04 January 2016

The energy demand of “the bones of cyberspace”

Terminal Commerce Building in North Philadelphia, originally an Art Deco furniture warehouse, now one of the largest data centers in the Northeast United States. Literally and figuratively behind the building, underlying the digital connectivity to ‘the cloud’, to cyberspace, is massive amounts of energy infrastructure. While we don’t really “see” even this building, it has a presence on north Broad Street, the main north-south road through Philadelphia. The energy systems maintaining this data center are in the background, inside and on top of, but mostly behind the massive structure. The corner of north 13th St. and Callowhill looking west. This view shows the immensity of the building nicely (note the yellow school bus for scale). October 2014.

This short essay is an addendum to “The urban, infrastructural geography of “the cloud”, a piece of writing that was chosen as an Ars Technica Editor’s Pick by Cyrus Farivar, shared widely on social media, and described by William Gibson as an exploration of “the bones of cyberspace”. Underlying these digital bones is energy in the form of electricity, and while a full examination of the energy demands of ‘the cloud’, of cyberspace, is certainly merited, here I present a snippet of a larger, ongoing project on the energy demands of mobile connectivity, the Internet, and our digital lives.

The first photo at the top of this post shows in wide-angle two sides of the Terminal Commerce Building, from the opposing sidewalk. If you head east on Callowhill, that one way street running right to left in the photo, past the block long, barren expanse of the building itself, you would reach into the energy-focused side of the data center. The first thing you’d notice is the hum of massive air-cooling units, each the size of a shipping container (see below for a street-level view of one of these units). Continue past the data center and you would reach an electricity sub-station. The presence of the sub-station highlights both how much energy is consumed to power this part of ‘the cloud’, so much that it links directly into the power grid itself.

Cooling equipment directly behind Terminal Commerce Building, on north 13th Street. There are many, many more of these cooling units in the area. October 2014.

In addition to the digital geography of ‘the cloud’, the constant, pervasive creation of wireless connectivity requires enormous amounts of electrical energy. ‘The cloud’ or cyberspace, whatever we term it and I’ll go back and forth in this essay, is co-produced through both digital systems and the electricity that enables the storage and transmission of digitized data. Moving ‘the cloud’s’ services to and between users necessitates an immense and dispersed energy infrastructure of regionally integrated electricity generation powering globally networked digital infrastructure. As discussed earlier, the storage of data in particular creates a distributed ecology of servers, fiber-optic cables, and network equipment that form an infrastructural geography of ‘the cloud’; this infrastructural geography has a material form in buildings like the Terminal Commerce Building in North Philadelphia. Maintaining and transferring this data around requires enormous amounts of electrical energy; in turn these electrical systems have their own interconnected infrastructures that are often found in close proximity to data centers.
Given how often and how quickly the technologies of ‘the cloud’ evolve and change, it is difficult to gauge the full measure of energy demand of ‘the cloud’. It is likely much of the facts and figures presented below are outdated, but they still offer a glimpse into the energy systems associated with ‘the cloud’ and its infrastructure. What is important is not finding the exact amount of energy consumed to power cyberspace, but to begin to acknowledge that distributed alongside our digital and mobile lives, always connected, always on, is also always electrical energy. Without it, there is no ‘cloud’, no cyberspace.

A PECO electricity substation behind the Terminal Commerce Building, powering cyberspace, enabling ‘the cloud’. The materiality of all this electrical infrastructure belies both the metaphors of ‘the cloud’ as ephemeral and also makes visible the mundane equipment that enables our mobile, digital lives. This electrical equipment sits adjacent to and behind the cooling equipment. Proximity to the electricity grid means the data center is less-likely to suffer from a power-outage, and if it did it would not last as long as areas further from PECO’s equipment. Like all data centers, Terminal Commerce Building maintains multiple-levels of redundant, uninterruptible backup power supply, both battery based and with diesel generators.

The energy demand of ‘the cloud’ is not insignificant:

“Direct electricity used by information technology equipment in data centers represented about 0.5% of total world electricity consumption in 2005. When electricity for cooling and power distribution is included, that figure is about 1%. Worldwide data center power demand in 2005 was equivalent (in capacity terms) to about seventeen 1000 MW power plants” (Koomey 2008)

“Worldwide electricity consumption for all communication networks is 1.8% of total energy use.” (Lambert et al. 2012)

“As of 2007, the average datacenter consumed as much energy as 25,000 homes. There are at least 5.75 million new servers deployed into new and existing data centers every year. Data centers account for at least 1.5% of US energy consumption and demand is growing 10% per year,” (Bartels 2011)

Globally, these digital warehouses use about 30 billion watts of electricity, roughly equivalent to the output of thirty nuclear power plants (Glanz 2012). Interestingly and importantly when considering this energy use, data centers “were using only six percent to twelve percent of their electricity powering their servers to perform computations. The rest was essentially used to keep servers idling and ready in case of a surge in activity that could slow or crash their operations” (Glanz 2012). Of that output equivalent to thirty nuclear power plants, only 1.8 to 3.6 of those nuclear power plants energy output went to delivering ‘the cloud’s’ services. The rest went to keeping the ‘the cloud’ able to transmit some piece of data to a user in the split-second connection we have come to expect for accessing our digital services.

