Facebook is obstructing our work on disinformation | Hacker News Reader

Facebook is obstructing our work on disinformation

theguardian.com

237 pointscharlysl90 comments

Many people here are conflating the research done through Ad Observer with the Cambridge Analytica fiasco. There is a huge difference in terms of user consent, and also the code review done by Mozilla on the open source Ad Observer plugin.

For a much better explanation of the distinct differences between these two situations, please see this EFF article: https://www.eff.org/deeplinks/2021/08/facebooks-attack-resea...

At the end, this is all about business and reputational risk arising from the research data being misused ala Cambridge Analytica. Risks are quantifiable based on past litigation history. So why can't FB require the researchers to provide indemnity secured by their college's fund or through some specialist insurer? If what they're doing is really above board and data collected is secure, the indemnity will never be called in and everyone can be happy.

If you're reading this and haven't deleted your account already, will you ever?

Probably not...

A lot of the people I know who still have an account, and are aware of Facebook's track record, use it for maintaining contact with friends and status updates, I don't think there is a whole lot that will get those people to drop, they would rather try to further close off the flow of information into Facebook, and they are already not trusting the service or what it does with the information it gets.

But they're still feeding the machine…

The presented line of reasoning makes no sense. But people aren't rational, that's no news.

You are saying that comfortable communication with friends is rationally less important than long-term ideological battle that you don’t really influence?

Yes. There are many options outside of Facebook, but compromising your morals sticks with you.

You don't need FB to comfortable communication with friends.

So there is no reason to use it.

If you know what FB is about, and still use it nonetheless (actually for no reason), that's highly irrational.

"I want to communicate with my friends" is just a very lame excuse for feeding the machine.

> You don't need FB to comfortable communication with friends.

You don't need planes to meet your distant relatives, but flying is often much more convenient than other modes of transportation.

> So there is no reason to use it.

Convenience is my reason. If you don't have reasons for yourself, that's fine, don't use it, but don't say for everyone.

> feeding the machine

This is a bit dramatic.

>You don't need planes to meet your distant relativesx

This is not a good parallel. You can use any modern instant messenger like Telegram to talk to your friends as comfortably (or more).

It is a good parallel. Obviously if your relatives live on another continent, flying is the optimal, but for example you can ride a bus from SF to LA instead of flying.

Bus takes much longer, whereas other messengers are practically as convenient as Facebook.

If you don't mind sharing your private conversations then sure it starts looking rational.

It is not just an ideological battle when your information is used directly against you personally. Companies pay for ads targeted to you by mining your personal content.

If you are not on facebook or rarely use it they have no opportunity to influence you.

You can use Facebook while not having any conversations on it you'd wish to stay private. I think most of my friends still on Facebook use it that way.

Facebook is a contact point, first and foremost. Nothing prevents you from greeting a friend and requesting that the conversation continues on a more private channel.

I have Facebook messenger app installed for the sake of a couple of people, and I check the Facebook main page once in a blue moon to respond to event invites from a friend.

Other than this I don't use Facebook at all any more and have used it very very little for the past several years.

I use a browser that tries to protect me from trackers and ads.

To me, this is what makes sense to do.

But they still get a significant amount of my attention because I use Instagram though.

The main issue here is that they also get to link all the other app events to your facebook account due to the facebook sdk and the fact you installed the app on the phone.

How is this not rational? Not feeding Facebook anything beyond the minimum (minimising anything extracted from the user) but keeping it open as a communications channel (maximising how much value the user is extracting) is exactly the rational strategy to adopt when network effect keeps Facebook as the #1 place people (re)connect.

I use facebook as an address book. When i need to get touch with someone I haven’t been talking recently, I open fb messenger and reach out. Then we can move it to a better form of communication. But it works. I don’t think I feed much info into it. Also fb market place has decent items.

"People aren't rational" is just a dishonest way of saying you disagree with someone's very logical reasoning due to different values.

I know Facebook is scum, but I have no other way to stay in touch with some of my IRL friends. So really I have competing agendas here, social connectivity vs. personal privacy, and I'm sacrificing some privacy to gain connectivity because I value one over the other. I, personally, mitigate the privacy loss by only sharing stuff I'd consider publishing publicly anyway, only liking stuff that's similar, avoiding groups and pages, periodically deleting all old content from my account, using adblockers and private browsing all the time, and customizing the FB web app via local scripts (thank you, FB Purity).

