Geography, Big Data, and Augmented Realities

1 August 2012 Oxford Internet Institute

Cities are no longer just made of bricks and mortar. They are no longer just confined to their material presences. They have become both digital and digitised. Because of the advent of the Geoweb, our experiences of cities are now intertwined with digital representations of those same places: layers of code, algorithms, photographs, reviews, videos, descriptions, and status updates.

Virtual layers in the cityThese new digital dimensions of place now profoundly matter for the ways that we interact with our urban environments. It is because of these changes, that OII Research Fellow, Dr Mark Graham, leads a research project to interrogate these virtual layers of the city. His work asks what they are, where they are, and why they matter.

Such interrogations of digital urban augmentations can be carried out in a range of ways. Ethnographies, surveys, and interviews of and with the creators and users of augmented content would all serve to elucidate some of the most important economic, social, and political implications of our increasingly hybrid urban environments.

However, because of the digital nature of urban data shadows and augmentations, it is now also possible to ask entirely new types of questions. The 'big data' surrounding many of our planet's urban environments can be mapped, measured, and analysed. It is in these areas that Dr Graham's work focuses. Specifically, his work on 'big data' and geography seeks to examine a few central questions.

First, it asks what the geographies of 'big data' are. In other words, which parts of the world are covered in dense layers of information, and why do we see such uneven geographies? The geographies of layers of information tell us much about how some places can be rendered absent or invisible in global conversations and representations.

One example of this work can be seen in the map below. Here, Dr Graham (along with Dr Bernie Hogan and Gavin Baily) downloaded every single Wikipedia article, parsed out every article about a place, calculated a range of metrics (such as conflict scores and word counts), and placed them all on an interactive map of the world. This map allows us to see some of the ways that 'big data' annotates and augments our world.

Mapping Wikipedia

Each one of the dots on the map represents human effort that has gone into describing some aspect of a place. The density of this layer of information over some parts of the world is astounding and it is hard not to be awed by the cloud of information about hundreds of thousands of events and places around the globe.

However, many of the smaller language Wikipedias have a very limited global focus. If your primary free source of information about the world is the Persian or Arabic or Hebrew Wikipedia, then the world inevitably looks very different to you than if you were accessing knowledge through the English Wikipedia. There are far more absences and many parts of the world simply don't exist in the representations that are available to you: potentially causing an entrenchment and a reproduction of the visibility of the already highly visible.

Second, Dr Graham's work on 'big data' and geography asks what the data shadows over places are able to tell us about the very places that they represent. Using conventional data collection methods (e.g. surveys), it can be extremely challenging, expensive, and time-consuming to draw broad-scale inferences about social, political, and economic trends around the world.

Mapping world religionsIf we take religion as an example, it can difficult to measure the precise and shifting geographies of the world's major religions. However, as in the previous example, augmentations of place can be mapped in order to draw unique insights in this area. In this map (left), all references to four religious keywords in Google Maps were captured and plotted. We then see very distinct geographies of online influence for Buddhism, Christianity, Hinduism, and Islam. These digital geographies mostly seem to mirror dominant material religious practices in those places. However, we also are able to observe a few unexpected results (e.g. the cluster of references to Buddha in Switzerland).

More broadly, Dr Graham's work in this area seeks to provide critical insight into the geography of virtual expressions in order to highlight the mutually constitutive, and at times contradictory, relationships between virtual and material layers and practices. Or, in other words, it allows us to think through how mappings of virtual representations of material practices are important tools for understanding how online activities simultaneously represent and reproduce the material world.

Mapping the world's TweetsFinally, Dr Graham's work seeks to understand the geographies of voice and authorship in layers of 'big data' that annotate our world. Here again, we see massive differences in the production of information and knowledge. Some people and producing and reproducing massive amounts of information, and many others are largely left out of these processes (see the map of the origin of geocoded tweets, left).

Dr Graham's work ultimately shows that 'big data,' in many ways, can allow us many new and unprecedented insights into material places and practices. We are now able to ask new questions and measure patterns, processes, and places in ways never before possible. However, there is also a need for critical scholarship about the thick cloud of information that blankets many of the places we inhabit.

We need to keep asking who is producing this information? Which voices and narratives are present, which are absent, and which are dominant? Who decides how contested places and processes get represented and reproduced? And whose interests do many of the layers of data that we consume and move through serve? Dr Graham's work continues to focus on these questions in order to ensure that we are able to measure, understand, and critique the socially constructed layers of information that augment our cities rather than accepting them as natural and true reflections of place.

Contact information

Oxford Internet Institute, University of Oxford, 1 St Giles Oxford OX1 3JS, United Kingdom Telephone: +44 (0)1865 287210 Fax: +44 (0)1865 287211 Email: press@oii.ox.ac.uk