Various underdetermination arguments against selective realism are explored and rebutted. The role of abstraction and idealization in templates is explored. The fact that most computational processes are epistemically opaque is emphasized, and the effect that opacity has on scientific knowledge is explained using the Ising model as an example. Arguments are given for emphasizing mathematical form over logical form in the philosophy of science, and for emphasizing methods applicable in practice over those applicable only in principle.
Keywords: abstraction, idealization, Ising model, logical form, mathematical form, methods, realism, scientific knowledge, template, underdetermination