Urban geographers have developed sophisticated methods to analyse segregation in modern-day mixed cities. This paper applies their techniques to turn-of-the-century Alexandria. This mixed city has various measures of its diversity. Census records provide the baseline dataset for this study. The decennial censuses of 1897, 1907, and 1907 parse the city into 100 street clusters, and offer detailed demographics of each of these clusters. This paper shows how these vaguely defined clusters can be mapped in space. It considers the nationality and religion labels attached to these populations and seeks to determine how Alexandria's population distribution looks in comparison to modern-day measures of segregation. For instance, the granularity of segregation is a current measure of interest. Can existing data reveal whether the supposedly mixed city of Alexandria was in fact a site of segregation? The well-known model of traditional Ottoman-Islamic cities suggests that self-contained, socially homogenous neighborhoods were a common means of spatial organization. Does a rapidly expanding city like Alexandria maintain or overrun this model? Does segregation at the turn of the twentieth century map differently from segregation in the early modern city? Beyond this comprehensive view of Alexandria's segregation, the paper examines more punctual evidence of diversity in space in particular sites. This evidence comes from a student-driven digitization of the Egyptian Gazette, a daily newspaper in Alexandria, for the years 1905-1907. The contents of this newspaper have been marked for person and place identifiers, and will be minded for comparison with the baseline maps of diversity and segregation. The paper will also draw on a database of prominent individuals generated from the Indicateur Egyptien of 1897 and the Goad fire insurance maps of 1905. These two supplemental data sources do not describe the city's peripheral regions, but they offer a great deal of microhistorical data concerning its central districts. Finally, the paper examines the problem of workable GIS for Middle Eastern cities. The author is not a GIS specialists. Instead, I seek a simple means of leveraging geo-spatial historical data, especially for classroom use. I will describe how I produced "pretty good" mapping by linking former street and place names to modern-day, user-friendly indices.