JEREMY W. CRAMPTON
Will Peasants Map? Hyperlinks, Map Mashups, and the Future of Information
In this essay, I examine the changing dynamics of how maps and information are interlinked. I argue that for most of its history, mapping has been the practice of powerful elites—the sovereign map.1 Nation-states, governments, the wealthy, and the powerful all dominated the production of maps, and knowledge of the world emanated from the elites for the benefit of the elites.2 This history is now being challenged by the emergence of a new, populist cartography, in which, through new forms of linking, the public is gaining access to the means of producing maps.
This is certainly not an isolated development. It is part of a larger movement of counterknowledges that are occurring in the face of ever-increasing corporatization of information, such as the consolidation of the news media into the hands of a few global multinationals and their dominance by fairly narrow interests. The Internet and Web, blogs, and the “netroots” (online political activism) are all reasons for this “people-powered” control of information.3 In this essay, I focus on some of the exciting new developments that can help create, visualize, and disseminate geographical information. I also note a number of obstacles that impede widespread dissemination of these tools.
Popular versus Populist Cartography
Maps are a powerful way of knowing about the world and have always involved linking certain types of information to spatial representations of that information. Evidence of map use dates back to earliest historical times (Greece, Rome, and Mesopotamia). Map popularity has fluctuated: at certain times, the public has embraced maps; at other times, maps were only made due to new geographical discoveries and technologies, without explicit demand by the public. However, if we look at these occasions, they all share something in common; they were popular but not populist events. In popular mapping, the control of geographical information remains in the hands of an elite. In populist mapping, by contrast, the public not only has access to maps as an end product but can control the means of production of maps. This populist project is a truly radical historical departure that has the potential to change the future of information. It faces some difficult challenges and obstacles, which I shall discuss.
Historical Examples of Popular Mapping
During the sixteenth century, as new territories were being explored, an explosion of new maps became available from the big European cartographic publishing houses, such as Ortelius. The map of the world produced by Ptolemy in the first century AD was rediscovered and republished in the mid-fifteenth century with very little modification. The Behaim Globe of 1492 (the oldest surviving globe) was indicative of knowledge at the time in that it obviously omitted the Americas but also pushed Asia eastward by 1500 miles, making it far more reachable from Europe. It is thought that this mistake confirmed Columbus in his enterprise of the Indies. If Columbus did not see that globe, he would be familiar with its general content as a navigator himself (and one who had a brother in the mapmaking trade), from world maps and maps of ports along the coast (known as portolan charts).
Subsequent to the “Columbian encounter,”4 however, information about far-flung territories and continents came in thick and fast, and publishers vied with each other to produce the most up-to-date maps. Juan de La Cosa, who sailed with Columbus, was the first European to map the American continent (1500), while Martin Waldseemüller's map of 1507 (recently purchased by the Library of Congress for ten million dollars) was the first to name it. The Flemish cartographer Mercator invented his eponymous projection in 1569, still in use in classrooms today. The sixteenth century also saw Abraham Ortelius issue the first modern atlas, the Theatrum Orbis Terrarum (Theater of the World) by combining maps into book form in 1570. It was Mercator, however, who coined the word atlas (for his collection in 1595). John Smith's map of Virginia of 1608 is also well known and includes a drawing of Chief Powhatan, father of Pocahontas. As these selected examples testify, the importance and number of maps and cartographic publications during this period cannot be underestimated, and they were embraced by the public. Every educated person considered their library incomplete without atlases and maps.
A different kind of popular mapping emerged during the nineteenth century. In this case, the new knowledges were not of territories but of science. Many types of “thematic” maps that then developed—such as proportional symbol, dot distribution, choropleth, and isoline maps—form the basis of today's mapping and GIS software.5 John Snow, for example, considered today to be the father of epidemiology and a keen exponent of the geographical nature of disease, is famous in both geography and public health for his work that used mapping to analyze the cholera epidemic.6 Snow's map identified the very water pump that was the source of the cholera–infected water. This was a full three decades before the germ theory of disease was accepted in the 1880s.
With the rise of industrialization and urbanization, the modern state mapped out a host of problematic subjects: crime, education, divorce, birthrates, education, poverty, disease, the distribution of languages, and new immigrants. All these topics received treatment at the hands of the century's political scientists, protodemographers, geographers, and governments. Even Florence Nightingale used data graphics to convince skeptical British officials that dirt and disease killed more men than fighting did.
Many of these maps were published in a new kind of atlas, the “statistical atlas.” It mapped not territory but, rather, the nation's human resources. Based on the census, it was aimed not just at officials but at the educated public. The first American statistical atlas was printed in sufficient numbers for both the public and libraries to purchase it.7 New editions were issued every ten years in time with the census.
