Category Archives: Privacy

How drones use algorithms to govern your life

How do drones use computational methods such as algorithms to govern your life? Here are ten ways.

Many people think (non-military) drones are only used by hobbyists, and then only to fly small Go-Pro cameras around.

This is mistaken.

Following is a partial listing of other ways drones perform algorithmic calculations on people. All of these are already here. The lesson is not that drones can do this and it’s about drones; rather the lesson is that drones are being used in algorithmic governance more generally.

These are examples from my files. Mostly these are non-military/intelligence usages but that distinction is not entirely tenable given the streams of expertise and knowledge between military and non-military drone research.

  1. Drones can assess abnormal or “suspicious” behavior.

    Japanese security company Secom, starting in December, will offer a surveillance service using drones designed to detect and track suspicious vehicles and people. The drones can also take pictures of license plates and intruders’ faces as they enter factory grounds or shops at night….

    Odubiyi said there was urgent need to upgrade and add to the existing 1,000 CCTV cameras in the state to complement the other crime prevention initiatives of the government which include the Security Trust Fund, Street Signage, House Numbering and the provision of three-digit number for emergency calls….

    This scheme [ALPR but could equally well be drones] makes, literally, a state issue out of legal travel to arbitrary places deemed by some — but not by a court, and without due process — to be “related” to crime in general, not to any specific crime.

  2. Drones can monitor the environment using a variety of sensors.

    I watched as the drone’s gas monitoring sensor was checked before the aircraft was launched by catapult for a 20-minute flight around the boundaries of the site….

    A drone can be nearly any size, from as small as an insect to as large as a 757 passenger jet. It can be outfitted with technologies including high-powered cameras, thermal imaging devices, license plate readers, laser radar, and acoustical eavesdropping, see-through imaging, scent detection, and signals interception devices.

  3. Drones can physically and forcibly shepherd you, move you along, or prevent your movements. A variety of levels of force can be used.

    Some in the law enforcement community, but not all, think there may be a time where it may be appropriate to have non-lethal weapons on a drone—such things as tear gas, pepper spray, etc., where a drone will be able to fly into a location where somebody is firing from a concealed position. Or a barricaded person in the drone would be able to drop a canister of pepper spray or tear gas to get a person to come out of hiding.

  4. Algorithms will be used in drone traffic management (UTM).

    Engineers at NASA’s Ames Research Center in Moffett Field, California, are developing UTM cloud-based software tools in four segments of progressively more capable levels. They design each “technical capability level” for a different operational environment that requires development of proposed uses, software, procedures and policies to enable safe operation, with Technical Capability Level One focusing on a rural environment. With continued development, the Technical Capability Level One system would enable UAS operators to file flight plans reserving airspace for their operations and provide situational awareness about other operations planned in the area.

  5. Drones are being used by law enforcement and emergency services.

    For first responders, surveillance teams and investigators, high-quality aerial imagery provides the real-time intelligence needed to assess a situation immediately, ensure safety on the ground, and capture detailed evidence and forensics.

  6. Drones are part of Big Data and data analytics.

    Keep that in mind as you examine the secret ISR study, and you’ll see that the Pentagon’s drone program uses data analytics in almost precisely the same way IBM encourages corporations to use it to track customers. The only significant difference comes at the very end of the drone process, when the customer is killed.

  7. Drones–and robots–are being equipped with algorithms that can predict your next move before you even make it.

    The algorithm, by two University of Illinois researchers, opens the door to software that can guess where a person is headed—reaching for a gun, steering a car into armored gate—milliseconds before the act plays out.

  8. Drones can learn to sense-and-avoid.

    One, Bio Inspired Technologies of Boise, Idaho, is tackling the problem with a hard-wired neural network, a type of device that is good at learning things. This can, the firm’s engineers believe, be trained to recognise and avoid aerial obstacles. Alternatively, a conventional, if high-end, computer can be programmed with algorithms predesigned to recognise and evade threats, by understanding how objects visible to a drone’s camera are moving.