You can disagree with my values and the resulting choices (maybe you value personal privacy over social connectivity), but those choices are completely rational even for people who don't necessarily think about them in painful detail like I do.

So get over yourself and quit thinking that other people are somehow beneath you because they make different choices than you.

If only there was a way to keep in touch with people if you didn't have a Facebook account. I wonder what people did before Facebook?

I deleted my account at the end of July 2019.

Facebook did not become a better place through that. But at least I now procrastinate on HN instead. The population here is visibly smarter and the mods actually do their work to keep the standard from sliding into abyss.

> The population here is visibly smarter

You pick your Facebook friends, you subscribe groups which you prefer with mods you like, and you are free to ban whatever Facebook recommends you. My Facebook feed is interesting, I even click ads from time to time because ads are good.

Comparison to HN is not fair, because Facebook is about friends and HN is about strangers. Still I kinda like Facebook because on HN there's very narrow Overton window and interesting comments are often downvoted while on Facebook I'm exposed to very broad range of ideas from far left to far right.

If you're frequently on Facebook, you still have to deal with comments from friends of friends, if reading their posts - and that's where the frustration starts.

"You pick your Facebook friends"

True, but this picking is not completely voluntary. If you are on Facebook and your relatives/coworkers/colleagues are, too, they will feel offended if you reject their friend request and you will have some personal hassle in your meatspace. Happened more than once to me and I reacted by cowardly accepting everyone and then filtering out most of them. Was not proud.

In that sense, being in a community of complete strangers here in HN is actually refreshing.

What I really value on HN is the fact that a lot of comments contain valuable info or at least interesting thoughts. I often changed my perspective on something I previously considered obvious after reading a HN discussion.

This is rare on Facebook, most of the user-produced content is low-effort, shallow, information-poor and/or fishing for likes.

I hope for the day when there are more options than stay or leave, when users of a platform have more voice to determine the direction of the platform.

Leaving FB (and IG and WhatsApp and GIPHY) seems like moving out of a country and never seeing or talking to many of my friends again. I'd much prefer to stay in the country and have more power in determining the laws governing how we interact with each other.

Yeah. Especially seeing the effects the pandemic has had on peoples' social lives, and as such quality of life, I'm not going approach grownups who are unlikely to form new strong friendships, and tell them to cut off a major source of casual semi-social interaction with old buddies.

For context, I'm a man in my thirties, and I notice that a lot of my peers are super busy balancing work/life, or just bury themselves in work.

I'm from a Nordic culture where we don't have Southern European social customs and strong multi-generational families. I'm really worried about how our norms basically push us into loneliness outside our attempts to run nuclear families, which often fail and leave people divorced with limited additional social bonds.

All reasonable, but why does it have to be Facebook?

Social networks and communications platforms have value because there are people using them. I'm sure you know this, it's called the network effect.

Facebook's services are places where people in my demographic have accumulated social networks that sometimes forms connections in actual socializing.

When we throw parties with my best friends for a wider social circles, friends and acquaintances included, we still use Facebook events, to coordinate them, as an example.

Similarly, although I use Signal to talk to my closest friends, Facebook Messenger is still a functional way to get hold of people when you don't have their phone number or e-mail. Where I'm from, people my age didn't really collect phone numbers, since Facebook's convenient for that.

Transferring the connection between myself and these acquaintances may be cumbersome or awkward, and mind you, I'm the kind of person who insists on moving conversations to Signal after a certain point.

As a case in point, it took years to make Signal something that I actually use for real social connections outside my closest circle of people who humor my nerdy demands. And only succeeded thanks to 1) me having moved some group chats to Whatsapp for encryption the minute Whatsapp released a desktop app in May 2016. This moved the social graph to my phone book, which made it easy to switch to Signal over time. The initial move to Whatsapp worked because people already used it. 2) the larger network effect of Signal taking off, largely over the last year.

So, with all of this hassle in mind, I'm not going to tell people who, bluntly put, are at risk of being lonely, to drop useful social tools.