But interest in knowledge of places had not disappeared; in line with the new scientific knowledges, it was oriented inward, at the home territory. As Schulten describes in her fine account of the Rand McNally mapping company,8 in America, at least, the continent was still relatively unexplored. By the time the “closing” of the American frontier took place in 1893,9 maps were required for the emerging automobile industry.10 The American Geographical Society (AGS), founded in 1851, and the National Geographic Society (NGS) also provided the public with prodigious quantities of new maps and explorer's accounts. These were popular with the public, if not among more serious-minded academics.11
The world wars of the twentieth century also stimulated a renewed public appetite for maps. During World War II, many Americans followed the progress of the war on wall maps, an activity encouraged by President Roosevelt himself. Such popular outlets as Fortune and the Los Angeles Times published incredible new maps by Richard Edes Harrison and Charles Owens suited to the “air age,” featuring views over the polar ice caps and perspectives of Europe as seen from Moscow.12 The war itself was not short on propaganda maps, often dropped in the thousands from the air over enemy territory.
Countermapping
The preceding examples demonstrate that the popularity of mapping has waxed and waned historically in conjunction with new demands for maps or new opportunities for maps. Noticeably, map popularity is associated with the production and dissemination of new knowledge by elites (the state, the wealthy). As maps are deeply cultural phenomena, this geographical knowledge does not exist in a sociocultural vacuum. Mapmakers share a top-down approach; information is disseminated from a cadre of cartographic experts to a largely ignorant public. This public has no control over what information is provided, when it is provided or in what form, how much it costs to access it, who can access it, the possibility of challenging this information and getting other information, and so on; that is, the system was profoundly undemocratic.
The fact that the distribution and circulation of geographic information was constrained in this way should not surprise us. The control of information and knowledge for the benefit of a political elite has been a hallmark of information for as long as there has been information, as writers on public opinion have long pointed out.13 Nevertheless, parallel to this control has been a current of opposition and critique, which in cartography takes the form of countermapping.
The idea of countermapping is to reverse power asymmetries. Maps can be created by small groups, communities, and even individuals to achieve goals not otherwise possible. They can be used by those in developing countries to work against dominant Western information. They are “counter” to the prevailing structures of power, especially those deployed by the state. For example, countermapping has been used in conservation to show that the way some areas are mapped affects their status as protected areas.14 Some spaces that include indigenous people with non-Western cultures might be construed as uninhabited and a candidate for environmental protection in ways that would disrupt their lives or ignore local knowledges. While biodiversity and species loss are critical ecological factors, the simple protection of areas can also be merely an extension of state control to the exclusion of local actors. Countermapping can be employed to give voice to these actors, whether they are in East Africa or impoverished American urban neighborhoods.
Indeed, one of the earliest examples of countermapping (although no term yet described it) was performed during the 1960s by the radical geographer Bill Bunge in urban areas, such as Detroit.15 Bunge's maps were produced with groups in the inner city struggling for civil rights and safe neighborhoods. One famous example mapped out rat bites on neighborhood children; another showed clusters where children had been hit by cars.
Countermapping often employs the very tools that have previously been used to assert dominant power relations. For example, in community GIS (sometimes known as participatory GIS, or PGIS), local communities may use cheap or Web-based GIS tools to map out neighborhood resources (e.g., community centers, parks, and open spaces) to resist development. PGIS is a grassroots phenomenon with the goal of empowering traditionally disempowered groups. “Map or be mapped” might be its motto.
Countermapping is an attempt to create maps based on different kinds of knowledge that explicitly embrace a political, partisan point of view. Countermappers claim that all maps have such points of view. Maps are not “mirrors of nature” that reflect knowledge but sites of knowledge production. Knowledge is created not in isolation but in conditions that privilege some knowledge over others.
These ideas have proven to be very influential in understanding spatial representations, and they parallel research in other areas, such as spatial cognition. For example, children appear to go through a process of understanding spatial relations as a creative process. As two leading investigators succinctly put it, “maps are creative statements about the world, not merely degraded versions of it.”16 Sarah Elwood, a leading researcher of PGIS, has argued that the conditions of spatial knowledge production are political.17 This does not mean that maps and GIS are biased; it means that knowledge is produced under conditions of power. Here we are close to a well-known idea in the work of Michel Foucault, that of “power-knowledge.” Foucault said that knowledges are usually produced under certain conditions of power and that some knowledges are privileged while others are “subjugated.” For example, he speaks of “a whole series of knowledges that have been disqualified as nonconceptual knowledges, as insufficiently elaborated knowledges, knowledges that are below the required level of erudition or scientificity.”18
Relevant for our purposes here is that sometimes counterknowledges can emerge and provide the basis for a critique of the prevailing way of doing things, likened by Foucault to an “insurrection” from below.19 The most obvious parallel to this insurrection are the “netroots,” a term coined in 2002 by Jerome Armstrong to describe the online grassroots political community. Armstrong has stated that he was attracted to “the whole netroots to grass roots type of political activism that the Internet enabled.”20 Armstrong and Markos Zúniga wrote that they were “crashing the gate” of establishment politics.21
The netroots has some interesting parallels to countermapping. It is organized from the bottom up and distributes messages through blogs and other social networks.
