9 ms·

How social-media platforms dispense justice

63 points by privong 8y ago

vladd 8y ago
http://archive.is/MMWqL http://archive.is/MMWqL
jimnotgym 8y ago
It will to see how the platforms fare once they have to exercise editorial control over their content. It will be especially interesting to see how their assumed competitive advantage over 'traditional media' holds once it is subject to a massive increase in overhead, which surely follows from having an army of moderators. I wonder if they will start to ebb, consolidate, and find a place alongside other media. Those that are scarcely profitable right now (Twitter for instance) should be the early indicators.
- village-idiot 8y ago
  Honestly, I’m surprised that social media is a thing. It appears to make its users miserable, especially twitter and Instagram, and some of them barely break even.
  - TeMPOraL 8y ago
    You could say the same about heroin. Or even alcohol, really. Taking immediate pleasure at the price of pain later is a failure mode of human brain; we all fall for it, to smaller or greater extent.
    - village-idiot 8y ago
      Good point.
raincom 8y ago
Now these platforms hide behind "external fact checkers", "black box algorithms", etc. The real problem is that every fact is theory-laden, a lesson learned from the history and philosophy of (natural) sciences. Today, we are in a state of disputing facts, because people are disputing theories behind stating such facts. Let me give you an example: a reporter finds two dead bodies. The fact that two dead bodies discovered may not be under dispute. The moment, one says that these two were killed, the other says that these two committed suicide--you see, two theories (homicide vs suicide) are presenting the phenomenon in two different ways. Further accept that these two were indeed killed. Now look at possible reasons behind these deaths. Multiple people offer often competing, contradictory, reasons behind these deaths. Basically, there are no facts. What we call facts are theory-laden facts that are agreed up on by all sides; in other words, all side agree on the underlying theory (say, homicide).
- philipov 8y ago
  There are facts. You said yourself "The fact that two dead bodies discovered may not be under dispute." That's the fact. Everything else you said isn't a fact, it's a hypothesis. Just because people don't know the difference between a fact and a hypothesis/conclusion doesn't mean there are no facts; unless you want to go full ontological skeptic and dispute that it is possible to ever know anything (never go full sophist!). The real issue is that we have a breakdown of good faith in civil conversation. That makes society impossible. At the point where you argue that there are not two bodies in front of me, even though there are two bodies in front of me, I might as well stop arguing with you and instead start punching you. Then you can argue that your pain is just theoretical.
  - apocalypstyx 8y ago
    What if they were conjoined twins who couldn't have previously been separated without killing both? *edit Or, in less cutesy terms, considering that much of the pro/anti-abortion argument (on both sides) often rests on tacit and irreconcilable definitions of what a person is, the point where the woman's body ends and another begins, when is it appropriate (other than perhaps under the most obvious of circumstances) to charge and assailant that there were two homicides, rather than one? That is, when does the definition become more than personal belief?
  - coldtea 8y ago
    >There are facts. You said yourself "The fact that two dead bodies discovered may not be under dispute." That's the fact. Everything else you said isn't a fact, it's a hypothesis. Just because people don't know the difference between a fact and a hypothesis/conclusion doesn't mean there are no facts; The distinction between fact and hypothesis is a red herring. First, unless one has verified something themselves, with their own eyes or carefully checked primary sources, there are no facts, just reporting of things claimed to be facts. In other words, there are no "two bodies in front of you" and you're not being denied by some fellow you discuss with while you're both looking at them. Just reporting that such two bodies exist somewhere, that one believes and the other doesn't. And more often than not, the reporting is of even more abstract things, like statistics (collected with who knows what methodology, and presented and baked to prove who knows what point, using all the tricks one can use to lie with statistics). Second, supposedly physical-domain facts can be fake as well. "This man was shot by person X" (leaving no room to hypothesis) while X might not have done it, or might have been framed - even if there were witnesses attesting that X did it and even a court found it so (many people have been later found innocent e.g. by DNA or further research decades after the fact, victims of over-jealous prosecutors, false testimony, setups, facial similarities, racist bias, and so on). Despite their guilt being not a hypothesis but a fact with "evidence", it was still bogus. Second, pure hypothesis is often presented as fact, and people are called out for not believing it, all the time.
bogomipz 8y ago
>"Although most of the moderators work for third-party firms, the growth in their numbers has already had an impact on the firms’ finances." I was kind of surprised to read this. Wouldn't content policy be much more coherent if these were employees of the company? I'm curious why they elected to use third-party firms. I understand that localization is an issue but this a global company with deep resources so I'm doubting that could be it.
- mayniac 8y ago
  Maybe it's easier for them to increase manpower temporarily if it's outsourced? i.e. there's probably going to be more need to moderate Facebook in the run up to the midterms this November, and even more so with actual elections. There's probably a lot of factors that go into determining how many people are needed to moderate and it could be easier/cheaper to hand over staffing to third party firms. That and as you mentioned, localisation.
- gaius 8y ago
