Exploring eye tracking in the context of spatial information, a key consideration is gazebased interaction. Geographic gaze-based recommender systems, for example, aim to provide an appropriate response to the user's gaze in geographic contexts, e.g., when looking at a particular location in physical or virtual space, on the screen or in 3D. Analysis of the viewer's gaze beyond fixation detection is required to infer the viewer's desire for information. Several approaches to developing an interest metric have been investigated. I will briefly review some advanced techniques for 3D gaze analysis, past interest metric efforts, and more recent developments of interaction styles.