Abstract: In this talk, we illustrate that computational issues can be typically tackled in four different ways, namely, (a) starting at the right point (like Shannon’s digitalization, Aryabatta’s zero, Turing’s universal machine etc.), (b) understanding and re-cycling the run-time uncertainties (like randomization, robustness, quantum uncertainty etc.), (c) flowing with finesse (like generality, feasibility, optimality etc.) and (d) stopping well-in-advance (like super-resolution, proactiviation, pseudorandomness etc.). For each of the four ways, we discuss famous examples that used them to solve important problems in the past and also speculate about what problems of the future could be similarly addressed.