In 1952, Overhauser proposed a method for transferring the large electron spin polarization of conduction electrons in a metal to the metal nuclei, enhancing their polarization one thousand fold. Such an enhancement method is called dynamic nuclear polarization (DNP). The article explains his idea, describes the experiments of Carver and Slichter that confirmed his proposal, their demonstration that the method was not limited to metals, describes the nature of immediate impact on the magnetic resonance community, discusses why DNP was not broadly utilized for many years, explains in simple terms how the method works and reports on the new developments in experimental methods that have given DNP an active and exciting future.