I have worked in GIS for well over fifteen years: primarily at Penn State and in Little Rock, Arkansas, and now in Toronto. I’ve been involved in teaching, analysis, operations and management within the field. Yet as soon as I mention to people what I do, I see a questioning look which glazes over when I start to explain. What are those three letters? I tell them I’m a geographer, and I work with maps: and the immediate connection is to cartography. And I do make maps… well, sort of. With the advent of Google Maps and Google Earth I can see another connection in my friends’ minds. But such visualization, though very cool, is only the smallest part of what I do. So I hope that I can provide a resource to people who want to know what GIS really is, how to use it and how to make connections in the industry. Although geared toward Toronto and Ontario, most of my experience has been in the States; specifics may vary but the principles remain the same.
Geographic Information Systems.
Geographic Information Science.
These are two interpretations for those three letters: and they don’t really say that much. I think they say more about the people who use them than the actual field itself. GIS is a tool, a method, for geographic concepts and ideas, and representing them in such a way that a computer can analyze them: over space. There is actually a certain irony embedded in here somewhere, because these two faces of GIS are so different. GIS is the marriage of a highly technical field (one related to computers, data systems and the Internet) to the more non-technical study of geography (related to places, landforms and social sciences). When I have taught previously, I discovered that one of the common reasons that people pursued geography as an academic profession was try to avoid math. Geography is more fuzzy, more personal; there is an entire branch of the field known as “social geography”. Yet I enjoyed geography precisely because I was good at math. I can see relationships in space between two objects: I can see the angles and the trig and the distances. GIS is a method of taking these relationships: the characteristics of an object in space and exploring how it relates to other objects around it… all in a computer, or a model.
At the centre of GIS is the description, or characteristics, of different objects in the world: of a variety of types and accuracies: but it is all tied to a location. The information itself is stored in a database, and each record is intended to represent a particular object in the real world. Those objects in space can be small and simple: a post, a car, a book; or they can large and complex: a city, a river, a national boundary. Descriptions become “attributes” and locations become “coordinates”, both occupying a particular “place” in the mathematical bits and bytes of the computer’s memory. Alongside the spatial information the attributes can be precise or more general: its name, size and anything else that might be useful. These two halves, the location and the description, make a GIS whole.
What makes GIS a tool is being able to analyze the relationships between these objects in the database. The tool is able to combine the information in a database: the attributes fused with geographic features, and perform operations on them. From simple to complex, it can perform spatial statistics and determine dependencies over space that are otherwise buried in the standard database structure. The computer is effectively able to understand concepts like: beside, near, far, contiguous, adjacent. Each can have very different meanings when considering analysis, and will be applied in different circumstances.
This is one of the reasons that I maintain that Google Earth and Google Maps are not real GIS products. They are great for visualization, certainly: and they’ve raised the bar in the popular mind so that more complex GIS tools have a lot to live up to. But in my mind one of the problems with the Google products is that they do very little analysis. You can’t actually figure very much with them. You can’t find very much with them at all. Even a simple query like finding everything within 2km of a particular address has no meaning in Google. Oh, you can find them, certainly: you can find the address and use the measure tool to draw a circle around the point and you can identify everything you can see. But a GIS can do that for you, and in extreme circumstances you don’t want to waste time figuring it out for yourself. If you’re dealing with emergency management and you’ve had an explosion at a propane facility (as happened in Toronto in 2008) and you want to evacuate everyone within 2km, a GIS can not only identify all the streets that will need to be evacuated, but can tell you individual addresses, filtered by residential, commercial and apartment complexes, including names and contact information. It can do so not only in a form that people can understand and respond to, but in a way that can automate responses and take advantage of a computer’s speed. GIS takes a fair bit of knowledge and skill to set up correctly, but once done it saves time and makes things more efficient… and because almost everything in the world is spatial, it allows us to see how different datasets relate to each other.
This post was originally published on Gather.com, and is reproduced here (primarily to try to keep my writing in one spot).