It is often said that one geophysicist's noise is another's data, but today it seems that it has never been more the case. Features of seismic records that once were considered noise are now routinely used to aid in imaging the Earth's interior and in deriving rock properties. Sacrosanct noise such as surface waves is now inverted to derive near-surface properties. Multiple reflections, once the interpreter's bane, have long been estimated and removed but now are being used in the imaging process. Even seismic interference is now actively pursued as a source of new data in the guise of simultaneous-source acquisition. And then there is the emerging use of passive seismic data, which once would have been considered as pure noise.