  I'm curious why they elected to use third-party firms. Plausible deniability, and no liability for the long term effects on the mental health of those exposed to extreme imagery. Knowing these companies almost certainly a tax dodge of some sort is involved too.
  - bogomipz 8y ago
    Sure plausible deniability sounds about right for these folks. However I would hope that the mileage on that is limited to single use. I'm sure they will use this excuse to point blame elsewhere during their next crisis but I doubt it will work for the one after that.
  - DoctorOetker 8y ago
    What if they found the "creative" solution to delegate the policing to the governments in question (as a "third party"), in exchange for not having to pay taxes. Then deniability goes both ways!
HONEST_ANNIE 8y ago
The root incentive to moderate is to keep the government from intervening or preventing people turning against the platform. Government starts interfering and people turn against them when there is a moral panic[1]. In democratic societies the best way for the social-media platforms to achieve their goal is to muffle moral panics directly. That's different goal from the moderation that government and public want from them (criminal or hidden actor propaganda.) I predict that the attention will be directed towards throttling sudden spikes in moral panic inducing content. Legitimate moderation will be just small part of the moderation effort. Most of the energy goes into protecting the sensibilities of "proper" society. There will be limiters for the reproduction number of any kind of controversial thinking. It will be just easier and cheaper. [1]: Moral Panics http://criminology.oxfordre.com/view/10.1093/acrefore/9780190264079.001.0001/acrefore-9780190264079-e-155 http://criminology.oxfordre.com/view/10.1093/acrefore/978019...
loup-vaillant 8y ago
> Internet firms in America are shielded from legal responsibility for content posted on their services. Section 230 of the Communications Decency Act of 1996 treats them as intermediaries, not publishers—to protect them from legal jeopardy. > When the online industry was limited to young, vulnerable startups this approach was reasonable. That's the wrong way to look at it. The size of the company should not matter here. It's how they operate that should determine which regulation applies to them. There are only two reasonable alternative: Either they are intermediaries, in which case they should have the duty to indiscriminately intermediate whatever they are asked to intermediate. This means no centralised moderation. Maybe allow moderation at the sub-community level (as Reddit does with its subforums), but no centralised censorship, except maybe upon reception of a warrant. Or they are editors, in which case they should have full legal responsibility of the content, and suffer full legal consequences of whatever illegal speech they allowed. I believe such a clean dichotomy is the only way to ensure free speech in the face of such centralised services. 'Cause I think the likes of Facebook, YouTube, Twitter… would rather choose the "intermediary" path than risk taking legal responsibility for their enormous user base.
- XorNot 8y ago
  This implies no suggestions algorithms and news feeds either. If these sites are pure intermediaries, then running machine-driven editorializing by controlling and suggesting the content people see is not a neutral act - those algorithms are biased and they choose how they act. Imagine a phone company which selectively didn't send calls to you because it's automated checkers determined you weren't interested. If you got nothing but telemarketing, well, now they're no longer a pure transit are they? Social media owners are already editors and they're trying to shirk their responsibility.
  - loup-vaillant 8y ago
    > This implies no suggestions algorithms and news feeds either. Yep, it does.
    - caseysoftware 8y ago
vinayms 8y ago
Social media companies must be allowed to do policing this way only if they abide by the national laws pertaining to freedom of expression. I mean, if a company wants to list as a social media company where people share information, and they also do policing, they must be answerable to local courts if they take down posts using their judgement which goes against the established freedom laws of the land, without hiding behind the garb of being a private enterprise etc and saying that its their prerogative to manage their platform as they please. I don't know if such a thing is already in place. The companies are having a knee jerk reaction to the misuse by public, and like all KJR this is heading the wrong way. The fact that the companies whose primary aim it to maximize profits are taking it upon themselves to do these sort of ethically and legally questionable activities itself will lead to bad ramifications. One obvious solution to controlling fake news is to remove the invisibility cloak of users by forcing them, by law, to use government issued ID to authenticate their accounts. This should make people more responsible before sharing misinformation. However, the fragility of technology that would lead to abusing authenticated user accounts is a big enough concern to put the onus on the companies. I don't care if the companies "employs executives who are thoughtful about the task of making their platforms less toxic while protecting freedom of speech" or use third party authenticating sources, or some fancy AI/ML, but unless they are forced to abide by the laws, things will only worsen. Whether they like it or not, social media companies are now entangled in social and legal aspects and have to deal with all that comes with it. They have enjoyed exponential growth in profits, now its time to get real.
xienze 8y ago
> or a photo alleging that Donald Trump wore a Ku Klux Klan uniform in the 1990s (leave it up but reduce distribution of it, and inform users it’s a fake). If they’re so concerned about misinformation, why would they leave a fake photo up, even with a disclaimer? Nah, I must be reading too much into this, surely this is a policy they apply evenly, and they’d do the same with a similarly fake photo of Obama.
- meowface 8y ago
  Not sure if you're being sarcastic. I assume they would do the same for doctored photos of Obama. Removing content entirely should only be reserved for when serious crime is likely to occur, I think.

