The earthquake source problem is characterized by the complexity coming from repeated interactions between many elements. The one-earthquake or one-fault problem is usually thought to be relatively well-understood and the excitement emerges when coupling (via long-range elasticity and relaxation processes) many faults presenting highly nonlinear responses (threshold dynamics) in the presence of rock heterogeneity and fault geometrical complexity. This is the vision that led us to propose an analogy between earthquakes deformations and self-organized criticality about a decade ago and to the recognition that earthquake dynamics offer for the solid earth a complexity and richness (and difficulties!) equivalent or even greater to that of hydrodynamic turbulence (usually considered to be one of the most important and difficult unsolved challenges nowadays).