The principal value of the blogosphere is that it democratizes our political discourse almost completely. Anyone can become a “pundit,” find an audience, report facts, create a community of like-minded citizens and activists, and influence the public discourse—all without having to mold oneself into what is demanded by the Washington Post and without having to care about pleasing the editors of Time magazine.22
To democratize “discourse” meaningfully in the case of mapping requires tools that are accessible to as many people as possible, the knowledge to use those tools, access to relevant data, and the ability to analyze and display that data on maps. In the next section, I discuss the development of these tools and what they mean for the future of geographical information.
The “Democratization of Cartography”
It turns out that when we talk about “the world's information,” we mean geography too.
—Google
Up until the 1980s, it had always been assumed that maps were essentially devices that communicated information that had been gathered and processed by the expert cartographer. As the historical examples previously cited testify, this had been the case for hundreds of years. The craft of cartography had a guildlike status, requiring years of training and the mastery of specialized techniques. These ideas about how maps worked were formalized in the postwar years by Arthur Robinson, a professor of geography at the University of Wisconsin–Madison. Robinson provided the conceptual apparatus of what later became known as the Map Communication Model (MCM), which explains mapping as a process of communicating information from the map expert or cartographer to the map reader. The information is acquired, marshaled, and selected by the map expert and set down on the map.
It is a very top-down model. For example, the cartographer-expert might acquire information on the distribution of crops across the Midwest, select and arrange the information (e.g., into categories of different crop types), and then symbolize it cartographically (e.g., as a dot distribution map). The map reader-novice then absorbs the information.
However, there were problems with this model. Cartographers had no way to decide how to present the information or even if their maps were being understood. Robinson's insight came in paying attention not only to the way the information was laid out (symbolized) on the map but to the abilities of the map reader to absorb it. His keystone work issued a call for research into “the physiological and psychological effects” of map design.23
This idea was based on that of one of the most influential scientists of the twentieth century, Claude Shannon. Shannon is the progenitor of communication (or information) theory.24 This theory is at the heart of our digital devices, such as computers. Shannon recognized that information was “countable.” Using his methods, it became possible to count the maximum amount of information that it was possible to transmit through a particular channel,25 such as a map. Shannon showed that communication could be improved if the “signal” (the information) was maximized and the “noise” (the unwarranted distortions or errors) could be minimized. This signal-to-noise ratio (SNR) is still used today in information theory to measure the quality of a communication. Using communication theory, Robinson cleared the way for the development of the map communication model in the late 1960s. By 1972, this model was firmly established in the discipline, with the International Cartography Association (ICA) establishing “the theory of cartographic communication” as one of its terms of reference.26
By the 1980s, however, there were a number of pressures on this account of mapping. For one thing, public control of information became more possible with the arrival of inexpensive desktop computers and the first mapping software. Mapping programs had been around since the 1960s and were later to prove very influential. The Harvard Laboratory for Computer Graphics, for example, nurtured early developments in GIS.27 But they were cumbersome, limited to expensive equipment, and required sophisticated programming skills. They were also very crude in appearance. By the 1980s and the advent of the Apple Macintosh (a platform quickly embraced in the graphic design, publishing, architecture, and cartographic communities), a new form of mapping—desktop cartography—was possible.
As a graduate student, I can still remember the thrill of those first Macs. The department taught cartography the “old” way (darkroom, camera, and photographic chemicals) until the late 1980s. Students were expected to buy ink pens and master free-form drawing on mylar as they had done for decades—in his 1948 cartography guide, Raisz had included a chapter on how to avoid smudging your ink.28 Now one could guarantee a straight line of constant width with a flash of the mouse.
Cartographers quickly realized that these new tools afforded new mapping possibilities. At the time, scientists were working on “scientific visualization,” a set of approaches for visualizing scientific data. In geography, this became known as geovisualization.29 Typically, visuals are of secondary importance in science or are only used to communicate findings—the “knowns.” By the late 1980s, scientists and cartographers realized that visualization could be a research tool to explore data to find hidden patterns. These exploratory tools focused on discovering the “unknowns” in a data set. Today, the GIS business is believed to generate anywhere from four to ten billion dollars a year,30 and the geospatial global business is possibly as large as fifty billion dollars a year.
Very large data sets, such as satellite imagery of deforestation, could now be interactively “data mined” for significant patterns without requiring prior knowledge of the situation. The power of the visual graphics was that they could display huge amounts of data at once. Today's Google Earth (GE) is an offshoot of this work—it provides a visualization of the earth with which the user can interact in any number of ways (zooming scale, adding or subtracting data layers, measuring distances, calculating directions, etc.). Both desktop mapping and visualization moved the production of mapping from the hands of the elite into those of the public. Mark Harrower, a leading proponent of populist cartography, has observed:
One of the themes of my profession right now is the democratization of cartography.... Mapping used to be a job of the elite, the Rand McNallys and National Geographics of the world. Now people are taking it upon themselves to map their passions.31
In other words, desktop mapping and geovisualization provided the beginnings of new forms of people's mapping. But the true democratization of cartography would only arrive with the advent of new advances in Web technology, often referred to as Web 2.0 functionality, such as massively distributed and hyperlinked data sets, mashups, and customizable open-source tools. These tools are profoundly different from their precursors because they allow collaboratively linked mappings.