63 points by privong 8y ago

http://archive.is/MMWqL http://archive.is/MMWqL

It will to see how the platforms fare once they have to exercise editorial control over their content. It will be especially interesting to see how their assumed competitive advantage over 'traditional media' holds once it is subject to a massive increase in overhead, which surely follows from having an army of moderators. I wonder if they will start to ebb, consolidate, and find a place alongside other media. Those that are scarcely profitable right now (Twitter for instance) should be the early indicators.

village-idiot 8y ago
Honestly, I’m surprised that social media is a thing. It appears to make its users miserable, especially twitter and Instagram, and some of them barely break even.
- TeMPOraL 8y ago
  You could say the same about heroin. Or even alcohol, really. Taking immediate pleasure at the price of pain later is a failure mode of human brain; we all fall for it, to smaller or greater extent.
  - village-idiot 8y ago
    Good point.

raincom 8y ago

Now these platforms hide behind "external fact checkers", "black box algorithms", etc. The real problem is that every fact is theory-laden, a lesson learned from the history and philosophy of (natural) sciences. Today, we are in a state of disputing facts, because people are disputing theories behind stating such facts. Let me give you an example: a reporter finds two dead bodies. The fact that two dead bodies discovered may not be under dispute. The moment, one says that these two were killed, the other says that these two committed suicide--you see, two theories (homicide vs suicide) are presenting the phenomenon in two different ways. Further accept that these two were indeed killed. Now look at possible reasons behind these deaths. Multiple people offer often competing, contradictory, reasons behind these deaths. Basically, there are no facts. What we call facts are theory-laden facts that are agreed up on by all sides; in other words, all side agree on the underlying theory (say, homicide).