  9. The variety of uses for drones is big and ever-expanding.

    These involved areas as diverse as agriculture (farmers use drones to monitor crop growth, insect infestations and areas in need of watering at a fraction of the cost of manned aerial surveys); land-surveying; film-making (some of the spectacular footage in “Avengers: Age of Ultron” was shot from a drone, which could fly lower and thus collect more dramatic pictures than a helicopter); security; and delivering things…Because drones are cheap, geographers who could never afford conventional aerial surveys are able to use them to track erosion, follow changes in rivers’ sources and inspect glaciers. Archaeologists and historians are taking advantage of software that permits drones fitted with ordinary digital cameras to produce accurate 3D models of landscapes or buildings. This lets them map ancient ruins and earthworks. Drones can also go where manned aircraft cannot, including the craters of active volcanoes and the interiors of caves. A drone operated by the Woods Hole Oceanographic Institution, in Massachusetts, has even snatched breath samples from spouting whales for DNA analysis. And drones are, as might be expected, particularly useful for studying birds.

  10. Drones are surveillant. As such they are ideal for all sorts of new mappings. This raises privacy concerns.

    We need to impose rules, limits and regulations on UAVs as well in order to preserve the privacy Americans have always expected and enjoyed.

What we should realize if that if it can be done it will be done, as long as it is legal (and often that is very much an unknown or grey area).

Papers from “Where’s the Value? Emerging Digital Economies of Geolocation” session

The written texts from the AAG panel session I co-organized with Agnieszka Lesczczynski entitled “Where’s the Value? Emerging Digital Economies of Geolocation” are now available. The panelists were Elvin Wyly (UBC), Rob Kitchin (National University of Ireland at Maynooth), Agnieszka Leszczynksi (University of Birmingham) and Julie Cupples (University of Edinburgh).

Two were posted to blogs (linked below) and two are reproduced below. Although I posted links to a couple of these previously, this blog entry collects them all. (Two panelists, Sam Kinsley and David Murakami Wood, were regrettably unable to attend.)

Thanks again to all!

~ ~ ~

Elvin Wyly: “Capitalizing the Records of Life” (see below)

Rob Kitchin “Towards geographies of and produced by data brokers

Agnieszka Leszczynski “What makes location valuable? Geolocation as evidence, meaning, & identity” (see below)

Julie Cupples “Coloniality, masculinity and big data economies

And here again is the audio from the session.

~ ~ ~

“Capitalizing the Records of Life”
Elvin Wyly, UBC

Let me begin with a confession.  I did some homework reading the cv’s of my colleagues on this panel, and this is where I found the answer to our central question, “Where’s the value in the emerging digital economies of geolocation?”  I.  Am.  In.  Awe.  It’s here, right here, right now, in the intersecting life-paths of extraordinary human geographers coming together to share the results of labor, creativity, critical insight and commitment.  Julie Cupples’ work on decolonizing education and geographies of media convergence intersects with Agnieszka Lesczynski’s inquiry into the gendered dimensions of the erosion of locational privacy and the “new digital spatial mediations of everyday life,” and Sam Kinsley’s ‘Contagion’ project on the movement of ideas through technologically mediated assemblages of people, devices, and algorithms.  David Murakami Wood’s Smart Cities project and editorial assemblage in the journal Surveillance and Society respond directly to the challenges and opportunities in Rob Kitchin’s (2014) call in The Data Revolution for “a more critical and philosophical framing” of the ontology, epistemology, ideology, and methodology” of the “assemblage surrounding” the production and deployment of geolocational data.  And many of these connections have been the subject of wise anticipatory reflections on Jeremy Crampton’s Open Geography, where the adjective and the verb of ‘open’ in the New Mappings Collaborative give us a dynamic critical cartography of the overwhelming political and knowledge economies of spatialized information.

As I read the cv’s of my panelists, it became obvious that the value of a record of a life — that’s the Latin curriculum vitae — is the new frontier of what James Blaut (1993) once called The Colonizer’s Model of the World, and what Kinsley (2014) has diagnosed as “the industrial retention of collective life.”  Smart cities, the social graph, the Internet of Things, the Quantified Self, the Zettabyte (270 bytes) Age analyzed by Kitchin:  all of this signifies a new quantitative revolution defined by the paradox of life in the age of post-humanist human geography.  In the closing lines of Explanation in Geography, David Harvey announced “by our models they shall know us” — a new generation of human geographers bearing the models and data of modern science; today, it’s the algorithms, models, and corporations that arrive bearing humans — millions and billions of them — whose curricula vitae can be measured, mapped, and monetized at scales that are simulteneously perrsonalized and planetary.  Facebook alone curates more than 64 thousand years of human social relations every day (four-fifths of it on mobile devices and four-fifths of it outside the U.S. and Canada) and LinkedIn CEO Jeffrey Weiner (quoted in MarketWatch, 2015) recently declared, “We want to digitally map the global economy, identifying the connections between people, companies, jobs, skills, higher educational organizations and professional knowledge and allow all forms of capital, intellectual captial, financial capital, and human capital to flow to where [they] can best be leveraged.”

