This essay outlines my interactions with Alan Wilson over more than four decades, focusing on my early work on gravity models and then my adoption of his entropy-maximizing procedure in the study of disaggregated voting patterns. Using that example, the essay explores its implications for the study of disciplinary history.