Since its initial development in the early 1970s, spectroscopic ellipsometry (SE) has evolved to become the primary technique for determining the intrinsic and structural properties of homogeneous and inhomogeneous materials in bulk and thin-film form, including properties of surfaces and interfaces. As an indispensible nondestructive approach for determining critical dimensions in integrated-circuit technology, its economic impact has been enormous. I review the development of theory as well as instrumentation, from the perspective of someone who has worked in SE essentially from its beginning. I provide comments about its present state, note some unresolved issues, then consider possible improvements and predict how the field is likely to evolve. (C) 2014 Elsevier B.V. All rights reserved.