Capitalized curricula vitae, however, are automating and accelerating what Anne Buttimer once called the ‘dance macabre’ of the knowledge economies of spatialized information, because the deceptively friendly concept of ‘human capital’ is in fact a deadly contradiction:  capital is dead labor, the accumulated financial and technological appropriation of surplus value created through human labor, human creativity, and human thought.  Buttimer’s remark about geospatial information being “a chilly recording by a detached observer, a hollow rattle of bones” hurt — because this is what she said in a conversation to the legendary time-geographer Torsten Hägerstrand, who in the 1940s spent years with his wife Britt in the church-register archives of a rural Swedish parish to understand “a human population in its time and space context.  Here’s what Hägerstrand (2006, p. xi) recalls:

 “[We] worked out the individual biographies of all the many thousands of individuals who had lived in the area over the last hundred years.  We followed them all from year to year, from home to home, and from position to position.  As the data accumulated, we watched the drama of life unfold before our eyes with graphic clarity.  It was something of stark poetry to see the people who lived around us, many of whom we knew, as the tips of stems, endlessly twisting themselves down in the realm of times past.”

Hägerstrand wrote that he was disturbed and alarmed by Buttimer’s words, and I am too, because Allan Pred (2005, p. 328) began his obituary for Hägerstrand by quoting Walter Benjamin, emphasizing that it is not only knowledge or wisdom, but above all real life — “the stuff stories are made of” — which “first assumes transmissible form at the moment of …death.”  But just as “every text has a life history” (Pred, 2005, p. 331) that comes to an end, now Allan Pred’s curriculum vitae has also assumed transmissible form of the market-driven, distorted sort you can track through the evolving Hägerstrandian time-space prisms of the digitized network society.  Hägerstrand is dead, but he has a Google Scholar profile that’s constantly updated by the search robots, and the valorized geolocatable knowledge of his citations put him in a dance macabre of apocalyptic quantification:  he is “worth” only 1.093 percent of the valorization of another dead curriculum vitae, that of Foucault, who’s also on Google Scholar.  The world is falling in love with geography, but we don’t need more than just a few human geographers to do geography, thanks to the self-replicating algorithms and bots of the corporate cloud of cognitive capital.

The geolocatable knowledge economy is thus a bundle of contradictions and the endgame of the organic composition of human capital.  Human researchers spending years in the archives to build databases are now put into competition with the fractal second derivatives of code:  how do I balance my respect and reverence for our new generation of geographers screen-scraping APIs and coding in R, D3, Python, and Ruby on Rails without giving up what we have learned from the slow, patient, embodied labor of previous generations working by hand?  I see the tips of stems, not just in Hägerstrand’s small Swedish parish, but right here, in this room.  Tips of stems, endlessly twisting down in the realm of times past — but in today’s times where each flower now faces unprecedented competition in every domain:  jobs, research support, academic freedom, human care, human recognition, human attention.  Tips of stems, endlessly twisting through time-spaces of a present suffused with astronomical volumes of geographical data in what the historian George Dyson (2012) calls the “universe of self-replicating code.”  Tips of stems, tracing out an entirely new ontology of socio-spatial sampling theory defined by the automated mashup analytics that now combine Hägerstrand’s time-space diagrams with Heisenberg’s observational uncertainties, Alan Turing’s (1950) ‘universal machine,’ and Foucault’s archaeology of knowledge blended with Marx’s conception of the “general intellect” and Auguste Comte’s notion of the ‘Great Being’ of all the accumulated knowledge of intergenerational human knowledge, tradition, and custom.  Tips of stems, tracing lifeworlds of a situationist social physics that treats smartphones as “brain extenders” (Kurzweil, 2014) converging into a planetary “hive mind” (Shirky, 2008) while reconfiguring the observational infrastructures and human labor relations of an empiricist hijacking of positivism:  if Chris Anderson (2008) is correct that the petabyte age of data renders the scientific method obsolete, then who needs theory?

