Hacker News Reader

Comment by twblalock | Hacker News Reader

At Facebook's scale, there is not a good way to do it. Even if they were able to scale it, it would not be good enough anyway.

You can automate it, but clearly that doesn't catch everything and automated systems can result in just as much bias as humans doing the job.

You can hire humans to do it and make them work fast, because the amount of posts per day is extremely high. You can use some automation to help them, but that will combine the problems of human moderation and automated moderation.

You can hire lots of humans to work at a normal pace and give them psychological counseling, and spend so much on their salaries that you'll go out of business, and even then they won't catch everything and will be frequently accused of bias.

It's a no-win situation for Facebook because even if they were able to objectively and fairly moderate every single post, they would be accused of bias by politically motivated people anyway.

This is not a technology problem, it's a social and political problem and Facebook will never be able to please everyone. Twitter is in the same boat.

View on HN

> You can hire lots of humans to work at a normal pace and give them psychological counseling, and spend so much on their salaries that you'll go out of business, and even then they won't catch everything and will be frequently accused of bias.

Then you should not fucking be in the business at all! What FB does is externalizing the cost onto societies and skimming the profit!

Cost here being two things: first the cost of mentally fucking up the moderators (so an extensive amount of psychological healthcare has to be shouldered by society) and second the cost to society that arises from letting Nazis and jihadists spread their propaganda (including the cost from real-world violence enabled by said propaganda).

Any online open forum would have the same problems no matter how hard they tried to prevent it. We can either accept that some level of hate speech is inevitably going to slip past the moderators, or we can give up on having open forums altogether. There is no perfect world with perfect moderation of online content.

I wish Facebook had the courage to say this themselves.

Is that provably inevitable for all online forms, or is “nobody knows how to yet”?

Governments act like all tech problems are the latter, but I would like to know if this is like encryption that only friends can break (never), or like self driving cars (most people call it impossible until five years after it’s been demonstrated).

For me moderating what amounts to open communication platform falls into third category: it is impossible and you should not try to do that.

Especially in FB's case where it is quite obvious that various opposing "hate groups" have learned how to game the system, and also how to present results of that such that it supports their rhetoric.

> it is impossible and you should not try to do that.

No, it is merely extremely expensive. But that's something that you need to tackle head on when your business model is to provide a service for free to a large chunk of the planet's population. The 'free' bit doesn't absolve you from your responsibility as an operator.

Seems like their recent move to get news off the feeds and focus on family and friends should solve the content issue. If someone is posting or liking hate speech or offensive content, I can just block them. I wouldn't add someone as a Facebook friend that would do that kind of thing anyway.

> it is impossible and you should not try to do that.

Can you prove, or at least demonstrate strong evidence for, either of those two claims? “FB are noobs at propaganda” isn’t sufficient, IMO.

The former, because words are subjective, therefore this is no way to get everything anyone anywhere would ever consider hate speech. It's just an impossible scenario.

> Any online open forum would have the same problems no matter how hard they tried to prevent it.

Facebook tries to get by with as few moderators as they can, with paying them as low as possible and generally trying to teflon away anything negative from their service (i.e. they negate being responsible for the fake news/hate speech on their platform).

I believe that there would certainly be a way to provide fair moderation - for example a ratio of 1 full time CSR (customer services representative) per 1000 users, not the 1 per 20k or worse they are currently trying to do.

> a ratio of 1 full time CSR (customer services representative) per 1000 users

That would require 2 million full-time CSRs, because Facebook has a little over 2 billion active users per month.

If that is the price to avoid externalizing the cost of a lack of moderation to society, then it is 2 million CSRs.

I'm sick and tired of Silicon Valley companies thinking they can save money on customer support and have societies pick up the tab.

And it's not like Facebook (or Twitter) couldn't get the money: 3€ per user and month should be enough to cover the cost of a properly salaried CSR - and at the same time create a barrier for bots and Nazi trolls.

> And it's not like Facebook (or Twitter) couldn't get the money

What makes you say that ? Facebook barely makes $5/quarter (3 months) in revenue. They couldn't even afford $1/month/user. Meanwhile Twitter has never turned a profit ever.

If you can't make more than $5 per quarter per user, maybe you shouldn't be in business. If you have never, ever turned a profit, then what are you, a charity?

Nope, you’re a venture backed startup!

