4 ms·

With respect, "Yes, so?" to the moderation problem is like standing next to Everest and saying "Yes, so?" It's far and away the single biggest concern. Policy

With respect, "Yes, so?" to the moderation problem is like standing next to Everest and saying "Yes, so?" It's far and away the single biggest concern. Policy is everything. Lots of engineers don't understand this, as engineers we concern ourselves mostly with the "mechanism" side of the "mechanism vs policy" duality. We think about doing things, moving forward, confronting problems as we go. In reality, we live in a bubble. The real issues are policy domain. Policy is mostly about what not to do. Non-corrupt government operating a platform requiring moderation involves so, so many "do not do that, you lose" nodes on the decision tree as we presently understand it that there is no winning strategy. The only way to win is to not play that game. Don't get into the business of moderation if you are a non-corrupt government. > Why couldn't it be structured in a way similar to how PBS operates? Well, for a few reasons. The main one is that government in general and the US government in particular is extremely powerful. That power attracts attention. Everyone with an interest impacted by the government- which is everyone- needs to attend to that power. The President "favoring" a platform will attract many many parties to that platform, who wish to carry influence or have an impact. Many of these parties have their own immense powers, both technical and policy/informational. PBS is in literal terms a TINY operator. $700M in revenues last year. A staff of a couple thousand people. As an organization it doesn't have any of the capabilities to operate a real time two way communication platform requiring it to act as a real time moderator against the myriad actions that a China or Russia not to mention countless other actors could throw against it. Of course, many would argue neither does Facebook. But with 100,000 staff and hundred billion+ revenue- it has more of a chance. Could such an organization be stood up? The policy questions again are- who funds it, to what end? What problem would it solve? There are no good reasons for a govt to spend the $Bs it would require and no good solutions to the first amendment problems it would create. I appreciate that there is a notion of the "public interest" that one would think everyone agrees on, which is what one imagines governments should be about. There can be a sufficiently common shared understanding of "public interest" in small communities, even scaling to homogeneous states (say, single digit millions of citizens). But "public interest" breaks down pretty quickly with sufficient community heterogeneity. > Yeah, communication via Twitter or Facebook was was never considered official. Well, in fact, it was. Prior to Musk, many government agencies had "official" voices/presence on twitter. And much work went into- from a policy perspective- ensuring what was communicated was suitable. Occasionally policy was "made" or announced first on twitter. You will now see those presences increasingly quiet (or loud) quit, due precisely to Musk's failure at the moderation problem.

rglullis 3y ago
My "yes, so?" is not that dismissing the issue, it is actually just looking at Twitter (and also Facebook) and thinking "how come using those platforms with numerous issues are not a 'policy problem?" Twitter was hacked, the POTUS account was used to promote crypto scam, and this is not a "policy" issue? Threads is owned by a company who has repeatedly involved in scandals about how data is handled, how is that not a "policy" issue? I honestly don't buy this argument. If the government position was that nothing on social media should be "official" and this is why any presence gets less oversight, okay. If the position is that the presence is official as long as the participants adhere to govt guidelines, okay. But those same guidelines could be applied on a public-owned platform. What I don't get is that they can claim that their presence is official while on the private platforms but are somehow unable to act in the same way if the platform was public.
- jonahbenton 3y ago
  > how come using those platforms with numerous issues are not a 'policy problem? Yes, it is, in a way- there is an implicit endorsement. That has to be balanced with the opportunity to craft messages directly to the audience the potentially problematic platform serves. But use does not entail responsibility. FB/Twitter were/are policy problems, but their operation is not the responsibility of any specific actor in government. The issue is more diffuse. > Twitter was hacked, the POTUS account was used to promote crypto scam, and this is not a "policy" issue? Yes, it was a problem for Twitter, because they were the operator. The statements made under the POTUS account were not actually made by POTUS. So not really a policy problem for POTUS. > Threads is owned by a company who has repeatedly involved in scandals about how data is handled, how is that not a "policy" issue? Sure, but the government is not the operator. It is not responsible for Meta's failings. > I honestly don't buy this argument. If the government position was that nothing on social media should be "official" and this is why any presence gets less oversight, okay. If the position is that the presence is official as long as the participants adhere to govt guidelines, okay. But those same guidelines could be applied on a public-owned platform. Neither of those is the rule. When a designated government voice speaks, that's "official", no matter what medium through which that speech occurs. There may be problems with the medium, and the government authorities responsible for regulating those problems will take those issues up with the medium. They are responsible for regulating, not responsible for operating. The government entity using the medium will often continue to do so, even when there is a seeming conflict. One sees this all the time in the financial sphere, where there are many regulators and many service providers which in theory need to be in compliance with the requirements of the various regulators but every day there are questions and disputes and claims and counter-claims and so forth. It happens all the time that one government agency is making use of a service provider for something while another government agency is suing that same service provider. > What I don't get is that they can claim that their presence is official while on the private platforms but are somehow unable to act in the same way if the platform was public. Yeah- it's completely different, if by "public" you mean "substantially controlled or operated by a government entity." Because then government is both accountable and responsible. When it's a private operator, the operator is responsible, and government at best is just accountable, if something really bad happens.