We all need theory — we humans.  Theory is the intergenerational inheritance of human inquiry, human thought, and human struggle.  Let me be clear:  I mean no disrespect to the extraordinary achievements of the new generation of data revolutionaries represented by my distinguished panelists, and all of you who can code circles around my pathetic do-loop rusty routines in FORTRAN, Cobol, and SAS.  Tips of stems, twisting themselves down into the realms of human history:  take a look around, at one of the last generations of human geospatial analysts, before we’re all replaced by algorithmic aggregation.  Yesterday’s revolution was humans doing quantification.  Today’s revolution is quantification doing humans.


Anderson, Chris (2008).  “The End of Theory:  The Data Deluge Makes the Scientific Method Obsolete.”  Wired, June 23.

Blaut, James (1993).  The Colonizer’s Model of the World.  New York:  Guilford Press.

Dyson, George (2012).  “A Universe of Self-Replicating Code.”  Edge, March 26, at

Hägerstrand, Torsten (2006).  “Foreword.”  In Anne Buttimer and Tom Mels, By Northern Lights:  On the Making of Geography in Sweden.  Aldershot:  Ashgate, xi-xiv.

Kitchin, Rob (2014).  The Data Revolution:  Big Data, Open Data, Data Infrastructures and Their Consequences.  London:  Sage Publications.

Kinsley, Sam (2014).  “Memory Programmes:  The Industrial Retention of Collective Life.”  Cultural Geographies, October.

Kurzweil, Ray (2014).  Comments at ‘Will Innovation Save Us?’ with Richard Florida and Ray Kurzweil.  Vancouver:  Simon Fraser University Public Square, October.

MarketWatch (2015).  “LinkedIn Wants to Map the Global Economy.”  MarketWatch, April 9.

Pred, Allan (2005).  “Hägerstrand Matters:  Life(-path) and Death Matters — Some Touching Remarks.”  Progress in Human Geography 29(3), 328-332.

Shirky, Clay (2008).  Here Comes Everybody:  The Power of Organizing Without Organizations.  New York:  Penguin.

Turing, Alan M. (1950).  “Computing Machinery and Intelligence.”  Mind 59(236), 433-460.


~ ~ ~

“What makes location valuable? Geolocation as evidence, meaning, & identity”
Agnieszka Lesczcynski, University of Birmingham

I want to invert the question that Jeremy and myself posed to the panel when organizing this session by asking, rather than ‘where is the value’ in geolocation, what is it that makes geolocation valuable? In contending that there are particular kinds of economies emerging around location, it is because geolocation itself is somehow intrinsically valuable, and I’d like to make some preliminary propositions to this end.

Over the last few years I have been particularly interested in the ways in which emergent surveillance practices of the securities agencies, made broadly known to us through the as yet still-unfolding Snowden revelations, are crystallizing around big data – its collection, mining, interception, aggregation, and analytics. And specifically, I’m particularly interested in the ways in which locational data is figuring as central within these emergent regimes of dataveillance. Indeed at the close of 2013, Barton Gellman and Ashkan Soltani, reporting in the Washington Post, identified at least ten American signals intelligence programmes or SIGADs that explicitly sweep up locational data – i.e., where location data is the target or object of data capture, interception, and aggregation.

  • Under a SIGAD designated HAPPYFOOT, the NSA taps directly into mobile app data traffic that streams smartphone locations to location-based advertising networks organized around the delivery of proximately relevant mobile ads, often unencrypted and in the clear. This locational data, which is often determined through mobile device GPS capabilities, is far higher-resolution than network location, allowing the NSA “to map Internet addresses to physical locations more precisely than is possible with traditional Internet geolocation services”
  • Documents dating from 2010 reveal that the NSA and GCHQ exploit weaknesses in ‘leaky’ mobile social and gaming applications that veil secondary data mining operations behind primary interfaces, piggybacking off of commercial data collection by syphoning up personal information including location under a signals intelligence program code-named ‘TRACKER SMURF’ after the children’s animated classic
  • In perhaps the most widely publicized example, the NSA collects over 5 billion cell phone location registers off of cell towers worldwide, bulk processing this location data through an analytics suite code-named CO-TRAVELLER which looks to identify new targets for surveillance on the basis of parallel movement with existing targets of surveillance – i.e., individuals whose cell phones ping off of the same cell towers in the same succession at the same time as individuals already under surveillance.
  • Just a few months ago, it was leaked that the CSE, or Canada’s version of the NSA, was tracking domestic as well as foreign travelers via Wi-Fi at major Canadian airports for up to two weeks as they transited through the airports and subsequently through other ‘nodes’ including other domestic and international airports, urban Wi-Fi hotspots, transport hubs, major hotels and conference centers, and even public libraries both within Canada and beyond in a pilot project for the NSA;
  • and, most recently, under a SIGAD code-named LEVITATION, the CSE has been demonstrated to be intercepting data cable traffic to monitor up to 15 million file downloads a day. Particularly significant in the leaked CSE document detailing this programme is that the CSE explicitly states that it is looking to location data to improve LEVITATION capabilities for intercepting both GPS waypoints and “[d]evices close to places” so as to further isolate and develop surveillance targets, including those carrying and using devices within proximity of designated locations.

