1) Why would a “perfectly logical being” compute (do) X and not Y? Do all “perfectly logical beings” do the same thing? (Dan’s comment: a system that computes your answer determines that answer, given a question. If you presuppose an unique answer, you need to sufficiently restrict the question (and the system). A universal computer will execute any program (question) to produce its output (answer).) All “beings” won’t do exactly the same thing, answer any question in exactly the same way. See also: No Universally Compelling Arguments.
1) Why would a “perfectly logical being” compute (do) X and not Y? Do all “perfectly logical beings” do the same thing? (Dan’s comment: a system that computes your answer determines that answer, given a question. If you presuppose an unique answer, you need to sufficiently restrict the question (and the system). A universal computer will execute any program (question) to produce its output (answer).) All “beings” won’t do exactly the same thing, answer any question in exactly the same way. See also: No Universally Compelling Arguments.
2) Why would you be interested in what the “perfectly logical being” does? No matter what argument you are given, it is you that decides whether to accept it. See also: Where Recursive Justification Hits Bottom, Paperclip maximizer, and more generally Metaethics sequence.
2.5) What humans want (and you in particular), is a very detailed notion, one that won’t automatically appear from a question that doesn’t already include all that detail. And every bit of that detail is incredibly important to get right, even though its form isn’t fixed in human image.