That would make them the world's second largest private employer after Walmart. It's an absurd number of people to hire for content moderation. And it likely wouldn't solve any of the world's problems.

We could always make it law and prohibit Facebook’s continued existence.

Similar to how GDPR is going to gut the business model of firms who suck up all personal data and sell to the highest bidder.

If you're tired of "picking up the tab" you should leave facebook and encourage others to do the same. What ever happened to the idea that people who say stupid things expose themselves.. it's kind of a favor

At a cost of $20,000 per CSR that is $40 billion a year, which is more than the revenue of Facebook.

If Facebook can't moderate properly because it is too expensive, then Facebook can't exist. We can go back to a more plural world of many specialist forums with community moderation. That way the content properly reflects the society that generates it.

Then leave Facebook and encourage others to do the same, rather than forcing your views on them.

Forcing my views? What if they were spreading porn to children, should society force it's views on them then? What if they were subverting elections? 'Forcing your views' is kind of how society works. Facebook is not a harmless forum where friends can adopt banter at the level they feel comfortable with. Facebook put things in your timeline that are unsolicited because they are a publisher and should follow the same laws as other publishers.

Externalizing the cost of society's speech to society?

How dare they?!

In a liberal society speech should be free in every way.

How is it different from companies building products that pollute the earth, and not cleaning it up? Food companies not worried about the impact of huge life stock, or burning down forest for agriculture, or car companies not worried about air pollution created by their cars?

> I'm sick and tired of Silicon Valley companies thinking they can save money on customer support and have societies pick up the tab.

Unless they're obliged to by law, they never will. These companies minimize customer support in order to improve the bottom line, and will keep these at a minimal level - just barely enough to get away with that. It's one of the elements of the so-called "disruption".

If Facebook can't work economically in a way that's also somewhat responsible socially, maybe it shouldn't have 2 billion active users per month. Maybe it shouldn't survive. Maybe I should finally quit it.

>Maybe I should finally quit it.

I did. You should, it will improve your peace of mind. Also Facebook is a culture swallowing monster.

I seem to recall in SVs earlier days, when tech started putting old business models out to pasture, the following phrase was thrown about quite a bit:

> No one owes you a business model

I don’t reslly care if FB doesn’t work at scale, if we put minimum responsibilities on them. They’re not owed that business model.

Well then I guess they took a bite larger than they could swallow, right?

It's pretty understandable they're not able to moderate over 2 billion active users per month.

So that leaves two options: Either not have 2 billion active users, or not have any moderation. The latter is plainly unacceptable because, like others pointed out, it simply externalises the cost onto society. The first option, well I'm sure people will find other ways to communicate and plan events; It's not like their service is so unique, just the network effect and the size (which is way too big any way). We were long overdue for something better to come along any way.

There's no fundamental "right" Facebook has to do this, to keep existing and growing in the way they do, if all they want to do is grow but not take responsibility with the consequences of their size. Edit someone else said it better: No one owes you a business model. If a crucial bit appears impossible to "scale", then well tough luck I guess.

>for example a ratio of 1 full time CSR (customer services representative) per 1000 users, not the 1 per 20k or worse they are currently trying to do

Hey, that's the Mastodon model! Relatively small independent servers with own admin/moderator teams responsible only for their own servers. That's scalable moderation.

Aren't those more like a maintenance team? While they are moderators, keep the server running and use it themselves, they don't specifically responsible for the posts on their server.

Mastodon has moderation tools[1] that allow administrators of a server to restrict not just users on their own server, but also how other users in the Fediverse can interact with their users. Users can also report content.

[1]: https://github.com/tootsuite/documentation/blob/7f81350fe4fe...

> they negate being responsible for the fake news/hate speech on their platform

Are they, though? I feel that they are no more responsible than internet providers are for the illegal shit people do with their connections.

I'm absolutely gobsmacked that people are starting from the position that hate speech should be censored at all. The determination of what constitutes 'hate speech' is made by whoever happens to be in power at the time. Winston Churchill would have given a much different answer than Adolf Hitler. From my point of view, it's chilling that the mechanism to do this has been built in the first place.

This is a common approach by groups who perpetuate hate speech- to exclaim with faux outrage that any limits on hate speech are tantamount to the dissolution of free speech at the behest of ostensibly powerful minority groups. I won't go so far as to accuse you of being a member of hate groups, but your response certainly echoes the propaganda that they spread, especially in forums like HN.