Thank you, I appreciate you going into some detail. It's frustrating to hear because it sounds like FB has now attached itself to a basic human need and can therefore never be removed -- but if you or those you refer to derive more benefit from that then the costs FB brings us then it's just another externality / tragedy of the commons.

I would not wish lonliness on anyone, though I hope in your example, FB is used to facilitate better social interaction, rather than subsume it entirely.

I really like the perspective you bring to this, especially this part:

> So, with all of this hassle in mind, I'm not going to tell people who, bluntly put, are at risk of being lonely, to drop useful social tools.

I think I'm playing a lot with the metaphor of Facebook or other internet properties as cities or countries—as locations where people exist and root. Similar to you, I have struggled to get people to leave the FB "city" or the iMessage "city" to follow me to the Signal "city." They may occasionally visit me on Signal but prefer to reside in iMessage or another place.

And then I feel bad asking them to move to Signal "city," as there aren't too many residents at the moment and not many whom they may know. So to take them to a sparsely populated foreign land where they don't really speak the language/dialect (different UI) can make me feel worried that it may disconnect them from the few connections they have back "home."

Anyways, thank you for sharing about your experience and inspiring me to explore this metaphor more deeply.

I feel a similar worry. I'm a man in my thirties as well, Midwestern American culture which has a similar focus on work and nuclear family, and yet my work for the last 9 years has been focused on emotional communication. I see lots of social pressure that if one has emotional intimacy anywhere, it's with the nuclear family, typically just with the romantic partner. Outside of that (or even in those relationships), it almost becomes taboo to express one's fuller range of emotions.

So yes, I wish we don't have to cut off our digital ties to others where we may still be maintaining some of those non-work, non-family relationships.

I did this year.

[Edit] though I had kept clear of it for 5 years.

Companies are not required to share data. Have you ever tried getting anonymized network traffic traces from a company for the purpose of analysis? There is a lot of open source data available (thanks crawdad.org!!!) but I doubt you will have a private company share data with you if you just go and ask them.

The same with Facebook. They're just paranoid that the data will make its way into places it shouldn't. Same with any other company.

No one is requiring Facebook share data, though it could easily be argued that they owe us.

It's users who install a browser extension that are sharing the ads they were shown.

If FB wants to serve people ads, why should they get to stop them sharing those ads with researchers / whoever they want?

They claim the FTC is forcing them to, even though the FTC has stated that's untrue.

This is fairly open and shut. Facebook are full of shit, quelle surprise.

I can't blame Facebook on this. Last time they allowed data for research purposes we got the Cambridge Analytica scandal, where that researches fed the data to this consulting company to be used for election purposes...

Why should these guys be trusted? "Trust me bro" --- that didn't work well before.

FB is right on this one.

Last time the data was accessed through an official API with (at best) dubious consent.

This time the data is voluntarily provided by using a client-side extension.

Seems quite different, doesn't it?

Let's say Facebook produces a newspaper with customized ads for each household. Should households be prevented from mailing their copy of the newspaper to researchers due to privacy concerns of the advertisers?

Should Facebook be allowed to stop sending newspaper to the researchers? Should Facebook be allowed to stop providing phone and mailing service for these researchers?

But they are not producing a newspaper with customized ads. They are showing you photos your friends took at the party last weekend. Photos that are shared with you and not with their boss.

If I am consenting to share with the researchers via Ad Observer the data that Facebook has shared with me, whatever that data is, great. I signed no NDA with Facebook.

Did all the people you have friended on Facebook also gave permission to give their data to the researchers?

I wish there was an official name for the fallacy of assuming that a given activity done at any scale are equivalent. You sharing screenshots from your Facebook and an add-on automatically slurping data from millions of people's Facebooks are not at all the same thing. They both might be okay to do but the reason for the latter is because it would be okay to do the former.

And not a single mention of the fine that Facebook got for giving access to this data for someone elses "work". The work that Guardian gleefuly led the assault on and also widely reported (with good reason! Cambrige Analytica was a huge privacy issue).

So not recognising the fine is disingenuous hypocritical crock and far from trustworthy reporting. Guardian, do better.

Expecting the Guardian to do better would be like expecting the NYT to do better- pretty pointless. They have specific agendas, and the stories are written to conform to the narrative.

As an aside, I don't use Facebook at all, and wouldn't trust them with my dog's name either.

