Randomizing works just as well as diversifying. Since the correlation between people’s decisions is far from perfect, it’s effectively randomized. No need to do anything yourself.
Alternatively, if you do find yourself in a group with similar preferences to you, you can collude. Failing that, you can assign your entire donation to an individual charity, choosing randomly weighted by your priority, and thus reduce the impact of transaction fees and other per-donor costs on your donations (such as marketing materials to encourage people to donate again).
My decision will correlate with other people’s, but there will still be enough variation that it’s not likely to result in a charity getting more money than they know what to do with.
Like DanielLC, I never really understood this application of Timeless Decision Theory. It seems that, in practice, people’s decisions rarely seem to correlate with mine in any sort of meaningful way such that if I chose to change my action, so would they.
Randomizing works just as well as diversifying. Since the correlation between people’s decisions is far from perfect, it’s effectively randomized. No need to do anything yourself.
Alternatively, if you do find yourself in a group with similar preferences to you, you can collude. Failing that, you can assign your entire donation to an individual charity, choosing randomly weighted by your priority, and thus reduce the impact of transaction fees and other per-donor costs on your donations (such as marketing materials to encourage people to donate again).
I don’t follow, can you elaborate?
My decision will correlate with other people’s, but there will still be enough variation that it’s not likely to result in a charity getting more money than they know what to do with.
Like DanielLC, I never really understood this application of Timeless Decision Theory. It seems that, in practice, people’s decisions rarely seem to correlate with mine in any sort of meaningful way such that if I chose to change my action, so would they.