Populist Mapping Applications: Web 2.0 and Web-Based Mapping
Google Earth
With the release of Google Earth in the summer of 2005, it became apparent that there was a tremendous public appetite for visualizing geographic information. GE-like tools had existed in scientific GIS for some years (and Vice President Al Gore had outlined an early vision of digital earth in 1998),32 but Google's popularity was far greater. The key to Google's success lay in providing open access to Google Maps, known as an application programming interface (API). Using this API, members of the public could “hack” (i.e., modify) these maps and link them up with their own data.33 The results are known as map mashups.
Google Earth is a data visualization tool—it does not perform analysis, run models, or manipulate data (create buffers, merge one layer with another, etc.). It provides realistic imagery and 3-D pictures rather than the abstract cartographic symbolization of traditional mapping. Despite these aspects—or, rather, because of them—Google Earth is easy to understand and is “natural” looking (although no view from space would ever look like GE).
Google Earth and other map open-access APIs are highly collaborative and provide fertile ground for other data to be linked and geographically visualized. For example, Google Earth now sweeps through Wikipedia and automatically makes maps of places mentioned in the articles (through Placeopedia). Google has created a feature that maps all the places mentioned in books and puts them into a Google map mashup. These maps give you a chance to see not only how “platial” (how rich in geographical reference) but also how concentrated or distributed the book is. “Where” does the book focus? Is it Westernized? Is it oriented to Europe and North America? One could also compare the maps from two different books on the same topic (e.g., the spread of a disease, like HIV or SARS) to see if they tell different stories.
Google has also implemented a layer of information called the Geographic Web, in which people can annotate the earth with their photographs or place descriptions. As with the Wikipedia project, Google seems to have realized a project that works because of user collaboration and contributions—data now come from the bottom up rather than from the top down.
Census Bureau Data
Every ten years, the United States collects reams of data about its population and the places people live. Additional data is collected on an annual basis. All sorts of topics can be mapped, including income, race, age, gender, ethnicity, and occupation. The census is probably the most important single source of sociodemographic data about America today, and its findings inform policy analysis and decision making of all kinds. All this data can be mapped—if you know how to navigate the Census Bureau's labyrinthine databases.
The bureau offers an online mapping tool called the American FactFinder, which is useful for an initial visualization of the data. The display is quite small however. Most serious users download the raw data sets and process them with GIS. Both of these approaches restrict usage of the data. Recently, a different approach was developed that allows users to interactively display census data without having to have GIS expertise. The tools to do this are distributed across the network, thus providing access for many more people than if it were desktop based. This is the Social Explorer project, based at Queens College at the City University of New York, in association with the New York Times. Social Explorer provides an easy-to-use interface to huge quantities of complex census data dating back as far as 1940.
Political Applications
Republicans still control the maps.
—Chris Bowers, MyDD.com
There is now some intriguing mapping evidence that suggests that access to, control of, and dissemination of geospatial information is changing political participation.34 While much political discussion occurs in the traditional, or “mainstream,” media, much is now also held in the emerging arena of blogs. Blogs now constitute a significant and noteworthy component in today's political landscape. Blogs and online political activism (the netroots) now play important roles in campaigns for getting out the vote (GOTV) and “getting out the dollar” (especially in online fund-raising). Since the 2004 elections and the success of Howard Dean and such organizations as MoveOn.org, the intersection of netroots and politics has only become stronger.
Working alongside and often in conjunction with the netroots are a range of mapping and GIS tools now available for the public. These tools often rely on making linkages between different kinds of things: for example, between different sources of data (e.g., between Google Maps and the U.S. Census Bureau or the Federal Election Commission) and between different software programs (e.g., between GIS and Google Earth). These linkages, effected through open-source software and APIs, mark a potentially new phase of political activism and collaboration characterized by more democratic access, control, and production of information and knowledge; a more local “micropolitics”; and potentially a way to break the hold of establishment “big money” incumbents.
For example, the FairData Web site provides community-based interactive maps for the whole nation down to level of precincts and census block groups.35 These data are linked to open-source mapping APIs, such as Google Maps, for visual display. Users can pan and zoom across the maps and display different layers of information (the site uses a sophisticated online GIS as a back end to the Web pages). For a GOTV effort, community organizers can create maps of the number of nonvoters by precinct. In the map of Philadelphia in figure 1, the voting turnout is shown for each precinct, allowing the GOTV team to target nonvoting neighborhoods.
The map shows that turnout varied quite considerably across the city and was below 40 percent in many areas. These areas can then be targeted by the GOTV effort. The maps can also show individual households that did not vote for even more targeted efforts. As far as I am aware, these are the first tools available to the public that were previously only compiled by political parties in secret political precinct maps.