philipov 8y ago
There are facts. You said yourself "The fact that two dead bodies discovered may not be under dispute." That's the fact. Everything else you said isn't a fact, it's a hypothesis. Just because people don't know the difference between a fact and a hypothesis/conclusion doesn't mean there are no facts; unless you want to go full ontological skeptic and dispute that it is possible to ever know anything (never go full sophist!). The real issue is that we have a breakdown of good faith in civil conversation. That makes society impossible. At the point where you argue that there are not two bodies in front of me, even though there are two bodies in front of me, I might as well stop arguing with you and instead start punching you. Then you can argue that your pain is just theoretical.
- apocalypstyx 8y ago
  What if they were conjoined twins who couldn't have previously been separated without killing both? *edit Or, in less cutesy terms, considering that much of the pro/anti-abortion argument (on both sides) often rests on tacit and irreconcilable definitions of what a person is, the point where the woman's body ends and another begins, when is it appropriate (other than perhaps under the most obvious of circumstances) to charge and assailant that there were two homicides, rather than one? That is, when does the definition become more than personal belief?
- coldtea 8y ago
  >There are facts. You said yourself "The fact that two dead bodies discovered may not be under dispute." That's the fact. Everything else you said isn't a fact, it's a hypothesis. Just because people don't know the difference between a fact and a hypothesis/conclusion doesn't mean there are no facts; The distinction between fact and hypothesis is a red herring. First, unless one has verified something themselves, with their own eyes or carefully checked primary sources, there are no facts, just reporting of things claimed to be facts. In other words, there are no "two bodies in front of you" and you're not being denied by some fellow you discuss with while you're both looking at them. Just reporting that such two bodies exist somewhere, that one believes and the other doesn't. And more often than not, the reporting is of even more abstract things, like statistics (collected with who knows what methodology, and presented and baked to prove who knows what point, using all the tricks one can use to lie with statistics). Second, supposedly physical-domain facts can be fake as well. "This man was shot by person X" (leaving no room to hypothesis) while X might not have done it, or might have been framed - even if there were witnesses attesting that X did it and even a court found it so (many people have been later found innocent e.g. by DNA or further research decades after the fact, victims of over-jealous prosecutors, false testimony, setups, facial similarities, racist bias, and so on). Despite their guilt being not a hypothesis but a fact with "evidence", it was still bogus. Second, pure hypothesis is often presented as fact, and people are called out for not believing it, all the time.

bogomipz 8y ago

>"Although most of the moderators work for third-party firms, the growth in their numbers has already had an impact on the firms’ finances." I was kind of surprised to read this. Wouldn't content policy be much more coherent if these were employees of the company? I'm curious why they elected to use third-party firms. I understand that localization is an issue but this a global company with deep resources so I'm doubting that could be it.

mayniac 8y ago
Maybe it's easier for them to increase manpower temporarily if it's outsourced? i.e. there's probably going to be more need to moderate Facebook in the run up to the midterms this November, and even more so with actual elections. There's probably a lot of factors that go into determining how many people are needed to moderate and it could be easier/cheaper to hand over staffing to third party firms. That and as you mentioned, localisation.
gaius 8y ago
I'm curious why they elected to use third-party firms. Plausible deniability, and no liability for the long term effects on the mental health of those exposed to extreme imagery. Knowing these companies almost certainly a tax dodge of some sort is involved too.
- bogomipz 8y ago
  Sure plausible deniability sounds about right for these folks. However I would hope that the mileage on that is limited to single use. I'm sure they will use this excuse to point blame elsewhere during their next crisis but I doubt it will work for the one after that.
- DoctorOetker 8y ago
  What if they found the "creative" solution to delegate the policing to the governments in question (as a "third party"), in exchange for not having to pay taxes. Then deniability goes both ways!

