A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
Geographic Information Systems (GIS) are computational systems that capture, store, analyze, and visualize spatial data, enabling geographers to overlay multiple data layers and detect spatial patterns invisible in tabular data. GIS represents the world as layers of vector (points, lines, polygons) or raster (grid) data, each carrying associated attribute information. Map projections are mathematical transformations of the spherical Earth onto flat surfaces; all projections introduce distortions of area, shape, distance, or direction, and the choice of projection is never a neutral technical decision. Remote sensing — data acquisition from satellites and aircraft — provides much of the raw imagery underlying GIS layers. The proliferation of GPS, open spatial datasets, and web mapping platforms has democratized spatial analysis but also raised concerns about surveillance and data privacy.
Use free GIS tools (QGIS, Google Earth Engine) to overlay demographic and environmental data layers and identify spatial correlations. Compare the Mercator, Robinson, and Peters projections and analyze what each distorts and preserves — and what political implications each carries. Practice reading choropleth, dot density, and proportional symbol maps to understand how design choices shape viewer interpretation.
A Geographic Information System is, at its core, a way of layering questions onto space. When you look at a traditional map, you see one representation of the world frozen at a moment in time. GIS replaces that single map with dozens of data layers — roads, population, rainfall, land use, pollution readings — that can be combined, queried, and analyzed computationally. The power of GIS is not just display; it is the ability to ask spatial questions: Where do high poverty rates and low supermarket access overlap? Which neighborhoods are within 500 meters of a toxic facility? How has the urban footprint of a city expanded over decades?
Data in GIS comes in two fundamental formats. Vector data represents the world as discrete geometric features — points (a hospital, a well), lines (a road, a river), or polygons (a country boundary, a census tract) — each carrying a table of attributes. Raster data divides space into a grid of equally sized cells, with each cell holding a value: a satellite image is a raster where each pixel records spectral reflectance; an elevation model is a raster where each cell records height above sea level. Neither format is inherently superior — you choose based on what you are modeling. Discrete, bounded objects are vector territory; continuous phenomena that vary across space are raster territory.
Map projections are the unavoidable compromise underlying all GIS work. The Earth is a sphere; a map is flat. Converting one to the other always distorts at least one of four properties: area, shape, distance, or direction. The Mercator projection — developed in 1569 for navigation — preserves direction perfectly (straight lines are true compass bearings) but catastrophically inflates area at high latitudes. Greenland appears roughly the same size as Africa, though Africa is nearly 14 times larger. The choice of projection is a design decision with real consequences for how viewers perceive the relative importance of different regions. There is no neutral choice.
The most critical thinking skill in GIS is resisting the visual persuasiveness of maps. When two spatial patterns overlap on a map, the temptation to infer causation is powerful — the patterns look like they "go together." But spatial correlation is not causation. Two variables can co-vary across space because they share a common cause, because they are both products of the same historical process, or even by chance at the scale you are examining (the modifiable areal unit problem: your results change depending on how you draw the boundaries). GIS surfaces patterns worth investigating; it does not supply explanations. The theoretical work remains yours.
Topics in reflective domains aren't scored by quiz answers. Read, reflect, and mark when you've thought it through.