- August 26, 2011
I was browsing one of my favorite blogs, Flowing Data, this morning, when I stumbled upon Derek Watkins. Derek is a graduate student in Geography at the University of Oregon who recently started a blog about his current research, which explores some ways that US borderlands extend into cyberspace, and how online representations might affect people’s thoughts about the region. What I find most compelling is the simplicity and universality of the questions that Derek attempts to answer through publicly available geo-tagged data.
For example, in the below visualization, Watkins addresses the question of perception of a place. Why do we think of Detroit as an icon of urban decay and stagnation when so many of us have never even visited the city? Our perception of the world is deeply influenced by the thoughts and opinions of others, particularly when they are widely shared and therefore, reinforced in the collective consciousness.
As Watkins points out in his original post, the relationship between perception and reality is complex and perhaps a map of geo-tagged Flickr photos may not tell the whole story of a place. But, oh, the beauty of adjacent data. In a very useful update, Watkins links to a post from Data Pointed, that maps population shifts over the past 10 years. The one below is of the Detroit metro area.
In urban areas, deep blue indicates that the population doubled (or more), pure red means that everyone left, grey denotes no change, and the intermediate tones represent the spectrum of increases and decreases in-between. Below 5000 residents per square mile, these colors fade with the square root of density towards white, where no people lived in either year.
So while it’s true that Detroit is experiencing huge population decline, it does not tell the whole story about this troubled city. The outlying areas and even some spots in downtown Detroit are experiencing surprisingly high growth. Why would the suburbs and the urban heart of a city in decline be experiencing such a boom? Perhaps we just need more data.