Apparently a stable balance exists between low temperature hydrogenation and oxidation reactions on grain surfaces. In contrast, an equally prominent ice component, consisting almost entirely of CO, must have accreted directly from the gas phase. Thermal processing, i.e. evaporation and crystallization, proves to be readily traceable in both these ice components. The spectroscopic signatures of energetic processing by cosmic rays and high energy photons from nearby protostars are weaker and not as well understood. A fundamental limitation in detecting complex, energetically produced (and also some simple) species is blending of weak features in the spectra of protostars. Sophisticated techniques are required to extract information from blended features. We conclude with a summary of key goals for future research and prospects for observations of ices using future instrumentation, including SIRTF/IRS.