The fact is that society has always drawn lines about what speech and behavior is and isn't acceptable in different venues and circumstances. Facebook's choice to censor hate speech is no different from a bar, restaurant, or department store asking someone to leave for shouting the N-word at fellow shoppers. It's a private non-governmental entity making a choice about how they want their users to act on their platform.

Furthermore, free speech claims in favor of hate speech ignore the real material costs in human lives that facilitating hate speech incurs. While I greatly value free speech personally, I also greatly value human lives and the ability of all people to meaningfully participate in society without facing systemic oppression, violence, and hatred.

Tolerance must inevitably come face-to-face with intolerance, and if we don't act to stop the worst kinds of hatred and intolerance they fester and decay the ideals that allow tolerance to flourish at all.

You’re drawing a false equivalency. Social media platforms like Facebook and Twitter are more like a public squares rather than restaurants or stores. It’s a space where free speech should be defended. Because free speech protects our society from authoritarianism and violence. It’s authoritarians who want to limit free speech, because their ideas have no merit and can only be enforced by violence.

> It’s authoritarians who want to limit free speech, because their ideas have no merit and can only be enforced by violence.

Let's not be disingenuous: democracies also want to limit the speech of authoritarians. And they probably should, because every idea that's broadcast will garner some following. It's human nature. Your knife doesn't only cut one way.

Good people want to limit the speech of bad people in principle, sure, just as good people would endorse violence if there were a way to ensure it were directed only against bad people. But no-one can fairly judge that, so limiting speech is a bad idea in the same way that permitting violence is a bad idea. The marketplace of ideas works slowly, but it does work: it does, ultimately, find the truth; good ideas succeed while bad ideas die out. Whereas the "marketplace" of violence can't tell whether an idea is good or bad, and while the violent people might be on the side of good for now, there's no reliable way to keep it that way.

> Social media platforms like Facebook and Twitter are more like a public squares rather than restaurants or stores.

Are they? Let's see the most important difference:

Is a public square privately owned? No. It is owned by the community, usually the council.

Is a restaurant privately owned? Yes. The proprietor has the option to evict people at will, as long as they do not discriminate against minorities in the process.

Do hate speech laws apply in a public place? Yes. If someone states racist things in public, they may be put in prison for hate speech.

Do hate speech laws apply in a private place? It depends on if this space is open to the public, and on the event being held. Generally, yes.

Is the owner allowed to evict any member at will from a public square? It depends. If the person is being harmful towards others, they may be removed from the square and detained, and sometimes even charged a crime.

Is the owner allowed to evict any member at will from a private square? Yes, as long as they do not discriminate against minorities when doing so.

As you can see, the attribute you have brought up does not matter. Facebook has the legal right to remove people for speech, as long as it does not discriminate against minorities in the process.

> Is a public square privately owned? No. It is owned by the community, usually the council.

Many public squares are privately owned these days.

> Facebook has the legal right to remove people for speech

No-one is disputing that. Grandparent wrote "It’s a space where free speech should be defended," not "It's a space where free speech is protected by the law." The lack of legal protection for speech in this kind of space makes it all the more important that individuals stand up for it.

I'm not even sure there's a real objective definition of "hate speech" - there's certainly speech you hate. And speech I hate. But its all rather subjective. Best we can say is that some speech grossly offends a shared, widespread, moral consensus, and so seems hateful from that perspective. But... lots of genuinely progressive speech is also this way.

Free speech policies that would effectively harbor hate speech - also protect your speech from being labelled as such, and then censored, or worse. There are a disturbing number of places still left in this world, where controversial speech can get you jailed, or killed - with the blessing of, or by a government.

Humans can't be trusted, by and large, to be benevolent censors.

> I'm not even sure there's a real objective definition of "hate speech"

Speech promoting violence against some group of people, and not for defense purposes. No subjectivity there.

So "punch a Nazi" should be censored, right?

The argument is that since the Nazi will inevitably punch you, preemptively punching the Nazi is inherently defensive, and thus urging people to punch Nazis is not hate speech.

Of course, you may very well ask, isn't it trivial to construct an almost identical argument to justify violence against almost any particular category of people you'd care to name? Hopefully your interlocutor won't respond by punching you in the face.

Yes? Righteous hate is still hate, even if enjoyable.

Absolutely. There is no excuse for violence except defense. The other comment talks about "proactive defense", but that too seems dubious.

