The determination of the optical point spread function (PSF) or a model thereof is one key step in the estimation of key astronomical quantities for most science cases. Yet it has proven quite challenging due to crowding or the total absence of point-sources in the field or variability (time, angle and wavelength). For Adaptive-optics (AO) assisted observations, alternative techniques exist such as PSF reconstruction (PSF-R) which relies on the AO control loop data. Our goal is to provide a standardized nomenclature and categorization of the techniques that use focal-plane data (numerical extraction, parametric model-fitting), recovery from telemetry (reconstruction, analytic or Monte Carlo modeling) or both jointly combined (hybrid, deconvolution) in an attempt to gain insight into the advantages and shortcomings of such techniques. Applicability to AO systems on Giant Segmented Mirror Telescopes (GSMT) is our main motivation.