Hacks 64-77

As geographers... crowd into the edges of their maps parts of the world which they do not know about, adding notes in the margin to the effect that beyond this lies nothing but sandy deserts full of wild beasts, and unapproachable bogs.

Plutarch, The Life of Theseus

At the far frontier of the world of maps, we find the province of Geographic Information Systems. Geographic Information Systems, better known as GIS, come in a number of shapes and sizes, but they all focus on the acquisition, management, analysis, and presentation of data with a spatial component. GIS tools are used by local and national governments around the world to collect and analyze the demographic patterns of their citizens and to plan roads, parks, and other civic infrastructure; by public utilities to record and plan telephone, power, water, gas, and sewer-line deployment and maintenance; by scientists to gather and interpret experimental data about our planet and the multitude of creatures that live on it; and by commercial enterprises for sales analysis and to plan the manufacture, warehousing, and shipping of goods all around the world.

It used to be the case that GIS tools were only available to academics and professionals at great expense, but the explosion of personal computing on commodity hardware and the revolution of open source software have combined to make Geographic Information Systems accessible to anyone with the patience to download a GIS tool and learn how it works. Not, mind you, that this is a trivial task to accomplish, but, if you're at all like us, we bet you'll be hooked from the instant you render your first map in GRASS or Quantum GIS and get that first glimpse of the possibilities inherent in GIS-based locative storytelling.

GIS is more than mere cartography, of course. The concept of Geographic Information Systems implies the ability to perform analysis on spatially connected data sets, to askand answerquestions with a geographic component. What is the median income of the residents of this area? How many earthquakes occurred in this region in the last decade? Which parts of the river will flood when it rains? Where is the best place to put a park, a gas station, a day care, a stadium, an airport? What endangered species or fragile ecosystems will be affected if we do so? All these questions, and many more like them, are approachable with a GIS and the right data sets.

Before we dive into the realm of GIS and desktop mapping, let's take a moment to consider what kinds of data are useful in Geographic Information Systems. Broadly, the two basic types of GIS data you'll find are raster data and vector data.

Raster data typically consists of a layer of cells laid out in a grid, just like a bitmap image, except that the raster layer corresponds to some explicit region of Earth's surface. Each cell is rather like the pixels of a bitmap, in that a numeric value is associated with each cell, representing something about the corresponding place. If the raster layer represents elevation, then the cell values might be meters above sea level. If the raster layer depicts ecosystems, then the cell values might correspond to ecosystem types. Alternatively, if the raster layer contains a scanned hand-drawn map or an aerial photo, then the cell might actually be a bitmap pixel!

In contrast, vector data is used to represent the geometry of place. Individual locations, such as archaeological sites and wireless network nodes, are often described as a point, with a single coordinate pair. Linear features, such as roads and rivers, are often depicted as a connected series of points, forming a line. Finally, sequences of connected line segments can form polygons, which are used to represent things that have physical boundaries, such as parks, municipal boundaries, national borders, and coastlines. With these three geometric primitivespoints, lines, and polygonsnearly every aspect of our world can be cataloged and depicted at any scale.

Geography students are often taught that "raster is faster, but vector is corrector." With that in mind, let's get started!

A word to Linux users about obtaining GRASS and other bits of open source GIS software: if you're running Debian GNU/Linux, you should add the DebianGis repository to your /etc/apt/sources.list. You can find out more about DebianGis at http://pkg-grass.alioth.debian.org/cgi-bin/wiki.pl. If you're running Fedora Core or any other RPM-based distribution (Mandrake, SuSE, etc.), check out the Mapping Hacks repository of GIS packages at http://mappinghacks.com/rpm.