There's not a lot of value talking about free speech on HN because it itself is subject to even stricter censorship of unpopular ideas. It's really a bubble of agreement and almost-agreement.

Suggesting that free speech advocacy, which has a very long tradition that predates contemporary social divisions, "echoes" the "propaganda" of "hate groups" is simultaneously meaningless and pernicious. You're drawing a line between free speech and "hate groups" while in the same breath denying that you're even doing so. In the process, you're not only inventing a nefarious association where none exists --- because causality flows forward through time, not backward --- but also completely ignoring danjayh's central point, which is that the definition of "hate" depends on the objectives of those with the power to define the word, and that machinery to suppress "hate speech" becomes, ultimately, a vehicle for reinforcing existing power structures and delaying needed change.

> Furthermore, free speech claims in favor of hate speech ignore the real material costs in human lives that facilitating hate speech incurs.

What material cost? Censorship advocates continually assume that hate speech must have large and personal costs, but in my experience, present no evidence. The benefits of censorship are not apparent. The costs, however, are clear in the historical record: slowed scientific progress, emotional distress, and atrocities that might have been avoided through vigorous public discussion. And it's always been in the name of the public good, or saving souls, or protecting the innocent, or some other unassailable and noble good that people with power have forced others not to say certain words. The idea that no, this time, it's different suggests a certain historical hubris.

I'd wager that throughout history, there's never been a society that's made its people happier or better-off through censorship. I'd love to see a counterexample if you can find one.

You have a number of points I agree with, but I find other points troublesome.

For instance, the costs of censorship are, interestingly enough, also the costs of not censoring free speech. Slowed scientific progress (due to the rise of anti-intellectualism), emotional distress (pretty obvious how some types of free speech cause this), and atrocities (e.g. Charlotesville) could have been avoided through censorship. Those are the material costs of allowing unbridled, "hate group" free speech.

On a side note, I am very personally conflicted on this topic. It was strange to read your comment and strongly agree with some statements and then very strongly disagree other statements.

Yes. People seem to be fixated on the perceived evils of the day and forget about how all of history is full of competing political groups and ideas, most of them promoting or using violence to gain power. If it was obvious how to pick goodies and baddies, hardly anyone would join the baddies.

Whether hate speech should be censored is also decided by whoever is in power at the time. The Adolf Hitler regime isn't going to be prevented from rounding up dissidents just because Weimarbook decided to be supportive of nazis.

I don't get the arguement that Facebook, (previously a way to stay in touch with friends) should have to broadcast every crank who wants to talk. If you want to be in a newspaper or radio you have to hit a certain quality level, and be the right kind of content. This editorship is not a restriction of speech, I can start my own website, print my own newsletter etc.

This is probably too reductive, but ultimately it's profitable to do so. Facebook is not in the news business, they're in the data/advertising business.

> Any online open forum would have the same problems no matter how hard they tried to prevent it.

Then don't make it open. Make it auditable and tied to your public identity, not to an email address you can invent at any time. The OP is right, FB is making huge profits by punting the costs onto society.

Do you want to forcibly prevent others from engaging in anonymous discussion? Have you given any thought to the implications of that?

FIDO.NET was a fairly polite society. There’s a lot to learn from then if there is will.

There's only one way I've seen in which one can reliably make and maintain a polite community: build it around a niche topic and moderate away things that stray too far away from it. The moment you let your community be about anything and everything is the moment the "unwashed masses" come and the whole thing starts catering to the lowest common denominator, which is a pretty shitty level.

It's not a complete recipe, but I feel it's a necessary condition. This is a big part of why HN is a polite place, or why most niche subreddits are polite places.

Your forum could still be "about anything", yet not cater to lowest common denominator.

It's only when you deliberately want to attract (and retain) as many eyeballs as possible, then the quality necessarily has to go out the window.

It works just fine with medium to large online forums, given they have a good and motivated moderator team. Just not for global online forums like FB.

> Then you should not fucking be in the business at all!

You seem to be venting incoherently. How does that follow from the statement you're replying to, at all?

Seems perfectly coherent to me. He's saying that if your business can't afford to solve the problems it creates (and instead pushes the problems onto society at large and/or disposable workers), then he thinks that business should go under.

The prevalence of fake news and hate speech on Facebook is caused by problems in society itself, not by Facebook.

"That house was flammable all on its own, all I did was provide a match".