Digital infrastructure requires the input of significant amounts of electrical energy with its own geography and material impacts, from power plants, nuclear and otherwise, to high-voltage transmission lines and urban sub-stations routing energy to data centers themselves. ‘The cloud’, while accessible, essentially, everywhere around the world, is inseparable from regional electrical power grids, which are a very grounded, equipment-heavy, resource-intensive infrastructure: not cloud-like at all. The bones of cyberspace are not just data centers, network equipment, fiber-optic cabling, and cellular antenna sites, but also and more importantly, the electrical energy that powers these interconnected digital systems.

The research underlying this essay was supported by an international research fellowship at the DEMAND Centre at Lancaster University. If you enjoyed this essay, check out plenty more writing about infrastructure and urbanization at my blog: http://www.everydaystructures.com/

A final photo to leave you with: at the south-east corner of the Terminal Commerce Building is a long-closed United States Post Office. According to an essay at Hidden City Philadelphia, the building had its own zip code at one time. Commenters mention that the post office was open as late as the 1990s. The transformation of the building into a data center necessitated the grafting-on of electricity and cooling equipment. The shift to digital uses of the building and the subsequent loss of workers employed within meant the closure of the post office, even though information, now in digital form, still moves in and out of here.

A closed United States Post Office on the back-side of the Terminal Commerce Building. The structure’s utility shifted from warehousing furniture and other physical objects, or holding offices that workers would come and go from daily, to now holding servers containing cyberspace/’the cloud’. The transmission of information in and out of the building has always mattered, even when that movement was paper-based and involved the post office. February 2011.

18 December 2015

The urban, infrastructural geography of ‘the cloud’

The little, vertical white rectangles on the topmost, towering part of the boarded-up warehouse are cellular antennas, connecting a passing cellphone or smartphone to ‘the cloud’. The pervasive connectivity that locative media like Google Maps rely on is produced through network equipment like this, bolted onto otherwise unused, tall buildings in the de-industrialized urban landscape. North Philadelphia, October 2010.
Data takes up space. The space it takes up — and the water, land and electricity that get used in taking it up — remains, for the most part, out of sight, out of mind and utterly uninteresting to actually look at.
-Ingrid Burrington, The Cloud is not the Territory
The relationship between data to space extends beyond the network equipment, services, and mobile devices that transmit and present information to a user. Pervasive wireless connectivity and ubiquitous computing, as ‘the cloud’ are central, common elements of contemporary urban life. Data centers translate, as it were, between individuals and their experience of the city by mediating experiences through digital augmentation. An example of this is Google Maps’ locative ability to place the user on the map and then orient said user to wherever they need to go. While data is largely immaterial except in the action it enables, like getting you to your meeting with that map, the storage, maintenance, and transmission of data require many layers of interfacing telecommunication infrastructure that function nearly everywhere but are always, inherently embedded in particular places.

In daily life, we typically experience ‘the cloud’ as the latent potential of data and digitized information in general to do things for us: to tell us information such as the weather today or when a bus will arrive, how to navigate between new places, or as a a quick glance at a document stored between multiple devices before a meeting. While we experience this data through a screen, whether smartphone or a larger-screened tablet/laptop/desktop, ‘the cloud’ as this electromagnetic, immaterial but impactful thing resides in specific places. The creation, maintenance, and transmission of data to and between users necessitates an immense, world-spanning telecommunications infrastructure that, as Ingrid Burrington’s recent writing for The Atlantic has explored in wonderful detail. ‘The cloud’ is extremely physical and not at all ephemeral nor ‘cloudlike’. For ‘the cloud’ to function requires a distributed, interconnected system of servers, fiber-optic cables, and network equipment such as cellular antenna sites, wi-fi routers, and so on that form an digital, infrastructural geography of data, a geography of ‘the cloud’. Even though this geography is out of sight and relatively uninteresting to look at (especially compared to the bits and bytes of email, social media posts, and cat videos contained within the digital infrastructure), it is a tangible antithesis of cloud-based metaphors. 

Buried fiber-optic cable markers alongside CSX railroad tracks in central Philadelphia. February 2010.