So the question is, why geolocation? Why is it of such great interest to the securities agencies? And here I want to argue that it is of interest because it is inherently valuable, and uniquely so amongst other forms of PII. And this value is latent in the spatio-temporal and spatial-relational nature of geolocation data.

  • the spatio-temporal nature of many spatial big data productions means that it may be enrolled as definitive evidence of our complicity or involvement in particular kinds of socially disruptive events or emergencies by virtue of our presence, or as in the case of CO-TRAVELER, co-presence and co-movement, in particular spaces at particular times
  • furthermore, longitudinal retention of highly precise, time-stamped geoloational data traces allow for the reconstruction of detailed individual spatial histories, which like the CO-TRAVELER example, similarly participate within what Kate Crawford has recently characterized as emergent truth economies of big data in which data is truth;
  • the relational nature of spatial big data productions, in which our data may be used to discern our religious, ethnic, political and other kinds of personal affiliations and identifies on the basis of the kinds of places that we visit and the ability to establish linkages with other PII across data flows;
  • and, in this vein, the ways in which locations are inherently meaningful – for example, they may be as revealing of highly sensitive information about ourselves as our DNA. For instance, on the basis of the specialty of a medical office that we visit, this information may be revelatory of the fact that we may have a degenerative genetic disease and the nature of that disease – information that socially we otherwise understand as some of the most private information about ourselves.
  • and, of course, the ways in which location is not only revealing of identity positions, but it is identity – for example, a group of researchers determined that unique individuals could be identified form the spatial metadata of only four cell phone calls at a very high confidence level.

So in asking where is the value in geolocation, my take is that it is valuable – to both the intelligence apparatuses that I have highlighted here but also corporate entities – because it is uniquely sensitive – revealing and identifying – amongst other forms of PII.

~ ~ ~

Talk at Colgate University on Big Data

Today I’m giving a talk at Colgate University, in New York. It was a bit of toss-up if the weather would allow me to get here (snow and flight cancellations), but thankfully I’m here safely.

My talk is entitled “Big Data Narratives: Surveillance and Privacy in the Age of the Algorithm.”

Thanks to my hosts Peter Scull, Adam Burnett and the geography department. The talk is made possible by the Dennis Fund endowed lecture series in the social sciences at Colgate.

PS: small historical note. Colgate was the place Peter Gould was evacuated to during WWII when he was a child.

World Economic Forum: privacy is not dead

A post on the World Economic Forum by someone from the SAS Best Practices team says the following five beliefs are wrong:

  • “I’ve got nothing to hide.”
  • Privacy policies apply to data, not users.
  • A single privacy policy will suffice for all your data.
  • Anonymized data keeps my personal identity private.
  • Privacy is dead.

The last one is particularly interesting since it is heard all the time. While there isn’t a true rebuttal on the site, the following comment is offered:

Hence, a more accurate statement would be: Privacy is not dead. Yet. The focus on big data privacy issues is escalating rapidly – and as more people begin to understand what’s at stake, things will change and are beginning to change.

More here.

Introducing the Databox to protect your personal data

Interesting concept, described in the Guardian:

The basic concept is a Databox, a piece of software that collects personal data and then manages how that information is made available to third parties. In essence, it’s “a networked service that collates personal information from all of your devices and can also make that data available to organisations that the owner allows”.

Continues here.

Science special issue on privacy

Science has published a special issue on privacy.