Do these tools by themselves mean that the political landscape is now more democratic? Not necessarily. Foucault's reminder (mentioned earlier) about power and knowledge is nowhere more salient than in the relationship between the military and digital mapping and geovisualization. The size of the military investment in GIS, such as the geospatial intelligence (GEOINT) community, is not known but was formally recognized in the creation of the federal National Geospatial-Intelligence Agency (NGA) in 2004, and the military's doctrine on GEOINT has been described in recent reports.36 Because GIS has historically been largely associated with government and industry (e.g., the GEOINT 2006 symposium was keynoted by the director of national intelligence, John Negroponte), there are many who view GIS as being just another mechanism of government control and surveillance.37
Pickles argues that many of the new mapping capabilities are wonderful.
They provide more powerful tools for local planning agencies, exciting possibilities for data coordination, access and exchange, and permit more efficient allocation of resources, and a more open rational decision-making process.38
Yet he concedes that these systems are taking place in a larger context of economic production and a “culture of military and security practices.”39 Trevor Paglen, a geographer at Berkeley, has investigated many of these “hidden geographies” and even provided a map mashup of the CIA's “extraordinary rendition” flights.40
Open-Source Access to Geospatial Data
The development of open-source data and tools is very attractive to those who seek to avoid political and cultural associations and retain control over the maps. For example, Microsoft offers MapCruncher, a technology that allows anyone to make their own map mashup in about ten minutes. NASA offers a global map projector—you can take any map and project it automatically. One listing gives over 230 ongoing open-source GIS projects (http://opensourcegis.org/).
One obstacle faced by the open-source mapping community is that many map data layers are protected by copyright, especially in the United Kingdom (the United States does not copyright federal data). The national mapping organization of the United Kingdom, the Ordnance Survey, can regulate these data through licenses. In response, the OpenStreetMap is a wiki-based collaborative mapping project to create mapping coverages that are copyright free (under the Creative Commons license).
Participants in the OpenStreetMap project take GPS systems with them wherever they go and then upload the recorded routes into the system. In the beginning, parts of the project were based on old copyright-expired maps. Other users who do not have GPS can edit or annotate the uploaded maps. (A courier company in London has also provided tracks of virtually all London roads.) For more inaccessible areas, such as Baghdad, the project has made an agreement with Yahoo to use its aerial imagery. This will provide digitized map layers of all features (roads, rivers, railroads, parks, etc.), which can be used in many different applications—for example, the data can be exported to Google Earth for wide viewing and distribution.
A “slippy map,” in the Google Maps style, allows users to browse across the map and zoom in and out to specific cities. The level of detail is at near-professional levels, which poses the real challenge of these projects: will they provide competition for the traditional top-down providers of geospatial information? It at least seems likely that open-source mapping will provide a parallel alternative set of publicly available data, but it does not seem likely that they will replace traditional data providers.
Another obstacle is that data providers sometimes curtail or suppress data that they have. The most well-known instance of this involves Google's imagery of India, China, and Korea. In these countries, Google has agreed to degrade the quality of imagery for certain military sites. (Other countries, such as Indonesia, have declined to make this request, stating that the reduced quality would simply indicate where such sites are located.) The sensitivity of this practice was illustrated when questions were asked in the U.S. Congress about Google's provision of imagery rather closer to the United States.41 Google revealed that in September 2006, they had replaced newer imagery of the devastated city of New Orleans with pre-Katrina imagery. Google responded that they continued to provide the newer imagery (which is actually lower quality) on a special Web site. However, Google's alterations to imagery, sometimes at the request of foreign governments, raises questions about the future of information supply.
Community and Participatory Mapping
When groups of people come together to address a problem, they can leverage economies of scale. Think of the online social networking communities, such as del.icio.us, Diggit, and Slashdot. In mapping, this leveraging has often taken the form of community or participatory mapping, which I already discussed in the context of countermapping.42
Scientific Applications of Map Mashups
Scientists are now using collaborative mapping tools to visualize and bring to light spatial patterns of things as diverse as bird migration patterns or the spread of the SARS virus and to demonstrate how logging will affect downstream communities.43 As we have seen, open-source geospatial APIs, such as Google Earth and Yahoo Maps, are popular and powerful tools. This point has not been lost on scientists, who are increasingly turning to these tools in order to visualize and communicate data. For example, Declan Butler, a senior reporter at Nature, regularly posts KMZ (Google Earth) files in articles showing the occurrence of avian influenza A (H5N1) and other public health issues.
Another fascinating application has been produced by the American Association for the Advancement of Science (AAAS) as part of its Science and Human Rights Program. The Geospatial Technologies and Human Rights (GaTHR) Project uses high-resolution global satellite imagery to examine areas of the world that are otherwise impossible to access, such as Darfur in Sudan. In 2004, this imagery confirmed the extent of ethnic cleansing in this area, and it is now available as GE layers. The AAAS says:
The QuickBird imagery used by the Department of State and USAID, together with other high-resolution imagery, has proven especially valuable as it can show damage to small houses, orchards, fields, and other features. Given the unequivocal time of image acquisition it can authoritatively document changes to these features, and in printed form the imagery helps compile and synthesize witness reports during interviews.44
Such a project can, of course, benefit from traditional GIS, but its public outreach and dissemination component is significantly improved by using publicly assessable outlets of data visualization. The GaTHR project also works with members of the human rights community (e.g., Amnesty International) who may not have access or expertise in costly technology.