The data centers that house the websites and other ‘cloud’ information we access daily are layered in the urban landscape even though the digitized utility of these spaces is not readily apparent. There is no symbology to data centers: no wi-fi symbol equivalent, for instance, to indicate what is housed within. While many of the data centers that encase ‘the cloud’, as the digitized, information-holding core of the Internet, are in out of the way places, in the suburbs or way past the suburbs, there are also many data centers within cities (http://www.datacentermap.com/ is a great source for learning more about where data centers are located). Some data centers are in re-used, industrial-era structures that were built long before our contemporary, digital-era began. Urban or not, data centers are often located where they are due to proximity to railroads and the industrial city that relied on rail to move goods around. The railroad track is itself not consequential to ‘the cloud’, but the railroad track’s right of way is of consequence because that is where the fiber-optic cables are laid, cables that move ‘the cloud’ between a data center and the user.
Data centers connect individual users but are typically separated from their proximate neighborhood, embodying the juxtaposition between highly designed, connective objects in the form of smartphones and other cloud computing devices, and the quotidian landscapes we all inhabit. We have shifted our information storage, retrieval, and processing needs into digital devices, but the information still resides somewhere, and in many cases, that somewhere is, today, a data center.

Top: 60 Hudon Street in Lower Manhattan (March 2010) and Bottom: One Wilshire in downtown Los Angeles (July 2014), two of the most important nodes in ‘the cloud’, at least in North America.

The best-known data centers, that both also act as co-location points, are likely 60 Hudson Street in Lower Manhattan and One Wilshire in downtown Los Angeles, the former originally the global headquarters of Western Union and the latter originally an office building, now both are among the most important nodes in the global Internet. Most of the Internet traffic between North American and Europe passes through this building, as well as data going elsewhere. Given New York City’s prominence for the global economy, there are a number of data centers in and surrounding the city, but all the major cities of the Northeast United States have these pieces of ‘the cloud’ nearby. Chances are good an email or a packet of digital information heading to Europe passes through this building.

AT&T buried fiber optic cables alongside the Amtrak Northeast Corridor train tracks in West Philadelphia. April 2011.

If you head south 100 miles from lower Manhattan into downtown Philadelphia, on an Amtrak Northeast Corridor train for instance, you will be running parallel to fiber optic cables buried in the gravel alongside the tracks. Once in Philadelphia you will find a number of urban data centers, including the Terminal Commerce Building.

Housing ‘the cloud’

The Terminal Commerce Building on the left, one of the largest data centers in the Northeast United States. The former offices of the Philadelphia Inquirer on the right, and City Hall in the center distance. Note the collection of backup power generators and cooling equipment poking up above the roof-line. A local real estate developer has proposed turning the currently-vacant Inquirer Building into a hotel, effectively moving the downtown’s commercial and foot-traffic many blocks north, and perhaps bringing some jobs back into the neighborhood. July 2014.
The eleven story, 1.3 million square foot Terminal Commerce Building sits just north of Philadelphia’s City Hall, across from the former offices of the Philadelphia Inquirer and the Philadelphia public school district’s main offices at 410 North Broad Street. This Art Deco building takes up an entire city block; it was completed in 1930 and originally contained wholesale furniture showrooms and warehousing for furniture. It has a railroad spur going into the basement to load and unload goods; this is where the fiber optic cables run into the data center, this otherwise unused railroad right of way cutting through downtown is what makes the building useful as a data center. According to a report submitted to place the building into the National Register of Historic Places, a status granted in 1996, the Terminal Commerce Building was “large enough to command its own post office, and later its own zip code, [and] by 1948 the building housed 175 companies employing some 5,000 people”. By the early 1970s this was the largest single office building in the city. Hidden City Philadelphia has a very detailed recent article about the building’s history here.

Since 1978 Sungard has occupied part of the building, offering offsite data storage and recovery in case of disaster (this information also comes from the above-mentioned report). Sungard is one of the larger global data storage and recovery firms, listed on the Fortune 500 and headquartered in suburban Philadelphia (Sungard was recently bought by FIS Global, a “financial services” company). While Sungard has remained in the building and today occupies 50% of the floorspace, the rest of the structure has slowly filled in with other data storage companies. By the late 1990s became a co-location point center where multiple carriers house their respective data centers in one building. Today there are approximately eighty networks in the building (to be clear, I’m not sure what constitutes a network in this citation taken from an article about the building’s new owner), which also acts as an interchange between data storage providers and telecommunications companies. Parts of ‘the cloud’ are both stored in this building directly, replacing couches and tables with servers and cables, moving bits of light instead of chairs and bed frames.

Whereas the building as a furniture showroom opened to the street with window displays and pedestrian traffic in and out, today the building presents frosted windows, closed metal doors, and a guarded entrance onto the main north-south thoroughfare through Philadelphia. The typical people seen outside are white male network engineers and technicians that maintain the computers, on a smoking break or walking to and from their cars. Data moves in and out, but a user of that data could not walk in off the street to ‘see’ their data without arranging to do so beforehand.