That analogy makes no sense to me. Can you clarify what you mean by this?

"all I did was being first to provide a match, in the town of Internet, where everyone is a distributor of matches."

Facebook is a massive enabler for the spread of fake news/hate speech. I do not doubt that there are problems in society, but Facebook needs to be held accountable for its (in)action.

Should paper manufacturers be responsible for the content that’s printed? Should ink manufacturers be held liable for what people do with it? Should the post office be held liable for messages being mailed? Remember — Facebook is voluntary. Nobody is forced to be there and if there are uncomfortable messages to which you are being exposed, turn it off. Sign out. Quit using it.

You seem to have an almost hysterical reaction to this story. Why are you so sensitive? Did Nazis and violence and Red Brigade terrorists and Communists and racists and Islamic extremists not spread their messages effectively before Facebook? Has the rate of violence from these groups increased since Facebook?

I don’t have stats in front of me, but I would argue that violence, terrorism and hate has decreased since Facebook. Sure we hear more about it now, but that’s a result of media penetration and not an actual increase in incidents. I would invite data that suggests otherwise.

One might argue that awareness of hate groups has increased because of Facebook thus leading to increased marginalization of their ideas.

Blaming the medium for the message is fascinating. The disease of hate and totalitarianism spread quite effectively before Facebook. We might argue that Facebook has actually reduced such things because it provides a marketplace where ideas can be evaluated and either accepted or shunned much more efficiently.

Banning the communication of ideas doesn’t stop the communication of ideas. We learned that from Solzhenitsyn. We should encourage more openness, less censorship. Sunlight disinfects.

> Should paper manufacturers be responsible for the content that’s printed

No, but publishers should be, and they are. FB is a publisher, a distributor of content. They are not the creator of electrons or LED displays via which your content is distributed, which would be the closest analogy to your paper and ink manufacturer.

No, the real publisher is that dumbass Facebook friend of mine, who shared a "fake news" story. And that dumbass Facebook friend of yours who saw the post of my dumbass friend and reshared it further.

That's the writer, the publisher is the entity providing the means of distribution. That's FB.

You make some good points. It's curious to see people attacking Facebook so vehemently, as if using it were somehow obligatory. I don't see much appeal in Facebook myself, but I don't feel the need to force my views on others. Is the Overton window shifting toward outright censorship? It's an interesting question.

Your comment was downvoted a lot though. Must've hit a nerve :-)

I don't see how they can do that. There is little qualitative difference between what you probably mean by "fake news" and the stuff that mainstream news outlet put out. To stay safe, Facebook would either have to refuse allowing any news content on their site, or else turn into a world's leading intelligence agency / think tank to validate all of it.

The qualitative difference is that traditional news outlets are held accountable for their "fake news". The New York Times tries not to publish untrue claims, and if it does, it will retract them and issue a correction or apology. If it won't do that, there are libel laws and other legal remedies.

Fake news on social media is subject to none of these controls. Facebook is a multiplier of some of the worst tendencies of human beings, and it lacks the institutions that the rest of society has evolved over centuries to contain those tendencies.

Facebook is not a publisher like the New York Times is, and it's very disingenuous to suggest otherwise.

I don't think it matters really. If Facebook's current practices cause, enable, or magnify harm to society, then there are levers that should be pulled to minimize that harm.

Facebook is, in part, a tool for distributing misleading information on a massive scale. We control tools that cause harm even if those tools are inert or even beneficial if they're not used by malicious people. Facebook has no institutional or economic reason to reduce the harm it causes, and it's the role of government to reduce harm when the market has failed to do so.

I think the commenter is making the point that they have grown to a level unsustainable to deal with the issue of exposing it's users to fake, hateful, or abusive content.

No, the crazies finding each other and organizing is entire FB's fault.

Are you being sarcastic? Sometimes it's hard to tell online.

Not really. I think this phenomenon is well documented at this point. How else would fake news spread so easily across social media? Everyone has a couple of crazies in their social circle, and they will broadcast false statements as facts. Due to a human cognitive quirk, we lend more credence to propositions we hear from multiple sources. The "crazies" that already believe in nonsense already do this on reflex. And so the crazy spreads, even to otherwise reasonable people.

That's not what you said. What you said was

> The crazies finding each other and organizing is entire FB's fault.

If it is entirely Facebook's fault, how come this phenomenon predates Facebook and exists outside of it?

