“…To begin, possible worlds have proven useful in a large variety of philosophical projects; consequently, many will regard the appeal to possible worlds in a theory of information as previously justified by their other fruits. In any case, commitment to possible worlds is only ontologically costly given inflationary understandings of possible worlds (e.g., Lewis, 1986) -and, of course, there are deflationary (e.g., fictionalist, abstractionist, instrumentalist) alternatives available (e.g., Rosen, 1990;Armstrong, 1989;van Inwagen, 1986;van Fraassen, 1980;Forbes, 1983;Yablo, 2001). Hence, we don't see that the counterfactual account, by being committed to possible worlds, carries onerous special burdens.…”