The social and economic utility of ‘the cloud’, of data centers holding key elements of the information-knowledge economy and our networked society is tied into processes of de-industrialization and economic restructuring. By virtue of the repurposing of industrial-era buildings, ‘the cloud’ as a particularly urban matter stretches from the multitude of users of data whether in the Philadelphia region or literally anywhere else around the world. We do not have to know which data center our own particular data is held within to be connected to buildings like the Terminal Commerce Building. That is the nature of ’the cloud’: it is everywhere and nowhere, but also in specific places like this.

Additionally, this data center and co-location node represents the manifestation of the new, information and innovation economy, which is largely inseparable from digital forms of connection to data and to dispersed social networks. As the industrial-era usefulness of inner city Philadelphia was to centralize, in Terminal Commerce Building, the storage and selling of physical goods, in the information economy the location still matters but for different reasons, reasons that do not have the same sort of impact on the neighborhood. The shift to a data center mirrors the shift in employment opportunities out of North Philadelphia, away from industrial warehouses and manufacturing in what was once called The Workshop of the World, into a revitalized downtown business including successful medicine, education, and tourism fields, but these workers today have no need to physically come to the neighborhood surrounding the data center. Their laptops and smartphones may access data within the building, but that is the extent of it.
The back side of the Terminal Commerce Building, taken from the railroad spur’s right of way that the carry the fiber-optic cables into the building. The rail spur connects into the Reading Viaduct, an abandoned elevated railway that supporters are working to turn into a park similar to the High Line in Manhattan. April 2015.


The disposition of data to transform social exchange is predicated on pervasive connection to global telecommunication networks, where the ‘high design’ of digital, mobile technologies like an Apple iPhone function through mundane and distributed infrastructural landscapes. These in-between, typically out of sight spaces transfer information across distance while also affecting proximate space in consequential ways. ‘The cloud’ may be everywhere a cellular or wi-fi signal reaches, but it also inhabits real, particular places like the Terminal Commerce Building. While data is largely immaterial except in the action it enables, the storage, maintenance, and transmission of data requires many layers of interfacing systems that are always, inherently embedded in particular places. Understanding the impact of ‘the cloud’ on cities necessitates conceptualizing the relationship between digital infrastructures and the urban fabric. Furthermore, data storage, maintenance, and transmission are themselves part of wider, longer-term patterns of economic restructuring and post-industrial, inner-city urban transformation. Data centers’ re-use of industrial-era structures ties the twenty-first century utility of ‘the cloud’ into nineteenth and twentieth century economic processes; buildings designed to hold heavy equipment or warehouse physical goods are useful for siting the heavy computer servers that hold ‘the cloud’, and inner city locations designed around proximity to a population of workers for manufacturing jobs are, today, often located close to central business districts and the needs of information-heavy, twenty-first century enterprise.

‘The cloud’s’ functionality, and the personal, mobile devices through which this data is consumed are inseparable from this infrastructural back-end. Without constant connectivity to data centers, ‘the cloud’ is nothing. While the location of data centers does not inherently matter, only the connection, data still embodies particular spaces, forming an infrastructural geography layered within and between cities, co-produced through or alongside other infrastructures such as transportation — especially the railroad — even within city streets where that railroad has been absent for decades.


Two final images, Top: one of a small homeless camp alongside a buried fiber optic cable, underneath Walnut Street in central Philadelphia (April 2011) and Bottom: a Verizon truck outside the Terminal Commerce Building (April 2015).

28 September 2015

the visual melody of infrastructure

A mini-hydroelectric station on River Brathay, Lake District United Kingdom. Photo by Alan Wiig, July 2015.

Teju Cole has a nicely digressive essay in this past Sunday's New York Times Magazine, Far Away from Here, that is a study of travel and photography in Switzerland. Cole writes:

Only direct observation can reveal [the particularities of places]. The way streetlights and traffic signs vary, the most common fonts, the slight variations in building codes, the fleeting culture of ads (different in each place, even when the company is a multinational), the noticeable shift in the range of hues that people wear in a given city, the visual melody of infrastructure as it interacts with terrain.
This last piece of the sentence is a wonderfully vivid description of infrastructure as a part of a landscape rather than a distraction from it. In certain settings, infrastructure both frames and focuses the wider scene, gathering the eye on particular details, then expanding the view to the way the equipment, the system, the concrete, the steel, the brick, the plastic, makes the place. The nature of infrastructure is to universalize. This is the work of the International Organization of Standardization among others, to make sure mobile phones connect in different countries, that automotive tires fit to rims, that steel and glass conform to pre-determined codes. But in the landscape, infrastructure takes on local character as it is adapted into particular cultural contexts, and these adaptations offer a "visual melody" that a makes place-difference distinct.

09 April 2015

Data Across Scales: The Geography of Data Centers

One Wilshire in downtown Los Angeles, one of the most important data centers in the world. The potential of data to reshape design requires the infrastructural landscape produced in and through spaces like this. Photo by Alan Wiig, June 2014.

