For various reasons relating to technology and history, Esri now has two major ways of storing vector datasets: geodatabase, and shapefile. Some of the differences relate to functionality and some mostly to terminology. Shapefiles and geodatabases have a lot in common. Geodatabases and shapefiles allow multipoints. They both delineate polygons with rings. They use polylines, composed of segments, with one or more paths. The similarities end there. Geodatabase segments may be circular arcs, elliptical arcs, or Bézier curves. Geodatabases allow for topological relationships; shapefiles do not. A major feature of geodatabases is the ability to use a geometric network (composed of junctions and edges that possess not only attributes, but behavior). Geodatabases calculate the lengths of lines and rings, and the areas of polygons—and shapefiles do none of that. Geodatabases allow subtypes of features, while shapefiles do not. So, the semi-witty comparison that a geodatabase is a “shapefile on steroids” greatly understates the case.