Where do you want to go today [ in attribute space ] ?

This paper proposes to conceptualize a geographic trajectory as an attribute-time path (ATP) traversing an n-dimensional attribute space. Geographic entities occupying a certain study area are conceptualized as discrete loci in that same n-dimensional space. Those entities undergo spatialization and the resulting point locations become vertices as the ATP is projected into lower dimensions to form a spatialized attribute-time path (SATP). Visual exploration of such a path supports novel perspectives on the nature of geographic space, including how it is traversed, transformed, and experienced by those that inhabit it. Examples based on two different data sets are presented, with population census data providing n-dimensional attributes of geographic entities in both cases. The first study illustrates how a transect of Vienna, Austria, reveals patterns of difference/similarity, once it is spatialized. The second set of examples is based on an extremely large neural network model derived from data for all 200,000+ U.S. census block groups. Commuter trajectories captured in New Orleans and San Diego are projected onto the spatialized block groups and the resulting SATPs interpreted.