The rate of progress in increasing the data storage density (b/mm
2) in magnetic recording is decreasing principally because of two difficult technical problems. These are: 1) the achievement reliably and repeatably, head-medium spacings of fractions of a micrometer that densities of

b/mm require, 2) reduction of track width ∼ bit length. Writing and reading with focussed light beams solves both problems and when combined with magnetic storage of data, the resulting technology has the strengths of optics and magnetics. This is achieved in Thermo-Magnetic Optic (TMO) recording on rare-earth, transition metal alloys on which data can be written, erased and re-written repeatedly at densities

b/mm
2. The question is "do these materials match conventional magnetic recording media in the life expectancy of the storage medium?" This calls for accelerated life testing with its inevitable extrapolations. The principal, potential failure mechanisms have been identified as the migration and oxidation of the rare-earth and the relaxation of the perpendicular orientation of the films. The evidence for these mechanisms will be reviewed and the conclusion is reached that films of the correct composition, properly undercoated and overcoated and deposited on the right substrate have a probable life expectancy of at least ten years.