rglullis 3y ago

My "yes, so?" is not that dismissing the issue, it is actually just looking at Twitter (and also Facebook) and thinking "how come using those platforms with numerous issues are not a 'policy problem?" Twitter was hacked, the POTUS account was used to promote crypto scam, and this is not a "policy" issue? Threads is owned by a company who has repeatedly involved in scandals about how data is handled, how is that not a "policy" issue? I honestly don't buy this argument. If the government position was that nothing on social media should be "official" and this is why any presence gets less oversight, okay. If the position is that the presence is official as long as the participants adhere to govt guidelines, okay. But those same guidelines could be applied on a public-owned platform. What I don't get is that they can claim that their presence is official while on the private platforms but are somehow unable to act in the same way if the platform was public.

jonahbenton 3y ago
> how come using those platforms with numerous issues are not a 'policy problem? Yes, it is, in a way- there is an implicit endorsement. That has to be balanced with the opportunity to craft messages directly to the audience the potentially problematic platform serves. But use does not entail responsibility. FB/Twitter were/are policy problems, but their operation is not the responsibility of any specific actor in government. The issue is more diffuse. > Twitter was hacked, the POTUS account was used to promote crypto scam, and this is not a "policy" issue? Yes, it was a problem for Twitter, because they were the operator. The statements made under the POTUS account were not actually made by POTUS. So not really a policy problem for POTUS. > Threads is owned by a company who has repeatedly involved in scandals about how data is handled, how is that not a "policy" issue? Sure, but the government is not the operator. It is not responsible for Meta's failings. > I honestly don't buy this argument. If the government position was that nothing on social media should be "official" and this is why any presence gets less oversight, okay. If the position is that the presence is official as long as the participants adhere to govt guidelines, okay. But those same guidelines could be applied on a public-owned platform. Neither of those is the rule. When a designated government voice speaks, that's "official", no matter what medium through which that speech occurs. There may be problems with the medium, and the government authorities responsible for regulating those problems will take those issues up with the medium. They are responsible for regulating, not responsible for operating. The government entity using the medium will often continue to do so, even when there is a seeming conflict. One sees this all the time in the financial sphere, where there are many regulators and many service providers which in theory need to be in compliance with the requirements of the various regulators but every day there are questions and disputes and claims and counter-claims and so forth. It happens all the time that one government agency is making use of a service provider for something while another government agency is suing that same service provider. > What I don't get is that they can claim that their presence is official while on the private platforms but are somehow unable to act in the same way if the platform was public. Yeah- it's completely different, if by "public" you mean "substantially controlled or operated by a government entity." Because then government is both accountable and responsible. When it's a private operator, the operator is responsible, and government at best is just accountable, if something really bad happens.