The world is nonlinear and linked. In this talk, I will present the origin of the small-gain theory and show that it serves as an important systematic tool for addressing two fundamental problems for networks: When is a dynamical network robustly stable? When can a dynamical network be made robustly stable by feedback? As an illustration, we show how the small-gain theory can be applied to unify solutions to event-triggered nonlinear control and provide novel solutions to distributed feedback optimization. Finally, if time permits, I will discuss briefly learning-based control, a new direction in control theory, that aims to integrate machine learning and nonlinear control techniques to relax the conservativeness of small-gain designs.