Harvard University's Graduate School of Design is hosting an international, interdisciplinary conference next Friday, "Data Across Scales: Reshaping Design". I'll be speaking in the 7pm session about the geography of data and the impact of data centers and colocation points, examining the storage, maintenance, and transmission of data as an urban concern. Details on my presentation follow.

The disposition of data to transform social exchange is predicated on pervasive connection to global telecommunication networks, where the ‘high design’ of digital, mobile technologies like an Apple iPhone function through mundane, distributed, operational landscapes that transfer information across distance while also affecting proximate space in consequential ways. Understanding the impact of data on the planetary, networked urban condition necessitates conceptualizing the relationship between data infrastructures and the fabric of the city.

The relationship between data to space extends beyond the sensors, services, and mobile devices that transfigure information into data. Pervasive connectivity and ubiquitous computing are central, common elements of contemporary urban life. Data centers act as objects of translation between individuals and the city. While data is largely immaterial except in the action it enables, the storage, maintenance, and transmission of data require many layers of interfacing telecommunication infrastructure that function across scales but are always, inherently embedded in particular places. Data centers, as a central element in the dispersal of data-based decision-making, operate within a variety of spatial contexts. Data centers connect individual users but are typically separated from their proximate neighborhood, embodying the juxtaposition between highly designed connective objects in the form of smartphones and other mobile computing devices, and the quotidian landscape of United States urbanism. I argue that understanding the impact of data on the planetary, networked urban condition necessitates conceptualizing the relationship between data infrastructures and the fabric of the city.

Information technologies and the data they function through have absolutely impacted urban life, yet the disposition of data to transform social exchange is predicated on infrastructural relationships with their own place in the built environment. The ‘high design’ of digital, mobile technologies like an Apple iPhone function through mundane, distributed operational landscapes that transform distance while also affecting proximate space in consequential ways. Visualizing data as a material construct is a means of considering the spatial effects of data.

This presentation investigates the spatial consequences of data through the data centers, fiber-optic cables, and cellular antenna sites enabling mobile, ubiquitous connectivity. These embedded systems form an elemental part of data-driven urban exchange, even as they often remain relatively invisible. Comparative examples of data centers, submarine and terrestrial fiber-optic cabling, and related digital infrastructure will be shown, drawn from fieldwork in California and the North East US over the last five years. A case study of the public face of data centers in Philadelphia will also examine the Philadelphia Navy Yard, a ‘smart city’, globalized free zone highly integrated into data networks across scales, as well as touching on the relationship of data centers and data in general to energy and electrical infrastructure. The centrality of data to our experience of the physical world, to debates over open governance and civic exchange, must also engage with the aesthetic of and politics of data centers themselves.

07 April 2015

Discussing African urbanism and infrastructure in the de-industrialized Northeast USA

Photo by Jonathan Silver. 

My colleague Jonathan Silver (@invisiblemapper) of Durham University's Geography Department is giving two talks next week, one on Monday afternoon at Temple University and one on Wednesday afternoon at Clark University. Details below; the talks are open to the public and I can answer questions directly or through the comments on this post. Note as well that the Temple University talk will be followed with a walking tour that I will lead to the site of the Stetson Hat Factory in North Philadelphia to compare Jonathan's research on African urbanization and the provision of networked infrastructures with the state of infrastructure and the urban condition in inner city Philadelphia. The event at Clark University will proceed as a walking lecture examining the topic through Main South, Worcester.

Talk One

Sponsored by Temple University Geography-Urban Studies, Graduate Student Association Colloquium Series:

Incremental infrastructures: Material improvisation and social collaboration across post-colonial Accra

Monday April 13, 2015, 3:00-5:00 PM
Weigley Room, Gladfelter Hall 9th Floor
Approaching the informal construction and extension of infrastructures through the terrain of “the incremental” opens up new platforms of analysis for post-colonial urban systems. This refers to ad- hoc actions on the part of slum dwellers to connect to energy networks or carve out informal living spaces. In this seminar Jonathan Silver argues that incrementalism is produced and subsequently secured and scaled through material configurations that seek to test and prefigure new forms of infrastructure and accompanying resource flows. Using a case study of energy and housing systems in a low-income neighborhood in Accra to define and examine these incremental infrastructures the seminar examines shifts in the Accra energy network as urban dwellers rework connections to flows of electricity and consider the material adjustment of housing and the role of cooperation in responding to threats of demolition and displacement. Together, incremental infrastructures and the ways that they are constituted may articulate a prefigurative politics in which residents seek to generate access to new infrastructural worlds.

The talk will be followed by a ‘walkshop’, co-organized with Alan Wiig (Temple University Geography- Urban Studies) exploring these issues in a North American context in the neighborhoods surrounding the university.

