Open OAGr opened 2 years ago
Nuno's previous experiments in this vein seem relevant to prioritizing this.
Yep, he's been doing some work in this broad area for a while now. One of the hopes was that that work would be useful for something like this. (I've been thinking about it for a while)
It would be really great to have some sort of rating for most EA projects. This would include:
It's not clear what the units should be. Some ideas include:
In the beginning, these estimates could be made by a small team. Later on, they could be made by forecasters, predicting what evaluators would say.