Since the discovery of spherical aberration associated with the primary mirror, the application of image restoration techniques to HST imaging data has become an almost routine procedure. It is not my intention here to argue for or against the use of these procedures generally, but simply to highlight certain features relating to the Faint Object Camera and the data it produces and how these features should influence an observer when designing observing proposals. If nothing else, this discussion may help those less experienced image restorers to understand some of the limitations of the techniques. The main sources of information for FOC Users should be the Faint Object Camera Instrument Handbook, Version 3.0 (Pre-COSTAR) (Paresce 1992), and Version 4.0 (Post-COSTAR) (Nota et al. 1993). Additionally, our accumulated experiences over the last three years are published in ``In-Flight Performance of the Faint Object Camera of the HST: II'' by Greenfield et al. (1993).
Clearly, for each individual user, the relative importance of these different effects will depend very much on the quality of their data, the type of target being observed, and the science they are trying to do, and therefore the following topics are addressed in no particular order. Readers should not attempt to infer any significance regarding their relative importance from the order in which they are presented. It should be noted that most of these effects are not independent in that their existence often influences, or is influenced by, other considerations. Note also that unless stated otherwise, all of our discussion is slanted towards the F/96 relay since this is the most used. We present a typical example in § 2. In the course of each discussion we will comment on how each of the problems will be affected by the deployment of COSTAR.