People got sick before the great flu pandemic of 1912, therefore the flu was not responsible for the 1912 flu pandemic.

Do you see the problem with this argument? FB was rapidly accelerated the organization of these types of extreme positions. My original comment follows from my last comment. What do you think the crazies do when they've gotten all outraged at some false proposition or other?

> People got sick before the great flu pandemic of 1912, therefore the flu was not responsible for the 1912 flu pandemic.

That's not the right analogy. This is closer to what you said:

Human epidemics are entirely the fault of the great flu pandemic of 1912.

An even closer analogy to what you said is:

Human epidemics are entirely the fault of widespread transportation networks. Ergo we ought to abolish such networks.

Err, no. I never said crazies never found each other before, I said the pandemic of crazies finding each other and organizing now are the fault of FB (and other social media of course).

Then it is not "entirely" Facebook's fault, as you suggested. Such hyperbole obscures rather than illuminates truth.

Do you have any empirical evidence to back this "pandemic" claim of yours? Or is that just a feeling you have?

Facebook is making this easier because it makes keeping with touch and finding people easier. But Facebook is only an incremental step forward here; the objection you're raising should be aimed at the entire Internet. It's the Internet that lets crazies find each other and organize; before Facebook, it was other sites, other services.

I'd go one step further - that's just consequence of better communication. You can't have one without the other. If you make life easier for everyone, you're also making life easier for the crazies, and for the bad guys. Personally, I'd still say it's worth it.

That's not what the comment said. Here is the conversation:

Original comment: You can hire lots of humans to work at a normal pace and give them psychological counseling, and spend so much on their salaries that you'll go out of business, and even then they won't catch everything and will be frequently accused of bias.

Reply: Then you should not fucking be in the business at all! What FB does is externalizing the cost onto societies and skimming the profit!

I still fail to see how that is a coherent and relevant reply. It comes across as shouting and venting rather than substantive.

Are you a non-native English speaker? It is perhaps not a very elegant response, but it's perfectly coherent and relevant.

Consider the original comment to be the first clause in an 'if-then' sentence, i.e. "IF the cost of hiring lots of humans to work at a normal pace and giving them psychological counseling causes you to go out of business, THEN you should not fucking be in the business at all, BECAUSE otherwise you are externalizing the cost onto society and skimming the profit."

Basically, Facebook should be taxed for the externalities they create like fake news (just as manufacturing plants should be taxed for the externalities they create by polluting). They should also be required to provide a safe workplace.

So if it turns out that their moderation policy creates big problems for their employees and the world, thought should be forced to shoulder a financial burden in the form of taxes/workers comp.

Yes, these are problems you come across in a nanny state. Which is why in the US, most all speech is protected, even hate speech. And for the government to impose laws or rules to force a private company to regulate that speech runs counter to that. So I'm not really sure why this is a conversation in the first place.

there is a large set of rules that define what can't be said or shown on radio or TV

Only on TV that is publicly broadcast over the public airwaves. Anything private like cable or satellite can say anything they want.

> Basically, Facebook should be taxed for the externalities they create like fake news

All economic activity produces externalities, positive and negative. That doesn't mean we should increase taxes and subsidies on everything.

I think parent's point is that "fake news" is a negative externality of the same kind as pollution.

Which it isn't. Pollution is a well-defined term; you can identify particular offending substances or disposal methods by their impact on humans and their environment. There are precise things that can be taxed to discourage their use. "Fake news", on the other hand, is a pretty uniform spectrum, from the most blatant falsehoods to the stuff CNN and NYT write. It's hard to even agree on where to draw the line (personally, I'd draw the line at "journalism should begin and end at stating verifiable facts + opinions on those facts clearly marked as opinion", but that's an extreme/minority position), and there's no way to start quantifying impact.

> You seem to be venting incoherently.

This breaks the HN guidelines. Please post civilly and substantively, or not at all—regardless of how bad another comment is.

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

Nazis and jihadists aren't that bad. They're just people promoting their political ideology. That's everywhere. Look at gangsterism - it's a major cause of murders in the US but it's glorified and promoted by rap music. Communism is arguably worse than Naziism but it's not treated with the same horror.

We've banned this account for trolling.

So what? Nobody builds social networks? Great idea, let's just all agree not to build social networks or make companies out of them. Because as we all know humanity is great at not doing things we have every capacity as well as a handful of reasons to do.

