Building the Salem Witchcraft GIS


Appendix A

About GIS Data

Q. Why use GIS software, which is designed to create highly accurate models of the geographic world, to model a printed map of doubtful accuracy?

A. Because GIS software provides us with a way to visualize data stored in a relational database system. This gives us more powerful analytic tools than is typically brought to bear on historical materials, and certainly on this subject.

GIS software was first developed by the Canadian government in the 1960s to inventory the vast land holdings across that nation. They needed a highly accurate way of automating information tied to geography. In recent years, GIS has grown rapidly and tremendously. It's major users are still those who manage or survey physical features and infrastructures: engineers, scientists, governments, facilities coordinators. GIS allows these users to visually organize and spatially query vast amounts of data to answer questions more rapidly and readily. Only recently have historians and social scientists begun to explore the power of GIS to study demography and temporal change in the physical world.

GIS data typically represents the world through a combination of vector graphics and tabular data, stored as a series of layers. Each layer represents only one type of feature, but may store many instances of that feature, each of which is represented by a record in a "feature attribute table." Features can be represented as points, for discrete features, lines, for linear features, or polygons for area features.

For example, if you were to model the road network in your hometown, you would probably include a layer that represented the streets. Each line in the layer would correspond to a street, and each line would have information about it--like the name of the street it represents, the addresses on that street--in the associated feature attribute table. If you wanted to show where the stoplights in town were, you might use points to represent this information: each point would overlay on top of the street layer, and should probably correspond to the street intersections. As with the street layer, the intersection layer would have associated information--perhaps each record would include not only the cross streets, but also the last time the lights were replaced at that stoplight.

The process of creating such GIS data layers from a printed source is called "digitization." Essentially, this involves tracing the points, lines and polygons of a map using the GIS software. To do this, one needs a set of known control points, or "tics" which would be digitized and used to anchor the rest of the digitizing. Without tics, it is difficult to impossible to assign real world latitude and longitude values to the data.

Once digitized, the GIS layer, or "coverage" can be associated with sets of data recorded in other ways.

For more information about basic GIS concepts, visit a tutorial written by the ESRI Corporation.