to understand something, when his salary depends on his not understanding it.”

- Upton Sinclair

In particular, stare too hard at that viral page, realize how scummy that page is, and then spend the rest of a career at facebook trying to tear down the systems that made that the most viral page, are likely going to earn you enemies, and cause the company to lose money (in the short term, at least).

Did you stare at that page on Facebook? The top post has millions of comments.

Which is totally fair in my opinion. It is no single individual's job to fix an issue this big and I don't think it's even feasible for a single person to do at a company like this. And if management hasn't set up incentives for people to proactively fix issues like this then they probably won't get fixed. Upper management can and should fix these issues but they probably don't care because they don't think it's an issue.

If it were a tiny thing that slipped under the radar, I'd buy it, but if my company were being dinged in the media every day for spreading misinformation, I would make sure the whole company is hyperaware of how misinformation spreads... which is often through viral posts and pages. I don't think this sort of oversight is something to blame on corporate bureaucracy.

maybe these junk posts are all that’s left to boost after tamping down the obvious misinformation and political propaganda

It doesn't seem like facebook pumped the whole page, all other posts have extremely low engagement, literally < 10 likes. What's wrong with pumping individual "good" posts? If anything it can be seen as honorable/egalitarian giving folks a chance to make it even if they aren't some media behemoth.

This is the central tragedy of Facebook.

They set out to build a platform where anyone, no matter how small, could compete with big media on a level playing field based on a straightforward metric: whether a user cares enough about a post that they're willing to engage with it. Publishers get rewarded for producing compelling content, users get to see more content they care about, and Facebook makes money since people spend more time on the site. Everybody wins.

And they succeeded wildly at this. But as it turns out, the things people respond to most are spam and hate and conspiracy theories.

I'm really not sure if that was ever what Facebook set out to do.

My mental construct of Facebook is analogous to Morty following the path defined by the death crystal but towards the end goal of 'the most money'.

Mine is Universal Paperclips, except it makes engagement statistics instead of paperclips.

The scenario describes an advanced artificial intelligence tasked with manufacturing paperclips. If such a machine were not programmed to value human life, then given enough power over its environment, it would try to turn all matter in the universe, including human beings, into either paperclips or machines which manufacture paperclips.[0]

Replace "paperclips" with "clickbait" and "human life" with "the right to privacy". "Artificial intelligence" then becomes Mark Zuckerborg.

[0]: https://en.wikipedia.org/wiki/Instrumental_convergence#Paper...

Collect data

And collecting data leads to what in this day and age?

Collecting money.

I enjoy this author's podcasts and newsletters, and an interesting angle they've covered recently is that Facebook, ostensibly trying to "clean up their act" with regard to being seen as a platform for disinformation and radicalization, has been tweaking their algorithm to not promote so much alt-right and similar content.

But the problem is, once they exclude all that stuff, there's just not a lot left. So you end up with inane garbage like this account being the most popular content on the platform. Like the whole platform is just Ben Shapiro and drop-shipping scammers. Really paints an "emperor has no clothes" picture of what's going on on Facebook these days.

You just don't understand the Metav3rse.

When corporations hide facts we call them immoral, and when they publish the facts we are scratching our heads?

Facebook does a lot of internal research which to me seems to be coming from a good place. Doing that research carries a lot of risk for them; e.g. the recent 'whistleblowers' mainly leaked that exact research reports. But instead of giving Facebook credit for even doing that, we always take a negative approach.

Note: I don't like facebook, I don't even have a facebook account, but unless we encourage good behavior (even at places we dont' like) how do we expect things to get better?

Yeah I also find it surprising that people are shocked that Facebook have internal studies into e.g. whether instagram causes depression in teenage girls (answer: inconclusive), like they’re a bad company for investigating whether their products have negative effects on people…

Its more that people are shocked* that facebook communicated to the outside that such problems were nonexistent, despite knowing that this was a lie.

*not sure wether shocked is the right term to begin with, it seemed more like indignation to me.

Did they actually know it was a lie? E.g. https://www.nytimes.com/2021/10/10/opinion/instagram-faceboo...