Try thinking about a real solution to the problem next time that isn't "let's not do this".

I'd flip this: if a society is so weak that it's damaged by "letting Nazis and jihadists spread their propaganda" (i.e. by letting people speak freely) then that society should fail.

(I do think that if Facebook is employing people to do something that predictably harms those employees then it should bear the cost of protecting them from those harms, and treating and compensate them when it fails to protect them)

Really! It's like someone founded a nuclear power startup and people in this thread are basically excusing them from properly dealing with the nuclear waste because, you know, it's really hard.

>At Facebook's scale, there is not a good way to do it. Even if they were able to scale it, it would not be good enough anyway

Facebook is picking the quota for decisions per hour or whatever, and choosing whatever help is available to reviewers. Certainly they have some culpability for the outcome.

I wasn't trying to apportion blame, because I don't see how it's useful to do so when there is no possibility of a good outcome.

If Facebook adjusts the quotas, they will just be doing a more effective job at pursuing an unworkable goal.

> At Facebook's scale, there is not a good way to do it.

That seems quite unpersuasive, especially with no citations given.

Just off the top of my head-- email exists at roughly the same scale as Facebook. Yet those shady "fwd:fwd:fwd" stories do not propagate over email at anywhere near the velocity of the propaganda campaigns which saturate Facebook. And at the end of the day those "fwd:fwd:fwd" stories don't reach nearly the number of end users as the propaganda campaigns do because of the critical mass of email users unwilling to forward that crap on to their contacts.

> That seems quite unpersuasive, especially with no citations given.

Well there is no good approach to censorship, it's censorship after all, you will always piss people off.

People need to grow up, there is stuff out there that may dicomfort your world view but get used to it. Also only because you are deleting comments on facebook does not mean - at all - that you are tackling the cause, you just swipe the dust under the carpet.

Well, maybe because email is by default designed to be 1 to 1, and 1 to many is considered to be borderline spam? While facebook is always 1 to many.

> Yet those shady "fwd:fwd:fwd" stories do not propagate over email at anywhere near the velocity of the propaganda campaigns which saturate Facebook.

Because people no longer use e-mail for private communication. I remember the times of "fwd:fwd:fwd" stories - they were popular before IMs and social media.

And let's not forget the ultimate, age-old source of "fake news" - a real-life, face-to-face conversation with a friend or family member that starts with "did you hear that $<fake bullshit news of the day>".

>At Facebook's scale, there is not a good way to do it. Even if they were able to scale it, it would not be good enough anyway.

Then maybe we need to place limits on their scale if they can't handle the problems their scale has created.

"You can hire lots of humans to work at a normal pace and give them psychological counseling, and spend so much on their salaries that you'll go out of business, "

They could hire 10000 people at 100k each. This would cost 1 billion per year making 3 billion every quarter.

And how much infrastructure would be required to keep those 10000 going? How much office space, IT equipment, managers, etc.

You’re suggesting that the company should grow its headcount by around 40%!

Plus, somehow in all this discussion, people seem to be accepting the idea on its face that somehow psychological counseling would fix said issues.

Maybe in some cases it would, but we are far from being able to cure many psychological problems consistently.

There are certain videos that contain extremely graphic violent content and could easily be automatically detected and blocked because they are reposts of the same video. That is a simple problem for which Facebook should be able to apply existing technology. There is a lot Facebook can do, but they are not, because they are lazy.

> apply existing technology. There is a lot Facebook can do, but they are not, because they are lazy.

Mmmh. You do realise that the state of the art technology that you refer to was developed by Facebook, right? I mean, there are no public communications around whether and which type of image recognition is used to detect harmful content at scale for obvious reasons but… take a wild guess.

Do you _really_ think engineers at Facebook who love to solve problems at scale would prefer to have an army of contractors handle something hurtful if there was a solution involving smart code and a battery of servers?

So, if FB has the state of the art and they are still employing armies of contractors then maybe they are just using said contractors to create their training data set.

The contractors’ feedback is certainly considered a high-signal dataset. I am not directly familiar with that aspect of the company (I worked on things somehow related but not this) but I suspect there are several ways of training and assessing estimations that certain content breaks certain rules (gore, direct threat, child endangerment, etc.). There could be challenges (legal, quality issues) of using automated decision to block or not block different types of flagged content.

