This study introduces a comprehensive representation of the public domain in urban space, discusses the role of machine learning techniques in such a representation, and presents a preliminary experiment. We define public domain as all places that people actually perceive as public space traditionally identified spaces such as streets, squares, parks, as well as privately managed collective spaces that function as public space. In representing this broad definition of public domain, we suggest that the vast amount of information available to designers can be managed with our proposed data model. We further suggest that interpreting the public domain requires representing experiential aspects, which few methods have attempted. We illustrate our experiential representation by discussing the concept that we call a Heuristic Nolli Map. We discuss its use in identifying characteristics that contribute to the “public-ness” of a common public domain type-main street, a typical linear commercial district in the centre of a residential area. We suggest that a data model of main street contain the GIS data that cities in the United States generally maintain: land parcel, building footprint, and street centre line. We also suggest that the data model contain additional data at a finer level of detail, e.g., tax assessment data. Finally, we describe our Heuristic Nolli Map methodology in terms of two steps: collecting opinion data about a user’s interpretation of public-ness, and using that data to build a user’s model of public domain. The model consists of a typology of public domain, e.g., characteristics of main streets, department stores, town centers; a classifier that employs machine learning techniques to interpret the public-ness of user-supplied data models; and a component that provides explanations for results of classifying particular data models. Finally, we discuss expected contributions of this research and the current status of the research in progress. The full file of this article can be downloaded for free