These are personal recollections and musings, written for the 50th Anniversary of the Journal of Magnetic Resonance. They are distilled from twenty years of writing simulation code, and filtered through hindsight and sarcastic intransigence. To me, the biggest recent achievements of the magnetic resonance community in the field of theory and simulation have been the successful war on the exponential scaling, the powerful and general simulation software, and the return to elegant notation. It appears that our future will be defined by computers. Three aspects are pertinent: simulation as the experiment is designed, optimal control as the experiment proceeds, and machine learning at the data processing stage.