GIS (geographic information systems) arose out of the need to perform spatial queries on geographic data. A spatial query requires knowledge of locations as well as attributes. For example, an environmental analyst might want to know which public drinking water sources are located within one mile of a known toxic chemical spill. Or, a planner might be called upon to identify property parcels located in areas that are subject to flooding. To accommodate geographic data and spatial queries, database management systems need to be integrated with mapping systems. Until about 1990, most maps were printed from handmade drawings or engravings. Geographic data produced by draftspersons consisted of graphic marks inscribed on paper or film. To this day, most of the lines that appear on topographic maps published by the U.S. Geological Survey were originally engraved by hand. The place names shown on the maps were affixed with tweezers, one word at a time. Needless to say, such maps were expensive to create and to keep up to date. Computerization of the mapmaking process had obvious appeal.
Computer-aided design (CAD) CAD systems were originally developed for engineers, architects, and other design professionals who needed more efficient means to create and revise precise drawings of machine parts, construction plans, and the like. In the 1980s, mapmakers began to adopt CAD in place of traditional map drafting. CAD operators encode the locations and extents of roads, streams, boundaries, and other entities by tracing maps mounted on electronic drafting tables, or by key-entering location coordinates, angles, and distances. Instead of graphic features, CAD data consist of digital features, each of which is composed of a set of point locations. Calculations of distances, areas, and volumes can easily be automated once features are digitized. Unfortunately, CAD systems typically do not encode data in forms that support spatial queries. In 1988, a geographer named David Cowen illustrated the benefits and shortcomings of CAD for spatial decision making. He pointed out that a CAD system would be useful for depicting the streets, property parcel boundaries, and building footprints of a residential subdevelopment. A CAD operator could point to a particular parcel, and highlight it with a selected color or pattern. "A typical CAD system," Cowen observed, "could not automatically shade each parcel based on values in an assessor's database containing information regarding ownership, usage, or value, however." A CAD system would be of limited use to someone who had to make decisions about land use policy or tax assessment.
Desktop mapping An evolutionary stage in the development of GIS, desktop mapping systems like Atlas*GIS combined some of the capabilities of CAD systems with rudimentary linkages between location data and attribute data. A desktop mapping system user could produce a map in which property parcels are automatically colored according to various categories of property values, for example. Furthermore, if property value categories were redefined, the map's appearance could be updated automatically. Some desktop mapping systems even supported simple queries that allow users to retrieve records from a single attribute file. Most real-world decisions require more sophisticated queries involving multiple data files. That's where real GIS comes in.
Geographic information systems (GIS) As stated earlier, information systems assist decision makers by enabling them to transform data into useful information. GIS specializes in helping users transform geographic data into geographic information. David Cowen (1988) defined GIS as a decision support tool that combines the attribute data handling capabilities of relational database management systems with the spatial data handling capabilities of CAD and desktop mapping systems. In particular, GIS enables decision makers to identify locations or routes whose attributes match multiple criteria, even though entities and attributes may be encoded in many different data files.
Innovators in many fields, including engineers, computer scientists, geographers, and others, started developing digital mapping and CAD systems in the 1950s and 60s. One of the first challenges they faced was to convert the graphical data stored on paper maps into digital data that could be stored in, and processed by, digital computers. Several different approaches to representing locations and extents in digital form were developed. The two predominant representation strategies are known as "vector" and "raster."