Geospatial technologies potentially offer human rights researchers and advocates a significant new tool for assessing human rights violations and monitoring developing crises in geographic areas where it is difficult to send observers.... These tools may also provide compelling documentation to encourage intervention and to determine responsibility. The initial phase of this project will enable AAAS to evaluate the potential uses and to determine the most feasible way to develop and disseminate these technologies within the human rights community.45
GE and similar applications, such as NASA's World Wind and Microsoft's Virtual Earth (VE) 3D, do not provide “real-time” data as many people believe (except in special circumstances). But they are vitally important for looking at “change detection” (comparing imagery between different time periods). Change detection can show whether villages or buildings have been razed to the ground, for example—as the AAAS found in Zimbabwe, despite governmental silence.
Barriers for Linking Geospatial Data
Interoperability
There are still many barriers to the use of open-source geospatial tools, map mashups, and map hacking. Some of these are technological—for example, ensuring that different software can operate with each other, or interoperability. The development of widely accepted standards and metadata is the most workable solution to this problem. We are currently in a situation analogous to the many standards for high-definition DVDs. They all work, but not necessarily together. Efforts such as the Open Geospatial Consortium (OGC), an international consortium of governments, universities, and corporations, can promote standards and inter-operability to a certain extent in a top-down model. The biggest problem here is not getting software to connect but getting the data and metadata into standard forms.
Institutional Barriers
As we have seen, there are presently two different realms of mapping and GIS data: GIS and Web-based mapping. The GIS industry is having to catch up to the popular applications, such as GE and VE. In the last year or so, programs that link between popular traditional GIS applications have appeared: an example is the Arc2Earth program, which links ESRI's ArcGIS and GE. GIS companies, after largely ignoring programs for digital earth visualization because they did not provide analysis, are now struggling to quickly catch up and leverage the tremendous popularity of Web-based mapping. A major breakthrough in thinking came when ESRI realized that they needed not just an exporter from ArcGIS to GE but one from GE to ArcGIS—that is, that you could bring pretty “pictures” into industrial-grade GIS to do serious work.
Another institutional barrier arises from the corporatization of information. The Internet has undergone tremendous corporatization over the last ten years, not solely in terms of content, but also in terms of ownership of the mechanisms of distribution (the cables and phone wires). In particular, there is mounting concern over the erosion of “net neutrality.” The concern is that Internet providers might no longer treat all Internet traffic equally. End-users, for example, may experience differential access to Web sites in accordance with fees that the sites and end-users may have paid (or not paid). This “tiered” access would resemble the current model often adopted by cable providers, whereby consumers receive different TV channels according to the package they have purchased. The fear is that access to content per se or even differential speeds of access to content (faster or slower) may become the norm on the Internet. One model to circumvent this is to switch to open-access WiFi broadband, but even the provision of that access is ultimately cable bound.
Advocates of tiered information access argue that it is a typical financial model found in many businesses. Proponents of net neutrality argue that the concept of business models should not apply to the provision of such an important source of information. The debate over net neutrality is currently being fought out in competing legislation at state and federal levels and is undoubtedly going to remain an important issue in the next few years.
The Digital Divide
The digital divide is a measure of access to the digital information economy. It includes access to technology (hardware and software) but also to knowledge itself (education). Recent research has demonstrated that the Internet is not free of the geographical restraints of the physical world.46 These divides occur at a plethora of scales: within a city, within a region, within a country, and between one country and another. For example, according to figures from the United Nations, Internet access rates are 19 per 1,000 people in sub-Saharan Africa; in high-income OECD countries, they are more than 30 times higher, on average 563 per 1,000.47 But even within the United States, broadband access (required for many of today's Internet applications) is currently installed in about 45 percent of homes—a high proportion, but certainly not at saturation level.48
The digital divide is enduring in the sense that new technologies are constantly being produced and constantly being spread unevenly. Each time we invent a great technology, we ironically also produce inequalities. As Shirky has argued, “diversity plus freedom of choice creates inequality, and the greater the diversity, the more extreme the inequality.”49 As the Internet increases in size and diversity, inequalities will also increase and replicate the digital divide patterns already found in the physical world.
Research also shows that there are some remarkable geographies of hyperlinks between blog clusters on the right and the left of the political spectrum. Looking not at the cross section of all blogs but, rather, at those that carry the most readership (the A-list bloggers), Adamic and Glance found that in the months preceding the 2004 U.S. presidential election, the degree of interaction between liberal and conservative blogs was very low. Both blogospheres linked mostly within their own communities and not across the political divide.50
Conclusion: Can Peasants Map?
Many obstacles to digital access, such as the digital divide and net neutrality, are not, at base, technological issues that can be addressed through market incentives; rather, they are complex sociopolitical problems. Lack of access to online information parallels the very underserved populations it could benefit. Community and participatory GIS, the netroots, and Web-based mapping are therefore not likely to provide solutions for underserved populations to bootstrap themselves out of poverty. But if underserved and well-served communities work together, then problems can be more ably addressed. This is a big if, and as this essay shows, there are enduring divides and connectivities. After all, we live not in isolated communities but in a world of networks.