As others here have pointed out, you are conflating two very different things.

Cambridge analytica deceived people to steal and sell data, and to invisibly manipulate them.

These NYU researchers are scientists, whose entire work is only valuable if put into the open, and who are boud do law and supervised by many checks and balances - soft and hard ones.

It must be possible to study platforms, and for people to volunteer their data with informed consent to science. Otherwise the entire mechanism of "self-observation" of society falls apart.

So yes, it would be good to mention the caveat that Facebook does need to protect users. But no, that's almost entirely not the issue here.

I’m not 100% familiar with the Cambridge Analytica situation, but wasn’t their tooling based around user data rather than just keeping track of ads?

From the article: > Facebook claimed that we were violating its terms of service, that we were compromising user privacy, and that it had no choice but to shut us down because of an agreement it has with the Federal Trade Commission. All of these claims are wrong. Ad Observer collects information only about advertisers, not about our volunteers or their friends, and the FTC has stated that our research does not violate its consent decree with Facebook.

If these NYU researchers have in fact created an identical system with identical strategies and goals, and are lying blatantly in the Guardian, that should be big news!

I’m somewhat skeptical about the equivalence here.

The threats posed by disinformation are too broad and numerous for these smaller academic organizations to handle. We need to consolidate our efforts under a single entity that will have more power to fight disinformation and protect truth. A "Ministry of Truth", if you will.

One of the frustrating things about political discourse is that, when conducted among a large enough audience, topics lose most of their nuance. Things tend to get reduced to A vs B. It's disappointing to see that when it comes to the subject of misinformation and propaganda in social media, for some people it seems, efforts to understand and reduce the spread of misinformation have become the enemy of freedom of speech and expression.

I don't understand why academics trying to understand the spread of misinformation on Facebook draws comparisons to the authoritarian government in 1984. Since when do we fear university researchers the same way we do oppressive governments? I get it, censorship is a problem, but seriously guys, "Ministry of Truth"? What the fuck.

Science is for everything that can be objectively measured, and politics is for everything else that people can't come to a consensus on. Bad things happen when you try to frame your subjective political opinions as objective scientific reality. "Misinformation" or not, you can't control what other people think, at least not within the confines of a free society.

So are we supposed to do, ban the study or research of topics that can't be objectively measured because the results might be interpreted as scientific reality by some people? The line between science and politics is not as clear cut as you might think. History and psychology can't be objectively measured. Anything, even hard science, can be politicized. The existence of climate change and COVID-19 are considered political opinions by a good chunk of the population.

Your premise that there are bad ideas that need to be studied and actively suppressed by authoritarian means is flawed.

You ignored my previous comment and replaced it with a straw man. No one is saying that the only way to combat bad ideas is through authoritarian means. That's a conclusion that you are drawing. How can you combat bad ideas in any way, if you don't first understand them, and how can you understand them if you can't study them?

And yes, I firmly believe there is such a thing as "bad ideas" in the sense that they are harmful. Can that be abused? Absolutely, but we can't over-correct and say that no idea should be labelled "bad". Take climate change. For years science told us it was coming. Misinformation about climate change has made it harder for us to do something about it, and now we're living the reality of extreme weather events. Why would we not want to know how that misinformation spreads, and how best to combat it?

It's not a strawman, because you said it yourself:

> Why would we not want to know how that misinformation spreads, and how best to combat it?

You don't get to suppress ideas in a free society, no matter how much you may disagree with them. You can discuss them and even try to win people over to your viewpoint through debate, but at the end of the day you can't force people to believe something without violating the social contract.

It absolutely is a straw man. There are many ways to combat something without resorting to —as you strawmanned their comment with–“authoritarian means.”

We combatted the ridiculous notion when masses of people thought they could cure diseases by burying a rotten fig under a dogwood tree on the 13th tuesday of the year if it was raining, and we did this by teaching people why that is ridiculous.

Understanding how and which absurd ideas enter people’s heads is incredibly important. And for many reasons that have nothing to do with “authoritarianism.”

