Some quick thoughts: - I’m continue to be thankful to the staff at CEA and hope the team good luck going forward, especially around the new ED and whatever direction that might bring. - I think it’s unhealthy for the ecosystem for CEA to get such a high ratio of funding from OP. When this happens, I assume that CEA adjusts to represent the interests of OP more than I’m comfortable with. As such, I’m excited by other donors becoming involved here. - At this point, I’m probably most excited about “long term institutional health” improving than anything else. Between FTX, some of this OpenAI fiasco, and other issues, I think that the “EA Bureacracy” has some maturing to do. If there are ways that CEA could help here (maybe by reinforcing/helping the community health team, or with some project like [these](https://forum.effectivealtruism.org/s/jxBRTDWZZYBbknuGK/p/Cvn6fwzdoLNLgTJif), I’d be most optimistic, though I realize these are tough to do well. - It seems tricky to raise money while waiting on a new ED, but so it goes. - I’m happy you provided BOTECs, but these strike me a bit like toy examples. Also, of course, happy to see Squiggle used! (I’m biased, of course). I’m not sold on the use of Shapley values for the first one—I’m personally not a big fan of Shapley values for these things—but this is a smaller point.
> we think that a more detailed model would obscure more than enlighten
I realize this might be controversial, but I’d be excited about a world where CEA posts generic costs+benefits of all of the big programs. I wouldn’t fault CEA, as no one else does this yet, but I think some of this would be very useful, though perhaps too confrontational.
Anyway, again, I’d really like to see CEA do well in the future, and would encourage motivated donors to ideally work with them to help make that happen. I’d be most excited about active donors who could really help give feedback on initiatives and help provide a contrasting viewpoint to OP.
I’d be excited about a world where CEA posts generic costs+benefits of all of the big programs. I wouldn’t fault CEA, as no one else does this yet, but I think some of this would be very useful, though perhaps too confrontational.
Agreed that this could be very useful. It could also (as I argued here) be useful to have more such models produced by independent evaluators.[1]
I’d be excited about a world where CEA posts generic costs+benefits of all of the big programs
Could you say more about what this looks like? In particular: are you suggesting that we pull together information that is already publicly available but spread out (e.g. metrics of how much we impact people are spread across various EA survey and OP LTist survey posts) versus publishing something which doesn’t already exist (e.g. our internal estimates of the dollar value we should assign to those survey results)?
Many of CEAs projects seem nicely split up. (Events, Community Health, Groups).
Costs
I could imagine a list of the money / counterfactual costs spent on each. A bit tricky, but doable. You’d have to include overhead, of course.
Benefits
Very simple models of the value of each project. These could be in some kind of relative value, so we could at least tell things like, “Project X seems to be more valuable in total than Project Y, though less efficient”
Even relatively simple versions of this could be illuminating and help for making consensus.
I’d be happy to put together a simple potential version of this if that could be useful. It could make sense for it to be private (within the org) for a while, maybe a long time, if you don’t have internal accounting like this already.
Some quick thoughts:
- I’m continue to be thankful to the staff at CEA and hope the team good luck going forward, especially around the new ED and whatever direction that might bring.
- I think it’s unhealthy for the ecosystem for CEA to get such a high ratio of funding from OP. When this happens, I assume that CEA adjusts to represent the interests of OP more than I’m comfortable with. As such, I’m excited by other donors becoming involved here.
- At this point, I’m probably most excited about “long term institutional health” improving than anything else. Between FTX, some of this OpenAI fiasco, and other issues, I think that the “EA Bureacracy” has some maturing to do. If there are ways that CEA could help here (maybe by reinforcing/helping the community health team, or with some project like [these](https://forum.effectivealtruism.org/s/jxBRTDWZZYBbknuGK/p/Cvn6fwzdoLNLgTJif), I’d be most optimistic, though I realize these are tough to do well.
- It seems tricky to raise money while waiting on a new ED, but so it goes.
- I’m happy you provided BOTECs, but these strike me a bit like toy examples. Also, of course, happy to see Squiggle used! (I’m biased, of course). I’m not sold on the use of Shapley values for the first one—I’m personally not a big fan of Shapley values for these things—but this is a smaller point.
> we think that a more detailed model would obscure more than enlighten
I realize this might be controversial, but I’d be excited about a world where CEA posts generic costs+benefits of all of the big programs. I wouldn’t fault CEA, as no one else does this yet, but I think some of this would be very useful, though perhaps too confrontational.
Anyway, again, I’d really like to see CEA do well in the future, and would encourage motivated donors to ideally work with them to help make that happen. I’d be most excited about active donors who could really help give feedback on initiatives and help provide a contrasting viewpoint to OP.
Agreed that this could be very useful. It could also (as I argued here) be useful to have more such models produced by independent evaluators.[1]
Although, I also think there is value in seeing models from the orgs themselves for ~ reasoning transparency purposes.
Thanks Ozzie!
Could you say more about what this looks like? In particular: are you suggesting that we pull together information that is already publicly available but spread out (e.g. metrics of how much we impact people are spread across various EA survey and OP LTist survey posts) versus publishing something which doesn’t already exist (e.g. our internal estimates of the dollar value we should assign to those survey results)?
Many of CEAs projects seem nicely split up. (Events, Community Health, Groups).
Costs
I could imagine a list of the money / counterfactual costs spent on each. A bit tricky, but doable. You’d have to include overhead, of course.
Benefits
Very simple models of the value of each project. These could be in some kind of relative value, so we could at least tell things like, “Project X seems to be more valuable in total than Project Y, though less efficient”
Even relatively simple versions of this could be illuminating and help for making consensus.
I’d be happy to put together a simple potential version of this if that could be useful. It could make sense for it to be private (within the org) for a while, maybe a long time, if you don’t have internal accounting like this already.
Yes this would be amazing! We’re trying to do something similar with EA Netherlands.