In his work on political Net-based activism, David Perlmutter explores the question of whether the online activism and the netroots are a representative constituency—specifically, whether bloggers are “the people.” He points out that at the moment, the netroots are overwhelmingly young, white, male, well educated, and technologically savvy and are thus not representative of the population as a whole. As he put it, “peasants do not blog.”51
In this essay, I have introduced a number of developments that both assist and create obstacles for access and usage of geospatial information. These tools are provided out of a genuine realization that the ways we visualize and understand the world around us—its places, geographies, and relationships—are undergoing a radical transformation. If the media (TV, newspapers, and news radio) has had to adapt and incorporate new models of information dissemination and participation, and if publishing is undergoing a similar transformation, then there would seem to be an equivalent transformation working on our mappings. The remaining questions, however, are to what degree, how much, and with what effects these tools will confront the obstacles and barriers. The answers to those questions will prove vital in deciding the future of information.
NOTES
1. C. Jacob, The Sovereign Map: Theoretical Approaches in Cartography throughout History (Chicago: University of Chicago Press, 2006).
2. D. Buisseret, Monarchs, Ministers, and Maps: The Emergence of Cartography as a Tool of Government in Early Modern Europe (Chicago: University of Chicago Press, 1992).
3. J. Armstrong and M. M. Zuniga, Crashing the Gate: Netroots, Grassroots and the Rise of People-Powered Politics (White River Junction, VT: Chelsea Green, 2006).
4. J. B. Harley, M. Warhus, and E. Hanlon, Maps and the Columbian Encounter: An Interpretive Guide to the Travelling Exhibition; American Geographical Society Collection (Milwaukee: Golda Meir Library, University of Wisconsin, 1990).
5. Maps that show geographical distributions (as opposed to road maps and general reference maps) can be categorized into several types. These maps are known generally as thematic maps and tend to show quantitative data. Isoline maps are basically contour maps (with lines of equal elevation) but can also represent abstract information (as in the familiar temperature map). Dot distribution maps show distributions of events or features with scatterings of dots (the more dots, the more features). Proportional symbol maps use size to indicate quantity (e.g., line width for traffic flows). Choropleth maps are one of the most familiar thematic maps. They take predefined areas (e.g., countries) and show quantity for each area (e.g., per capita income by country on a world map).
6. S. Johnson, The Ghost Map: The Story of London's Most Terrifying Epidemic—and How It Changed Science, Cities, and the Modern World (New York: Riverhead, 2006); T. Koch, “The Map as Intent: Variations on the Theme of John Snow,” Cartographica 39, no. 4 (2004): 1–14.
7. F. A. Walker, Statistical Atlas of the United States (New York: J. Bien, 1874).
8. S. Schulten, The Geographical Imagination in America, 1880–1950 (Chicago: University of Chicago Press, 2001).
9. F. J. Turner, The Frontier in American History (New York: Henry Holt and Company, 1921).
10. J. R. Akerman, “American Promotional Road Mapping in the Twentieth Century,” Cartography and Geographic Information Science 29, no. 3 (2002): 175–91.
11. H. Clout, “Geographers in Their Ivory Tower: Academic Geography and Popular Geography in Paris 1931,” Geografiska Annaler Series B Human Geography 87, no. 1 (2005): 15–29.
12. D. E. Cosgrove and V. della Dora, “Mapping Global War: Los Angeles, the Pacific, and Charles Owens's Pictorial Cartography,” Annals of the Association of American Geographers 95, no. 2 (2005): 373; R. E. Harrison, Look at the World: The Fortune Atlas for World Strategy (New York: Knopf, 1944).
13. E. S. Herman and N. Chomsky, Manufacturing Consent: The Political Economy of the Mass Media (New York: Pantheon, 2002); W. Lippmann, Public Opinion (New York: Free Press, 1922).
14. L. M. Harris and H. D. Hazen, “Power of Maps: (Counter) Mapping for Conservation,” ACME 4, no. 1 (2006): 99–130.
15. W. Bunge, Fitzgerald: Geography of a Revolution (Morristown, NJ: Schenkman, 1971); W. Bunge, “The First Years of the Detroit Geographical Expedition: A Personal Report,” in Radical Geography, ed. R. Peet (Chicago: Maroufa, 1969), 31–39.
16. R. M. Downs and L. S. Liben, “Through a Map Darkly: Understanding Maps as Representations,” Genetic Epistemologist 16 (1988): 16.
17. S. Elwood, “Beyond Cooptation or Resistance: Urban Spatial Politics, Community Organizations, and GIS-Based Spatial Narratives,” Annals of the Association of American Geographers 96, no. 2 (2006): 323–41; S. Elwood, “Negotiating Knowledge Production: The Everyday Inclusions, Exclusions, and Contradictions of Participatory GIS Research,” Professional Geographer 58, no. 2 (2006): 197–208.