The researchers we're talking about are in a spat with Facebook over the notion that Facebook isn't doing enough to censor ideas they disagree with. You're projecting your own bad faith arguments onto other people: we're not talking about burying figs under trees. We're talking about the kind of "misinformation" that includes the covid-19 lab leak or the current predident's son moonlighting for oil and gas companies in eastern Europe. You can't have institutions deliberately trying to narrow the window of acceptable discourse in a free society because there is nobody who will be able to watch the watchers make sure that they're not abusing their authority.

> I don't understand why academics trying to understand the spread of misinformation on Facebook draws comparisons to the authoritarian government in 1984.

Its because this "concern" over miss/disinformation is not isolated to academia, the ruling party has been pushing this narrative for some time now.

The ruling party has now aligned itself with the security state, the military industrial complex, and is now working on wrestling control of information away from big tech--all while expressing great concern over moms on Facebook getting their news from unauthorized sources as viewership of conventional mass media, which they have control over, continues to fall. I don't think it is unreasonable to view this as an authorization posture. Sure it's not Ministry of Truth status, but the comparison is there

Dude, what are you talking about. The ruling party? You mean Democrats, who have been held the presidency for less than a year? The ones that barely have a majority in congress, and the ones that appointed only 3 out of 6 judges in the Supreme Court?

What does it even mean that the "ruling party" is now aligned with the "military industrial complex"?

Democrats control conventional mass media? Because the behemoth that is Fox New's and the rest of Rupert Murdoch's media empire is, not mass media?

The democrats currently control two of the three branches of the federal government, the majority of Americans are democrats, establishment republicans are defecting a la Lincoln Project thanks to the fascist bent of the new Republican party, the democrats control the majority of the cities and the states with the most economic output and influence, and big tech workers and executives are overwhelming democrats. Sure it's not 100% but they are the ones in power rn and it has been growing for the last three decades despite setbacks at the federal level.

News Corp is the only major media corporation out of _six_ that is not aligned with the democrats. They happen to be the only one not pushing the disinformation narrative and are gaining viewers, not losing them like the others.

The military sided with the establishment, which is now embodied by the democrats as the republicans have become proto fascist populists, during January 6th. The military is now having hearings on equity and inclusion, requiring reading from liberal-co-oped leftist academics. The CIA recently put out that commercial trying to present themselves as holding progressive values. The democrats are talking about a new domestic war on terror and have acquired funding to expand the capitol police. The majority of officials in both the CIA and the FBI are democrats now. I could go on and on.

I'm not trying to say that all of this is bad or that the republicans are better, I am just pointing out that establishment power is aligning behind the democrats after the whole Trump fiasco and this whole dis/misinformation narrative has strong authoritarian undertones

Based on the examples they gave in the news report, it appears their motivation is partisan. So they intentionally pick one party line to conduct a “research” on “disinformation”. What exactly do they want to know from this “research”?

Yes, the Grauniad (Guardian misspelled - long running British insider joke) is left wing. The researchers writing the article are from New York University, and we are presuming the researchers are left wing too.

So what? Your implied alternatives are:

* Only allow politically neutral science (impossible for any subject containing politics)

* Only allow right wing science?

* Disallow all scientific study of politically charged topics?

What do you mean "allow" or "disallow"? It is common sense for scientific purpose, limiting your research subjects to one political group will lead to biased/distorted/misleading results, unless the purpose of the "research" is non-scientific but simply partisan hack.

So much of the discussion on disinformation is completely disingenuous and a distraction from the fact that Facebook and other social media magnates need to radically restructure their business practices and strip away the last 15 years of "innovation" in how our feeds are presented to us.

Return everyone's feed to a reverse-sorted chronological list of posts from friends only and so much of the misinformation problem goes away!

How about you just let me read whatever I want and come to my own conclusions about what to believe, thanks

Note that the FTC consent decree does not require blocking like this: https://www.ftc.gov/news-events/blogs/consumer-blog/2021/08/...

But they're enhancing other disinformation projects, so it balances out.

Of course they would fight it, disinformation is very profitable for them.

It's really odd that these researchers think they deserve access to this information. The successful will probably go to work at competitors to FB down the road. Such entitlement just because they give themselves the "research" title.

"Being entitled" is a weird way to say:

Researchers routinely work long hours

with little pay

out of altruistic motives

to the benefit of everyone.

Would you say doctors are entitled when they seek patients to study?