Talk Two

Sponsored by Clark University's Graduate School of Geography

Postcolonial urbanisms and a comparative theory of infrastructure: A walking lecture

Wednesday April 15 from 1:15-2:30pm
Meet at 1:15 outside the Geography Building
Across the social sciences, the growing importance of understanding cities through the infrastructures that sustain urban life create new challenges in how we research global processes of urbanization. Despite this attention there has been surprisingly little work that has sought to compare the various conceptions of infrastructures in ways that could develop a global theorization, relevant to the global North and South. This walking seminar is predicated on asking a series of questions concerning infrastructures from a post-colonial perspective. Can theories of infrastructure in cities of the global South help in understanding urban infrastructures across the global North? For instance, are there commonalities between informal, slum neighborhoods in Africa and de-industrialized cities in the Northeast United States? Does the changing nature of infrastructure operation, maintenance, provision and politics generate the need and context for a new global (urban) theory of infrastructure? This walking lecture seeks to develop an empirically and conceptually informed response to such questions by engaging with the landscape of Worcester. We will meet outside the Geography Building and proceed through the Main South, returning to Clark University. Wear comfortable shoes, bring a camera and a notebook.

Jonathan Silver is a post-doctoral researcher at the Department of Geography, Durham University. His PhD at Durham sought to understand the political ecologies of networked systems across Accra and Cape Town’s electricity network. Jonathan is currently working on the SAMSET project investigating energy geographies across cities in Uganda, Ghana and South Africa together with work on understanding post-colonial urbanisms through work on sanitation and everyday infrastructures.

02 April 2015

New essay published: IBM's smart city as techno-utopian policy mobility

A view of Providence's smart city downtown redevelopment at sunset, December 2014. Photo by author. A full set of photos of Providence's smart city project from 2012, 2013, and 2014 can be found on Flickr.

The latest issue of City: analysis of urban trends, culture, theory, policy, action was published yesterday. It contains an essay of mine about the work of IBM's Smarter Cities Challenge to improve urban problems through information technology-focused policy solutions. I'm happy to finally be able to share this work publicly. A link to the pre-press copy of the essay is at the bottom of the blogpost, as are details about contacting me for the final version. The publisher, Taylor & Francis, offers fifty free downloads of the essay here.
 IBM's smart city as techno-utopian policy mobility


This essay explores IBM’s Smarter Cities Challenge as an example of global smart city policymaking. The evolution of IBM’s smart city thinking is discussed, then a case study of Philadelphia’s online workforce education initiative, Digital On-Ramps, is presented as an example of IBM’s consulting services. Philadelphia’s rationale for working with IBM and the translation of IBM’s ideas into locally adapted initiatives is considered. The essay argues that critical scholarship on the smart city over-emphasizes IBM’s agency in driving the discourse. Unpacking how and why cities enrolled in smart city policymaking with IBM places city governments as a key actor advancing the smart city paradigm. Two points are made about the policy mobility of the smart city as a mask for entrepreneurial governance. 1) Smart city efforts are best understood as examples of outward-looking policy promotion for the globalized economy. 2) These policies proposed citywide benefit through a variety of digital governance augmentations, unlike established urban, economic development projects such as a downtown redevelopment. Yet, the policy rhetoric of positive change was always oriented to fostering globalized business enterprise. As such, implementing the particulars of often- untested smart city policies mattered less than their capacity to attract multinational corporations.


The smart city has arrived, albeit unevenly and in different manifestations, through the continued implementation of information technologies in mediating urban governance, civic exchange, and the flow of people, goods, and data through cities (Hollands 2008; Luque et al. 2014; Townsend 2013). The critical engagement with the smart city in urban scholarship articulates the problematic role major information technology corporations play in pushing this ‘techno-utopian’ vision of urban change (Luque et al. 2014, 74). This paradigm advances a ‘smartmentality’ (Vanolo 2013) of urban management through data-driven metrics, verging on a new era of ‘governing through code’ (Klauser et al. 2014). IBM has emerged as a leading proponent of the smart city discourse over the last five years, and the global information technology corporation worked at becoming what Söderström et al. (2014) term an ‘obligatory passage point’ delimiting and defining the smart city governance paradigm (citing Callon 1986). And yet, little of this scholarship looks beyond the policy narratives of smart city initiatives to actively ground the work in cities that adopted these policies (Kitchin 2014a). 
The techno-utopian vision of and discourse around the smart city matters, a point Söderström et al. argue (2014), but the agency of this discourse, and IBM’s role in furthering the discourse, must be balanced against the rationale cities gave for enrolling in smart city policymaking to begin with. For instance, IBM’s bombastic and prolific promotional materials, white papers, and policy reports all offer a rhetoric of transformative change that is not necessarily reflected in the outcomes of the initiatives, a point that will be expanded on below. The continued critique of the smart city must be mediated through examinations of how the policies are assembled, adapted, and implemented. 