18. M. Foucault, Society Must Be Defended: Lectures at the College De France, 1975–1976 (New York: Picador, 2003), 7.
19. Ibid., 9.
20. W. Safire, “Netroots,” New York Times Magazine, November 19, 2006.
21. Armstrong and Zuniga, Crashing the Gate.
22. G. Greenwald, “Blogs, Alternative Political Systems, Funding,” http://glenn greenwald.blogspot.com/2007/02/blogs-alternative-political-systems.html.
23. A. H. Robinson, The Look of Maps: An Examination of Cartographic Design (Madison: University of Wisconsin Press, 1952), 13.
24. C. E. Shannon, “A Mathematical Theory of Communication,” Bell System Technical Journal 27, no. 3 (1948): 379–423, 623–56.
25. D. Mindell, J. Segal, and S. Gerovitch, “From Communications Engineering to Communications Science,” in Science and Ideology: A Comparative History, ed. M. Walker (London: Routledge, 2003), 66–96.
26. L. Ratajski, “Commission V of the ICA: The Tasks It Faces,” International Yearbook of Cartography 14 (1974): 140.
27. N. Chrisman, Charting the Unknown: How Computer Mapping at Harvard Became GIS (Redlands, CA: ESRI Press, 2006).
28. E. Raisz, General Cartography (New York: McGraw-Hill, 1948).
29. J. Dykes, A. M. MacEachren, and M. J. Kraak, Exploring Geovisualization (Amsterdam: Elsevier, 2005).
30. Daratech, “GIS/Geospatial Market Grew 17% in 2005 to Top $3.3 Billion,” http://www.directionsmag.com/press.releases/index.php?duty=Show&id=14697&trv=1.
31. “New Technology Helping Foster the ‘Democratization of Cartography,’” ScienceDaily, http://www.sciencedaily.com/releases/2006/09/060920192549.htm.
32. A. Gore, “The Digital Earth: Understanding Our Planet in the 21st Century,” http://www.isde5.org/al_gore_speech.htm.
33. S. Erle, R. Gibson, and J. Walsh, Mapping Hacks (Sebastopol, CA: O'Reilly, 2005).
34. E. Talen, “Bottom-up GIS: A New Tool for Individual and Group Expression in Participatory Planning,” Journal of the American Planning Association 66, no. 3 (2000): 279–94.
35. The FairData/FairPlan site is so vast that no description can really encompass it. It provides interactive maps, census data, precinct maps of registered non-voters by race, racial profiling data, GOTV data, and much more.
36. United States Joint Forces Command, Geospatial Intelligence Support to Joint Operations (Washington, DC, 2007).
37. J. Pickles, A History of Spaces: Cartographic Reason, Mapping, and the Geo-Coded World (London: Routledge, 2004); N. Smith, “Real Wars, Theory Wars,” Progress in Human Geography 16, no. 2 (1992): 257–71.
38. Pickles, History of Spaces, 148.
39. Ibid., 152.
40. T. Paglen, “Unmarked Planes and Hidden Geographies,” http://vectors.usc.edu/index.php?page=7&projectId=59; T. Paglen and A. C. Thompson, Torture Taxi: On the Trail of the CIA's Rendition Flights (Hoboken, NJ: Melville House, 2006).
41. Associated Press, “House Panel: Why Did Google ‘Airbrush History’?” http://edition.cnn.com/2007/TECH/03/31/katrina.google.maps.ap/index.html.
42. Elwood, “Negotiating Knowledge Production.”
43. S. Herhold, “Technology Builds Bigger Soapbox,” Mercury News, December 3, 2006.
44. American Association for the Advancement of Science (AAAS) 2007. “Geospatial Technologies and Human Rights.” http://shr.aaas.org/geotech/whatcanGISdo.shtml (assessed December 28, 2007).
45. E. W. Lempinen, “New AAAS Project Will Explore Geospatial Technology and Human Rights,” http://www.aaas.org/news/releases/2006/0127geospatial.shtml.
46. J. Chakraborty and M. M. Bosman, “Measuring the Digital Divide in the United States: Race, Income, and Personal Computer Ownership,” Professional Geographer 57, no. 3 (2005): 395–410; M. Crang, T. Crosbie, and S. Graham, “Variable Geometries of Connection: Urban Digital Divides and the Uses of Information Technology,” Urban Studies 43, no. 13 (2006): 2551–70.
47. United Nations Development Program, Human Development Report 2006: Beyond Scarcity; Power, Poverty, and the Global Water Crisis (Basingstoke Hampshire and New York: Palgrave Macmillan, 2006).
48. L. Rainie and J. Horrigan, “Election 2006 Online,” http://www.pewinternet.org/pdfs/PIP_Politics_2006.pdf.
49. C. Shirky, “Power Laws, Weblogs, and Inequality,” http://shirky.com/writings/powerlaw_weblog.html.
50. L. A. Adamic and N. Glance, “The Political Blogosphere and the 2004 U.S. Election: Divided They Blog,” Proceedings of the 3rd International Workshop on Link Discovery (New York: ACM, 2005), 36–43; see also Adamic's essay in the present volume.
51. D. D. Perlmutter, “Are Bloggers ‘the People’?” http://policybyblog.squarespace.com/are-bloggers-the-people.