Thanks for the article, it was an interesting read, since I was mostly going of the WSJ stuff before. I nonetheless feel that with social research in the private sector absolute certainty or even studies that would pass peer review are a bit too high of a bar to meet before we can talk about the organisation knowing something. But I guess thats more of a semantic argument.

After reading the NYT article it still seems to me that a more ethical company would have done more to address the probable issues. Meta did very little and seemed to cherrypick the most favourable public studies. Anyyway, all this doesn't seem suprising or even unusual to me. I generally don't think that we should expect companies to act very ethical. Rather, incentive structures should be created/amended in a way so that purely self interested companies act as if they were ethical.

Thank you for this article. I was beginning to lose hope that there was anyone with a measured view on this issue.

The shock is that someone at Meta looked into these studies and decided not to take actions to fix it because it might hurt their revenue.

We are told again and again that correlation is not causation, but we readily ignore this maxim when we are looking for an account that we hope is true. At a time when Facebook is regularly vilified (sometimes deservedly), wanting to believe that its practices have caused teenagers’ mental health to suffer is understandable. But wanting doesn’t make it so.

[0]: https://www.nytimes.com/2021/10/10/opinion/instagram-faceboo...

> We are told again and again that correlation is not causation

I'm beginning to suspect that maybe we're told this too often, so we're starting to take it for granted that correlation can't be causation. But the saying actually only means that correlation is not necessarily causation. Quite often, correlation is actually there because of causation.

The shock comes because it doesn’t fit the narrative that Facebook is all evil, all the time. Nuance, complexity, and evenness are lost on some. No one and nothing is all bad all the time.

I think the shock is how old the reports are, and how since then things have not changed, but in fact worsened.

Didn't their own research show that it made 13% of teenage girls who use the platform seriously depressed?

The reporting on this study has been terrible. It was a really small study of female teenage Instagram users who reported having various problems. Of them, 13% blamed Instagram for making things worse, which was actually less than the number that said Instagram made things better. And this is all based on self-reported data anyway, so none of us should be taking it very seriously.

Just because you were honest with your partner about cheating on them doesn't mean they have to stay with you. That isn't a mixed message about whether your partner wants to hear the truth or not. You're not being punished for telling the truth about what you did, you're being punished for what you did.

> When corporations hide facts we call them immoral

Not really.

When corporations hide externalities that negatively impact the entire population we call them immoral.

I mean, nobody has ever called KFC immoral for having "secret spices."

I think you are just missing the backstory. Of course if you are approaching what Facebook does with naivety you end up sounding.. naive.

The whole reason this report was put out by Facebook in the first place is because people used their own tool Crowdtangle to point out how the most popular pages of all were pushing political and medical misinformation to billions and were run by foreign spam mills. So they gutted Crowdtangle and published this report cleansed of any reference to "questionable" pages. But as a previous submission [1] showed, the numbers they state don't add up. And as this submission shows, if you dig into what Facebook wants to tell you is popular, it's all still the most garbage, devoid of value scam content - and it is their algorithm that is promoting it above all.

1: https://ethanzuckerman.com/2021/08/18/facebooks-new-transpar...

In hindsight, I can see this quote is generating a lot of discussion about the content reporting, but I intended to focus the latter part of the sentence, which I thought really highlights the ridiculousness of social media.

Well, perhaps whoever generated this report was simply reporting the truth and not trying to fabricate or distort the truth in order to make Facebook look better?

The fact that the author of the linked article apparently just presumes that they would do so and "[can't] figure out" why they wouldn't says a lot about their own integrity.

I don't think it says anything about their own integrity. I think it's just a tongue-in-cheek remark about how this information makes Facebook look ridiculous.

Burying the lead, to be sure.

They're not paid to care.

Even worse: They’re paid to not care.

When conducting scientific research, I'd think "not caring" is a desirable trait. You don't want anyone's personal feelings to get in the way of reporting accurate information.

I think this is just being pedantic. You have to care about something to formulate and test a hypothesis. You have to care about what the data say to bother gathering them and collating them into some report in the first place. "To care" can be "to be curious about something" or "to have an interest in something", not just "to be concerned about something."

No, you have to be interested in something. Caring implies that you have an emotional desire for the hypothesis to be either true or false and that leads to bias. Scientific interest drives motivation.