By charting the evolution of IBM's Smarter Cities Challenge, a three-year event beginning in 2010 where IBM donated their consultation services to strategize technology-driven solutions for a variety of urban problems, this essay examines IBM's smart city policymaking as an example of globally-circulating policy mobilities (McCann and Ward 2011). Via the Challenge, cities adopted IBM’s smart city proposals to achieve what McCann (2013) termed ‘extrospective policy boosterism’ using policy, in this case untested techno-utopian policy, to highlight ambitious economic potential. The smart city acted as a mask for entrepreneurial governance strategies. Instead of improvements specifically targeted to business enterprise such as downtown redevelopment, the policies proposed widespread urban change through digital augmentation. The cities that participated in the Challenge were able to present an image of competitive, creative, and strategic governance immediately following the global financial crisis, a time when municipal budgets were cut by shrinking tax revenues. Successfully enacting IBM’s policies was not necessarily a city’s priority. 

With a case study of Philadelphia’s smart city initiative, Digital On-Ramps, I argue that aligning policy rhetoric with a city’s needs was much more complex than implementing a technological fix. IBM’s Philadelphia initiative called for solving chronic underemployment among 600,000 residents in the de-industrialized inner city through a workforce education software application (typically called an ‘app’) (IBM 2011a). This was a step in the longer, messier process of engaging government, non-governmental organizations, private enterprise, and community stakeholders in creating job opportunities in Philadelphia. This smart city app could be useful to city residents, but as proposed by IBM and celebrated by Philadelphia’s mayor it was successful primarily as a promotional device. Implementation of the smart city policy was secondary to the utility of the initiative in selling the city as a promising location for globalized enterprise to set up businesses. For this particular presentation of the smart city, the intended audience was outward-focused beyond the city and its residents, intended to signify the city as smart much more than to advance a new regime of data-driven urban governance. Smart city policymaking proposed citywide benefit, instead of direct benefit to business enterprise through downtown redevelopment, but the overall goal was to signal the city as attractive to global business. 


To continue reading, you can download a pre-production version here. The final, published version is available at City's website. Leave a comment, reach out on Twitter @alanwiig or email me - details in the About section at the bottom of the page. 

Wiig, A. 2015. IBM’s smart city as techno-utopian policy mobility. City 19 (2-3): 258–273.

24 February 2015

Upcoming talk: The smart city is a techno-utopian fantasy

Philadelphia City Hall, 2013. Photo by Alan Wiig.
This coming Thursday 26 February I've been invited by Cities@Manchester to speak. My talk will critique the smart city through a case study of Philadelphia's recent work with IBM.

The smart city is a techno-utopian fantasy: A case study of IBM in Philadelphia

Smart city initiatives have been adopted worldwide, proposing techno-utopian solutions to urban problems big and small. These policies are indicative of the digitization of urban life, where social and economic exchange rely on globalized telecommunications networks and digitally-focused governance strategies. Propelled through events such as IBM’s Smarter Cities Challenge, the smart city offers a data-driven logic of widespread benefit to a city and its residents that masks the underlying advancement of entrepreneurial development objectives. The rhetoric of intelligent, transformative digital change works much more to “sell” a city in the global economy than to actually improve urban conditions. Through a case study of IBM’s policy consultation in Philadelphia, I argue that the promotional capacity of the smart city worked to drive economic growth in specific, already-well-off areas of globally-oriented enterprise. Philadelphia proposed an online application (app) for training up to 600,000 low-literacy residents for jobs in the information and knowledge economy, while at the same time highlighting as “smart” everything from the city’s relative location in the megalopolitan Northeast United States to the ready potential for creating new industries in the region. By tracing the smart city initiative into the fabric of the city itself, this presentation highlights the inconsistencies of a policy aimed at reducing inequalities through enterprise that cannot support the individuals in need. The spatial consequences of the smart city in Philadelphia reinforced a zone-based urbanism more-typically found in Asia or the Middle East. In its supposed efforts at civic engagement, the smart city instead furthered the splintering of the urban landscape.

Further details can be found here.

22 February 2015

Upcoming talk: Batteries, chargers and plugs: Charting the energy demands of mobile communication

This picture I took on SEPTA's Regional Rail in Philadelphia last year exemplifies this emerging social practice: this student (I assume as much since he got off the train at Temple University's stop) chose the seat by the door to access this electrical outlet, to top off his smartphone before heading to class. The outlet apparently works, but it decidedly was not intended for this use. Photo by Alan Wiig, 2014. 

For the last week I've been in residence at Lancaster University's DEMAND Centre, working on a research project examining the energy demands of mobile communication and digital connectivity, specifically focusing on changing social practices around charging mobile devices in public, while in transit. This coming Wednesday February 25 I'll be giving a talk on the research.