> out of altruistic motives

> to the benefit of everyone.

This is debatable or even blatantly false in many cases. Researchers are still human, and many of them are driven by greed and self-interest.

> The successful will probably go to work at competitors to FB down the road.

What do you base this on?

All of the indicated research interests and criticism points in the article are left-wing (fair enough), but then it follows that this may not be good faith research in the sense of being a neutral look at disinformation.

Facebook may well be underhand here, but the title claim seems to be disingenuous enough to undermine sympathy disinterested observers may have had with the researchers.

If in fact these researchers are in effect at least partially being seen as political campaigners themselves, then they are on shaky ground expecting special treatment being accorded due to an independent status claimed but not justified.

No. Seems like they mention that Bidens ads were not origin marked, why is partisanship so imporant in this issue?

> "it failed to include ads supporting Joe Biden ahead of the 2020 elections"

Thank you for that correction, I had misread the origin mark point.

> Using data collected through Ad Observer, and also data collected using the transparency tools Facebook makes available to researchers, we’ve been able to help the public understand how the platform fails to live up to its promises, and shed light on how it sells the attention of its users to advertisers.

So they exploited the data from the main source of revenue of a corporation and then to use that data to smear that same corporation and then they complain that the corporation do not want to cooperate anymore. This is a textbook definition of blind entitlement.

> We can’t let Facebook decide unilaterally who gets to study the company and what tools they can use. The stakes are too high. What happens on Facebook affects public trust in our elections, the course of the pandemic and the nature of social movements. We need the greater understanding that researchers, journalists and public scrutiny provide. If Facebook won’t allow this access voluntarily, then it’s time for lawmakers to require it.

Here's the problem the people who want to fight "misinformation" need to realize: they won't be in charge and the loyalty of those who gets to decide might change and those who gets to decide will not be the same people you would place your hopes in. I can delete my Facebook account but I can't delete the government.

Why should these guys be trusted? "Trust me bro" --- that didn't work well before.

FB is right on this one.

China also likes to detect and sensor all "disinformation". GTFO out here trying to research on what ads/subjects people are seeing so you can categorize them based on what your classification of that information is.

Entire community is in uproar on apple scanning your pictures but this "research" group gets a pass on monitoring what people are engaging in non-public internet FB groups/feeds?, and gets to classify if that information is "disinformation"?

I like reminding those "misinformation fighters" that they will never be in charge. The look of disenchantment in their face is usually quite telling.

You seem to have the wrong end of some sticks.

Users are installing a browser extension to specifically share the ads they were shown.

Facebook are claiming this is against an FTC agreement and so they are sadly forced to shut the researchers down - the FTC says that's bs.

But it is too high risk that these extensions will share something else like friend list or phone numbers. Even if extension authors have honest intentions (we cannot verify that), we cannot trust their account and the extension itself won't be hacked and extension replaced with something else.

And if that happens everybody will forget these newspaper analogies and will go straight to blaming Facebook. No, it is in everybody's best interest (except these researchers) not to have this extension.

You mean after FB has already shared your friends list and phone numbers? And tracked you and your friends, most if not all of who did not give you permission to give FB their private contact info, all over the web?

I see.

Since they have shared it with Cambridge Analytica, therefore, they must share with every single group that asks?

That just doesn't make sense.

I'm happy that FB is more stringent on this since guardian isn't exactly great either.

Not what I said. Did your friends and family give you specific permission to give FB their private contact info, which FB uses to associate private relationships whether they are on FB or not and track people across the web? Nothing to do with Cambridge or anyone else I didn't mention.

So, your comment has nothing to do with the current topic them?

Yeah, a lot of people give that permission. It is in the user agreement.

Not that I agree with wide user agreement, but I'm sure user agreement covers that part, arguably.

Yes, arguably, your friends are rightfully allowed to share your private info like phone numbers and photos. Because they have access to those information.

Not that I agree with it, it is just the current state we live in.

> I'm happy that FB is more stringent on this since guardian isn't exactly great either.

The Guardian has nothing to do with Ad Observer and the research it is used for. They simply ran the article written by the NYU researchers that Facebook banned.

Why did they choose to run it on a shady newspaper though?

There are at least 10 other more credible newspapers to choose from.

Reply on news.ycombinator.com