3 ms·

Here is one of the challenges with this space. What can be used to suppress misinformation can also be used to suppress information as well. Doing nothing has

• makomk 5y ago
  Any attempt to suppress "harmful misinformation" will be seen as partisan because, well, it pretty much is. There was a huge amount of utter bullshit pushed on social media about the 2016 election being stolen by fraud, including stuff that the author obviously could not possibly have worked as described or been used to steal the election, and a large proportion of the US population literally believed Russians had hacked the voting tallies. It obviously wasn't harmless either: someone radicalized by Facebook went and shot a Republican senator, and it was only through masses of luck and intrusive medical interventions that he survived. Yet the only thing the mainstream media showed an iota of concern for was the fact that anyone objected to this. Hell, even in the run-up to the 2020 elections the press were pushing the narrative that it'd be impossible to know the results weren't hacked and the election was valid - right up until it became clear Trump had lost, at which point it became absolutely certain they were valid and the audit chains and processes so robust only a conspiracy nut would question them. The same kinda happened in reverse in 2016.
  • btilly 5y ago
    I'm pretty sure that the "senator" you are naming was Congressman Steve Scalise. I have not specifically heard that the shooter, James Hodgkinson, believed in conspiracy theories about a stolen 2016 election. But he may have. However the big difference between partisan beliefs about the two elections is this. After 2016, Democrats mostly believed that they lost at the polls due to an effective disinformation campaign run by the Russians. After 2020, Russians believed that they lost due to widespread vote counting fraud. As https://www.rollcall.com/2021/02/24/partisan-voters-claim-we-wuz-robbed-no-they-werent/ https://www.rollcall.com/2021/02/24/partisan-voters-claim-we... says, Democrats and Republicans believe this by almost exactly the same margins. But what is believed matters. Democrats may have been furious, but they believed in rule of law. Radical Republicans, by contrast, attempted to overturn the election via an insurrection.
• sixothree 5y ago
  Let’s not pretend Facebook ever was or ever will be some bastion for democratic values. We shouldn’t expect that not “suppressing information” is ever in their best interest. Or that it would be something new for them. It seems like fantasy to think in those terms.
• pessimizer 5y ago
  > So how do we suppress harmful misinformation, such as that the 2020 election was stolen through widespread fraud. And not provide tools that can be misappropriated against truths which are inconvenient to people with money and influence? We recognize that the "we" you're referring to are the "people with money and influence." "Harmful misinformation" from their perspective is information that they dispute either factually or through implication or perspective that could harm them, the people with money and influence. > So how do we suppress harmful misinformation, such as that the 2020 election was stolen through widespread fraud. With clear and extensive explanations and transparency. The people who insisted loudly that there was no widespread fraud the moment the election ended in their favor were operating with as little factual, auditable information as the people who were insisting that there was. The people who were insisting on a fraud narrative were of course consuming a lot more misinformation, but both sides were pretending that they were knowledgeable about something they weren't, at all.
  • btilly 5y ago
    First, the "we" that I was referring to was "we as a society" and, more specifically, "we as technologists". As in, how can we create a technology and social norms that both encourage a fact-based narrative while being resistant to political manipulation. Also to your specific example of the election, a lot of people who spoke out fairly quickly against the fraud information were operating on the factual, auditable information that both sides were presenting their data to judges of various persuasions around the country, and the judges were virtually unanimously concluding that there was no case. And then various recounts began coming back, likewise concluding that there was no fraud. I don't know what standard of evidence you think people should have spoken out at. But that seemed at the time to be a reasonable level of evidence. And it still does. The judges in this case were literally a collection of people chosen to be trustworthy, with varying political alignments, who were making informed decisions on the basis of more data than I have, and consistently came to the same conclusion. A similar kind of thing for fact checks would be a standard that I could be comfortable with. But it has to be similar. A collection of independent people. Chosen on the basis of methodology. With different political alignments. And it is only when they broadly agree that we are willing to impose rules.
• dredmorbius 5y ago
  It's almost as if algorithms and procedures aren't enough, that values and judgements matter.