Batteries, chargers and plugs: charting the energy demands of mobile communication 

The use of smartphones to access the Internet while on the move is a common aspect of everyday, personal mobility in the twenty-first century. Transit, weather, and social media applications (apps) engageindividuals' attention during commutes, but the energy demands needed to power computing devicesleads many users to employ creative, informal actions around charging batteries in inappropriate places. There is a growing pressure fortransportation planners to provide electrical outlets and charging stations systemwide. Using fieldwork conducted on trains and in train stations in Northern England and the Northeast United States prompts the following line of inquiry: What is the role of electrical energy and batteries in enabling both mobile communication and ubiquitous connectivity while in motion? How is the social practices around constant, wireless connectivity shaping the spaces of transportation? What are the infrastructural provisions that transit providers must now consider around plug loads for battery charging? And finally, what are the larger implications of battery charging and general electricity use for patterns of local, regional, and global mobility?

If any readers are in the area the talk is open to the public. Further information here.

16 November 2014

the materiality of planetary urbanization

A Sunoco gas station in Worcester, Massachusetts. November 2014.
If a global urban age is indeed currently dawning, this circumstance cannot be understood adequately with reference to the formation of global cities or large-scale megacity regions but requires systematic consideration of the tendential, if uneven, operationalization of the entire planet--including terrestrial, subterranean, oceanic, and atmospheric space--to serve as an accelerating, intensifying process of urban industrial development. Insofar as the dominant model of capitalist urbanization continues to be based upon the generalized extraction, production, and consumption of fossil fuels, it is directly implicated in a form of global ecological plunder that has permanently altered the earth's soils, oceans, rivers, and atmosphere with unprecedented levels of pollution and toxic waste. (Brenner 2014, 47)

Planetary urbanization may be the manifestation of “the ongoing creative destruction of political-economic space under early twenty-first century capitalism” (Brenner 2013, 94). Urbanization as a capitalist economic phenomena, especially at the scale of the planet, is produced through many infrastructures layered and interconnected through and between cities and the greater landscape. These networks create the urban of planetary urbanization. Some infrastructures are more global than others: petroleum in particular illustrates the underlying systems literally and metaphorically fueling planetary urbanization. The extraction of crude oil, its distribution through transmission pipes and oil tankers, ports, routes of oceanic travel, refineries, then dispensation to gas stations is a continuous process crucial to planetary urbanization. At the station, the end user in a private automobile withdraws and pays for the fuel that facilitates automobility (the typically credit card-based payment for the fuel ties the individual into another planet-spanning infrastructure: the telecommunications systems for banking and finance that process the payment). 

Gas stations operate as components in a distributed terrain transcending any one city as it enfolds many distinct, material spaces within a repeated pattern, a logic of "spatial products" (Easterling 2005) that sit at the core of planetary urbanism. As Brenner's long quote argues, "planetary" condition is created through fossil fuels, which are consumed in any number of ways, the most visible via gas stations. Gas stations are the tangible point where fuel appears, transferring from underground storage tank to a vehicle's gas tank. The fuel never appears except in a spill, even though its existence is undeniable and necessary. These quotidian filling stations are inescapable, especially in a car-centric country like the United States. Even so, they occupy particular spaces along thoroughfares, near freeway on-ramps, and so on, in a design lexicon that both accommodates cars and is similar everywhere; you always know how to interact with a gas station.   

The logic of contemporary capitalism as a global construct is not inherent: it is created and maintained. Planetary urbanization may explain the outcomes of global capitalism today, enclosing all spaces and exceeding all municipal boundaries, yet conceptualizing the theory necessitates engaging with the flows of many infrastructures that have global reach, to the point of re-conceptualizing urban infrastructure itself around the importance of, and local-to-global impacts of, these systems that constitute capitalism and consequently planetary urbanization. By its very nature there is no singular artifact or space of planetary urbanism. Applying the theory to particular places and processes we engage with regularly and see daily is, perhaps, a means of locating theory everywhere, globally. 

An Xtra Fuels gas station in Beacon, New York. November 2014.

Note about this post: 

As a theoretical framing of the contemporary urban-global condition, planetary urbanization is gaining traction through the work of urbanists Neil Brenner, his writing partner Christian Schmid, and Brenner's Urban Theory Lab at Harvard's Graduate School of Design. While their work thus far has focused on mapping territories and flows of planetary urbanization, there is a need to contextualize emblematic or representative spaces of this theory as well, the "operational landscapes" of this theory where the planetary congeals into particular, material terrains. This post is merely a starting point of an ongoing discussion that Rob Holmes of Mammoth have begun, leading up to the 2015 Association of American Geographers' Annual Meeting in Chicago, where we have organized a session engaging the debate, Infrastructure as landscape: investigations into contemporary theories of global change (description of session here).

  • Brenner, N. 2013. “Theses on Urbanization.” Public Culture 25 (1 69) (February 18): 85–114. 
  • Brenner, N. 2014. Urban Theory Without an Outside. Harvard Design Magazine 37.
  • Easterling, K. 2005. Enduring Innocence: Global Architecture and Its Political Masquerades. Cambridge, MA: The MIT Press.