I think the issue you’re addressing is a real and important one. However, I think current norms are a response to disadvantages of blurting, both on an individual and movement level. As you note, most people’s naive divergent first impressions are wrong, and on issues most salient to the community, there’s usually somebody else who’s thought about it more. If we added lots more blurting, we’d have an even greater problem with finding the signal in the noise. This adds substantial costs in terms of reader energy, and it also decreases the reward for sharing carefully vetted information because it gets crowded out by less considered blurting.
Hence, the current equilibria, in which ill-considered blurting gets mildly socially punished by people with better-considered views frustrated by the blurter, leading to pre-emptive self-censorship and something of a runaway “stay in your lane” feedback loop that can result in “emperor has no clothes” problems like this one. Except it wasn’t a child or “blurter” who exposed SBF—it was his lead competitor, one of the most expert people on the topic.
I’ve said it before and I’ll say it again, EA’s response to this fraud cannot be—not just shouldn’t, but can’t—to achieve some combination of oracular predictive ability and perfect social coordination for high-fidelity information transmission. It just ain’t gonna happen. We should assume that we cannot predict the next scandal. Instead we should focus on finding general-purpose ways to mitigate or prevent scandal without having to know exactly how it will occur.
This comes down to governance. It’s things like good accounting, finding ways to better protect grantees in the case that their funder goes under, perhaps increased transparency of internal records of EA orgs, that sort of thing.
If we added lots more blurting, we’d have an even greater problem with finding the signal in the noise.
The EA Forum is a hub for a wide variety of approaches and associated perspectives: global development randomista, anti-factory-farming activist, pandemic preparedness lobbyist, AI alignment researcher, Tomasik-style “what if electrons are conscious?” theorist, etc. On top of that, it has a karma system, post curation, and many options for filtering/tagging and subscribing to particular kinds of content.
So both in terms of the forum’s infrastructure and in terms of its content and audience, I have a hard time imagining a more ideal venue for high-quality 101-level questions and conversations. What specific signal are people worried about losing? Is there any way to (e.g., with tags or mod curation or new features) to encourage freer discussion on this forum and preserve that signal?
(One bit of social tech that might help here is just to flag at the top of your comment what the epistemic status of your statements is. That plus a karma system addresses most of the “wasting others’ time” problem, IMO.)
Except it wasn’t a child or “blurter” who exposed SBF—it was his lead competitor, one of the most expert people on the topic.
Sure. SBF isn’t my real crux for thinking EA is largely bottlenecked on blurting. It’s an illustrative example of one reason we’re likely to benefit from more blurting.
Discovering the fraud sooner probably wouldn’t have been trivial, especially if the fraud started pretty recently; but there are many outcomes that fall short of full discovery and yet are a fair bit better than the status quo. (As well as many other dimensions on which I expect blurting to improve EA’s ability to help the world.)
I’ve said it before and I’ll say it again, EA’s response to this fraud cannot be—not just shouldn’t, but can’t—to achieve some combination of oracular predictive ability and perfect social coordination for high-fidelity information transmission. It just ain’t gonna happen. We should assume that we cannot predict the next scandal. Instead we should focus on finding general-purpose ways to mitigate or prevent scandal without having to know exactly how it will occur.
I fully agree! But it sounds like SBF already should have been a fairly scandalous name throughout EA, based on reports about the early history of Alameda. Never mind whether we could have predicted the exact specifics of what happened at FTX; why did the Alameda info stay bottled up for so many years, such that I and others are only hearing about it now? This seems like a misstep regardless of how it would have changed our relationship to SBF.
I think the issue you’re addressing is a real and important one. However, I think current norms are a response to disadvantages of blurting, both on an individual and movement level. As you note, most people’s naive divergent first impressions are wrong, and on issues most salient to the community, there’s usually somebody else who’s thought about it more. If we added lots more blurting, we’d have an even greater problem with finding the signal in the noise. This adds substantial costs in terms of reader energy, and it also decreases the reward for sharing carefully vetted information because it gets crowded out by less considered blurting.
Hence, the current equilibria, in which ill-considered blurting gets mildly socially punished by people with better-considered views frustrated by the blurter, leading to pre-emptive self-censorship and something of a runaway “stay in your lane” feedback loop that can result in “emperor has no clothes” problems like this one. Except it wasn’t a child or “blurter” who exposed SBF—it was his lead competitor, one of the most expert people on the topic.
I’ve said it before and I’ll say it again, EA’s response to this fraud cannot be—not just shouldn’t, but can’t—to achieve some combination of oracular predictive ability and perfect social coordination for high-fidelity information transmission. It just ain’t gonna happen. We should assume that we cannot predict the next scandal. Instead we should focus on finding general-purpose ways to mitigate or prevent scandal without having to know exactly how it will occur.
This comes down to governance. It’s things like good accounting, finding ways to better protect grantees in the case that their funder goes under, perhaps increased transparency of internal records of EA orgs, that sort of thing.
The EA Forum is a hub for a wide variety of approaches and associated perspectives: global development randomista, anti-factory-farming activist, pandemic preparedness lobbyist, AI alignment researcher, Tomasik-style “what if electrons are conscious?” theorist, etc. On top of that, it has a karma system, post curation, and many options for filtering/tagging and subscribing to particular kinds of content.
So both in terms of the forum’s infrastructure and in terms of its content and audience, I have a hard time imagining a more ideal venue for high-quality 101-level questions and conversations. What specific signal are people worried about losing? Is there any way to (e.g., with tags or mod curation or new features) to encourage freer discussion on this forum and preserve that signal?
(One bit of social tech that might help here is just to flag at the top of your comment what the epistemic status of your statements is. That plus a karma system addresses most of the “wasting others’ time” problem, IMO.)
Sure. SBF isn’t my real crux for thinking EA is largely bottlenecked on blurting. It’s an illustrative example of one reason we’re likely to benefit from more blurting.
Discovering the fraud sooner probably wouldn’t have been trivial, especially if the fraud started pretty recently; but there are many outcomes that fall short of full discovery and yet are a fair bit better than the status quo. (As well as many other dimensions on which I expect blurting to improve EA’s ability to help the world.)
I fully agree! But it sounds like SBF already should have been a fairly scandalous name throughout EA, based on reports about the early history of Alameda. Never mind whether we could have predicted the exact specifics of what happened at FTX; why did the Alameda info stay bottled up for so many years, such that I and others are only hearing about it now? This seems like a misstep regardless of how it would have changed our relationship to SBF.