This is probably too late a response to ever be read, but with regard to the latter point, Kripke did not have to invent the 'quus' operation , as there are act

This is probably too late a response to ever be read, but with regard to the latter point, Kripke did not have to invent the 'quus' operation , as there are actual examples to be found, for example in modular versus conventional arithmetic, real-valued versus complex-valued functions, etc. These cases do not create any unavoidable problems; we simply disambiguate where necessary. It is a non-issue unless your philosophical intuition depends on the semantics of words being something more fundamental than a convention among the users of a language. As for your first claim, it appears to be begging the question, predefining, without justification, 'real' computation as being something that only humans can do, while machines can only simulate so doing.