Admissible estimation of multivariate normal mean

The James-Stein (1961) estimate of a multivariate normal mean has spawned a
spectrum of related admissibility and inadmissibility results. Nevertheless it has
not had much impact at the grass-roots level of applied statistics and continues to
be almost totally ignored (after almost a half century) in introductory textbooks.
Possible reasons for its low pro le are discussed.