HONEST_ANNIE 8y ago

The root incentive to moderate is to keep the government from intervening or preventing people turning against the platform. Government starts interfering and people turn against them when there is a moral panic[1]. In democratic societies the best way for the social-media platforms to achieve their goal is to muffle moral panics directly. That's different goal from the moderation that government and public want from them (criminal or hidden actor propaganda.) I predict that the attention will be directed towards throttling sudden spikes in moral panic inducing content. Legitimate moderation will be just small part of the moderation effort. Most of the energy goes into protecting the sensibilities of "proper" society. There will be limiters for the reproduction number of any kind of controversial thinking. It will be just easier and cheaper. [1]: Moral Panics http://criminology.oxfordre.com/view/10.1093/acrefore/9780190264079.001.0001/acrefore-9780190264079-e-155 http://criminology.oxfordre.com/view/10.1093/acrefore/978019...

loup-vaillant 8y ago

> Internet firms in America are shielded from legal responsibility for content posted on their services. Section 230 of the Communications Decency Act of 1996 treats them as intermediaries, not publishers—to protect them from legal jeopardy. > When the online industry was limited to young, vulnerable startups this approach was reasonable. That's the wrong way to look at it. The size of the company should not matter here. It's how they operate that should determine which regulation applies to them. There are only two reasonable alternative: Either they are intermediaries, in which case they should have the duty to indiscriminately intermediate whatever they are asked to intermediate. This means no centralised moderation. Maybe allow moderation at the sub-community level (as Reddit does with its subforums), but no centralised censorship, except maybe upon reception of a warrant. Or they are editors, in which case they should have full legal responsibility of the content, and suffer full legal consequences of whatever illegal speech they allowed. I believe such a clean dichotomy is the only way to ensure free speech in the face of such centralised services. 'Cause I think the likes of Facebook, YouTube, Twitter… would rather choose the "intermediary" path than risk taking legal responsibility for their enormous user base.

XorNot 8y ago
This implies no suggestions algorithms and news feeds either. If these sites are pure intermediaries, then running machine-driven editorializing by controlling and suggesting the content people see is not a neutral act - those algorithms are biased and they choose how they act. Imagine a phone company which selectively didn't send calls to you because it's automated checkers determined you weren't interested. If you got nothing but telemarketing, well, now they're no longer a pure transit are they? Social media owners are already editors and they're trying to shirk their responsibility.
- loup-vaillant 8y ago
  > This implies no suggestions algorithms and news feeds either. Yep, it does.
  - caseysoftware 8y ago

vinayms 8y ago

Social media companies must be allowed to do policing this way only if they abide by the national laws pertaining to freedom of expression. I mean, if a company wants to list as a social media company where people share information, and they also do policing, they must be answerable to local courts if they take down posts using their judgement which goes against the established freedom laws of the land, without hiding behind the garb of being a private enterprise etc and saying that its their prerogative to manage their platform as they please. I don't know if such a thing is already in place. The companies are having a knee jerk reaction to the misuse by public, and like all KJR this is heading the wrong way. The fact that the companies whose primary aim it to maximize profits are taking it upon themselves to do these sort of ethically and legally questionable activities itself will lead to bad ramifications. One obvious solution to controlling fake news is to remove the invisibility cloak of users by forcing them, by law, to use government issued ID to authenticate their accounts. This should make people more responsible before sharing misinformation. However, the fragility of technology that would lead to abusing authenticated user accounts is a big enough concern to put the onus on the companies. I don't care if the companies "employs executives who are thoughtful about the task of making their platforms less toxic while protecting freedom of speech" or use third party authenticating sources, or some fancy AI/ML, but unless they are forced to abide by the laws, things will only worsen. Whether they like it or not, social media companies are now entangled in social and legal aspects and have to deal with all that comes with it. They have enjoyed exponential growth in profits, now its time to get real.

xienze 8y ago

> or a photo alleging that Donald Trump wore a Ku Klux Klan uniform in the 1990s (leave it up but reduce distribution of it, and inform users it’s a fake). If they’re so concerned about misinformation, why would they leave a fake photo up, even with a disclaimer? Nah, I must be reading too much into this, surely this is a policy they apply evenly, and they’d do the same with a similarly fake photo of Obama.

meowface 8y ago
Not sure if you're being sarcastic. I assume they would do the same for doctored photos of Obama. Removing content entirely should only be reserved for when serious crime is likely to occur, I think.