One thing that, as a data scientist, I noticed: none of those reports complains about seeing the same image repeatedly. I understand that posters are often tempted to spread new content, so I would expect there to be many repeats. It could be because they see so many similar and shocking things they don’t recognise. Or companies could use techniques to pool together content that is identical or similar enough.

> There is a lot Facebook can do, but they are not, because they are lazy.

Therefore they contract it out rather than automate it? Your comment makes no sense.

I'm saying they should be able to automate flagging of the really bad stuff because it will likely be copies of the same video (by virtue of being so bad) ...

So why not provide multiple Facebook views? That is, put everything in the same databases. Automate creation of flags for posts and accounts. Then let users set moderation level as they desire. Some would want everything. The rest would specify what they don't want to see. And they'd be warned about false positives. Some governments would limit options, I'm sure. But we already have that, so hey.

That's exactly what they do...the algorithmic feed is basically what you described except instead of it being very explicit upfront, the signals are a combination of explicit and implicit and the resulting view is always evolving independently for each person.

OK, I get that. But there's no option for zero filtering.

I'm thinking of the old talk.masked Tor onion service. Except near the end, there was zero censorship. But users could set their own filters.

I kinda get what you're saying but in practice zero filtering would be a horrible horrible user experience. The content velocity on Facebook is so very high that unless you filter the feed in some way - it becomes a different product entirely, or perhaps even unusable.

And in the case of "Let the user choose the filters however they want", the business portion of the company takes on a massive risk. Remember this isn't a charity. Facebook wants to be able to make a lot of money and exist independently as an organization. Furthermore, they want to be able to predictably grow the business.

The reason I push back against your suggestions is because I see them often - solutions that swing too far into the realm of idealism and entitlement. Facebook is a business, so you have to be able to provide solutions that recognize the importance of the business side of things.

OK, I get that. Unfiltered Facebook would be a very different thing. And maybe not viable as a business. Because public outrage about stuff that offends them. And because laws against hate speech, subversive agitation, and so on. It'd probably need to be a Tor onion service, to protect the provider. Such as Diaspora or whatever.

And yes, I am an idealist. As Crowley defined Thelema, "Do what thou wilt shall be the whole of the Law", and "Love is the law, love under will." The problem, however, is the haters. Burroughs joked about just killing them where they stand, and I suppose that's consistent with Thelema. But I can't see how it'd work in practice.

Because it's not about protecting people from being hurt by seeing what they don't want. It's about protecting people's minds from being exposed to ideas that Facebook or the government thinks they shouldn't think.

I'd say that it's both. Laws against hate speech are intended to protect people against seeing what might hurt them. But yes, many governments censor subversive speech, and force Facebook etc to follow their lead.

Why do they need to moderate every post. The problem posts will be coming from a minority of accounts. Ban them on first offense, or maybe allow one warning, and people will figure out pretty quickly that Facebook is not the place to post that sort of thing.

1) Because you don’t know who the offenders are until you show their content to a moderator.

2) Because zero-tolerance tends to result in over-blocking, and they already have a problem with banning content that shouldn’t be banned.

I think Facebook knows enough about its users to precisely identify the accounts that are likely to be offenders.

Perhaps, but to the extent that it does that’s a combination of moderation and machine learning — exactly what post several layers above was saying is insufficient.

Bans are easy to circumvent, and the worst trolls tend to be technically savvy enough to create new accounts. They'll do it forever. They'll hide behind VPNs and anonymizing proxies if that's what it takes.

Also, what do you do if the hate speech is posted or linked to by a notable public figure, such as the President of the United States or a member of Congress?

So? The system doesn't have to be perfect, it only needs to make trolling time consuming enough to reduce it to tolerable levels.

> hate speech is posted or linked to by a notable public figure

Here's how I would approach it. Accounts that are tied to real identities have a much higher threshold for bans -- public figure or not. Sans a credible threat of harm to specific individuals they can largely say whatever they would like, hateful or not. When pseudoanonymous accounts get flagged their choice is to attach a publicly visible identity to the account or be banned.

It has to be perfect. People will be enraged if they can find any example of it not working. The human brain only understands examples, not rates. Facebook is now an all-knowing god that cannot make mistakes or not know something.

Even a single mistake will end up in the news as evidence that Facebook is "right wing" or "hard left," and perhaps both at the same time.

If you think this is easy, you don't understand the problem.

Reply on news.ycombinator.com