I'm not sure what you mean by "vanilla anthropics".
Am working on it - as a placeholder, for many problems, one can use Stuart Armstrong's proposed algorithm of finding the best strategy according to a non-anthropic viewpoint that adds the utilities of different copies of you, and then doing what that strategy says.
Both SSA and SIA are "simple object-level rules for assigning anthropic probabilities"
Yup. Don't trust them outside their respective ranges of validity.
if you assume [stuff about the nature of the universe]
You will predict [consequences of those assumptions, including anthropic consequences]. However, before assuming [stuff about the universe], you should have [observational data supporting that stuff].
Am working on it - as a placeholder, for many problems, one can use Stuart Armstrong's proposed algorithm of finding the best strategy according to a non-anthropic viewpoint that adds the utilities of different copies of you, and then doing what that strategy says.
I think this essentially leads to SIA. Since you're adding utilities over different copies of you, it follows that you care more about universes in which there are more copies of you. So your copies should behave as if they anticipate the probability of being in a universe containing lots of...
Nick Bostrom's self-sampling assumption treats us as a random sample from a set of observers, but this framework raises several paradoxes. Instead, why not treat the stuff we observe to be a random sample from the set of all stuff that exists? I elaborate on this proposal in a new essay subsection: "SSA on physics rather than observers?" At first glance, it seems to work better than any of the mainstream schools of anthropics. Comments are welcome.
Has this idea been suggested before? I noticed that Robin Hanson proffered something similar way back in 1998 (four years before Bostrom's Anthropic Bias). I'm surprised Hanson's proposal hasn't received more attention in the academic literature.