A problem with this approach is that the ordering of the things in the sequence matters ((1,0,1,0,1...) reorders to (1,0,0,1,0,0,1...)). This method works here, where the ordering is by moments of time, but not for, say, summing the utility of infintely many agents, where there is no clear ordering.
Thanks! Your idea is interesting – I put a comment on that post.
Something you are probably aware of is that accepting “anonymity” (allowing the sequence to be reordered arbitrarily) requires us to reject seemingly intuitive principles like Pareto (if you can make someone better off and no one worse off, then you should).
Personally, I would rather keep Pareto than anonymity, but I think it’s cool to explore what anonymous orderings can do.
A problem with this approach is that the ordering of the things in the sequence matters ((1,0,1,0,1...) reorders to (1,0,0,1,0,0,1...)). This method works here, where the ordering is by moments of time, but not for, say, summing the utility of infintely many agents, where there is no clear ordering.
I have a method of comparison that doesn’t depend on the ordering: https://agentfoundations.org/item?id=1455
Thanks! Your idea is interesting – I put a comment on that post.
Something you are probably aware of is that accepting “anonymity” (allowing the sequence to be reordered arbitrarily) requires us to reject seemingly intuitive principles like Pareto (if you can make someone better off and no one worse off, then you should).
Personally, I would rather keep Pareto than anonymity, but I think it’s cool to explore what anonymous orderings can do.