Daniel’s Heavy Tail Hypothesis (HTH) vs. this recent comment from Brian saying that he thinks that classic piece on ‘Why Charities Usually Don’t Differ Astronomically in Expected Cost-Effectiveness’ is still essentially valid.
Seems like Brian is arguing that there are at most 3-4 OOM differences between interventions whereas Daniel seems to imply there could be 8-10 OOM differences?
Here is my first draft, basically there will be a plan money prediction market predicting what they community will vote on a central question (here “are the top 1% more than 10,000x as efffective as the median”) then we have a discussion and we vote and then resolve.
What is a big open factual non community question in EA. I have a cool discussion tool I want to try out.
Daniel’s Heavy Tail Hypothesis (HTH) vs. this recent comment from Brian saying that he thinks that classic piece on ‘Why Charities Usually Don’t Differ Astronomically in Expected Cost-Effectiveness’ is still essentially valid.
Seems like Brian is arguing that there are at most 3-4 OOM differences between interventions whereas Daniel seems to imply there could be 8-10 OOM differences?
Similarly here: Valuing research works by eliciting comparisons from EA researchers—EA Forum (effectivealtruism.org)
And Ben Todd just tweeted about this as well.
Here is my first draft, basically there will be a plan money prediction market predicting what they community will vote on a central question (here “are the top 1% more than 10,000x as efffective as the median”) then we have a discussion and we vote and then resolve.
https://docs.google.com/document/d/14WpLjsS6idm8Ma-izKFOwkzy-B2F6RDpZ0xlc8aHlXg/edit