At birth, your data trail began. You were given a name, your height and weight were recorded, and probably a few pictures were taken. A few years later, you were enrolled in day care, you received your first birthday party invitation, and you were recorded in a census. Today, you have a Social Security or national ID number, bank accounts and credit cards, and a smart phone that always knows where you are. Perhaps you post family pictures on Facebook; tweet about politics; and reveal your changing interests, worries, and desires in thousands of Google searches. Sometimes you share data intentionally, with friends, strangers, companies, and governments. But vast amounts of information about you are collected with only perfunctory consent—or none at all. Soon, your entire genome may be sequenced and shared by researchers around the world along with your medical records, flying cameras may hover over your neighborhood, and sophisticated software may recognize your face as you enter a store or an airport.

There’s already been a story on it at NPR about the power of metadata.

Get the whole special issue here.

Cyberwarfare and Encryption

The recent cyberhacking attacks on Sony, CentCom and a number of well known companies over the past few years (especially Home Depot and Target) highlight a major development in the online world known as cyberwarfare. Cyberwarfare is defined as offensive and defensive measures designed to achieve superiority of control of cyberspace (Internet, networks, software and hardware) or parts thereof.

Although cyberwarfare is often thought of as an activity practiced by the state, attacks against corporations and the theft of credit card data can also be considered attacks against state sovereignty and symbols of statehood, and, hence, cyberwarfare. The information released in the Sony attacks, despite it being a foreign company, largely pertained to American operations and that bastion of Americana, Hollywood. Especially as it was an American subsidiary, Sony Pictures Entertainment, based in Culver City, Ca. And of course governments can, and do, invoke such attacks as justification for new cyber crime laws, some of them pretty restrictive.

Nevertheless, it may be useful to distinguish types of cyberwarfare, or to permit that some forms of cyberwarfare (offensive or defensive) may be more easily carried out by states. An attack on the electrical power grid may be considered cyberwarfare while the theft of corporate credit card numbers may better be considered hacking. Nevertheless the line is hazy (and there is of course a long-standing form of warfare known as economic warfare).

Here are some forms of cyberwarfare I identified for my political geography class last fall.

Methods—what is cyber warfare

There are undoubtedly others and the definition is flexible enough to include “insider threats,” or those who work from within (the Sony attacks have often been attributed to insiders, although the US government has consistently maintained that North Korea is ultimately responsible.). Some of these are well-known, such as Zero-day capabilities (so called because they been worked on for zero days) and “advanced persistent threats” (APT) a euphemism for government-sponsored attacks (usually from China, the world’s most persistent source of cyber attacks, along with…. the USA).

There are some interesting geographies of cyberwarfare, including:

–who is the target of attacks (who is vulnerable)
–who is perpetrating attacks
–differential effects of attacks, including collateral damage


In my class, I suggested that we could understand cyber vulnerability as occurring at geostrategic chokepoints where the cables that carry the bulk of internet traffic come to shore. If you like, this is the new geopolitical map after Mackinder, where who controls the cables controls the global flows of traffic!

Norse, which has been the most widely quoted source of doubts about North Korea being responsible for the Sony hack, have a very nice dynamic map of live attacks here. It nicely illustrates some of the geographies involved (click map below). (They use the “honeypot” system of offering faked targets that attract attacks. For example they may have a site which looks like an unsecured piece of factory control software, or a computer without a firewall. According to their website, they have over 8 million honeypots scattered around the world.)


Chokepoints are also critical for surveillance, which is why GCHQ and NSA have programs to tap into them. Skewjack Farms near Bude in Cornwall is one such site, and some of its operations have been revealed by Channel 4 (UK) using Snowden documents.

But there are other geographies too, as this graphic about the now well-known Stuxnet infection demonstrates:


Although Stuxnet targeted the Iranian enrichment plant at Natanz by the Bush and Obama administrations in conjunction with Israel, a programming error let the virus into the wild, where it infected over 100,000 hosts in more than 155 countries, especially Indonesia, India, and Azerbaijan. This creates a new geography of collateral war damage from cyber operations.

Although geographers have not looked into these geographies all that much (Derek Gregory has glossed on Stuxnet [Update see also “Geography, Territory and Sovereignty in Cyber Warfare” by David Midson, and “Wikileaks and the Global Geography of Cyberwarfare” by Gastón Gordillo), a new paper by Robert Kaiser in this month’s issue of Political Geography attempts to document what he calls the “birth of cyberwar.” Kaiser chooses the 2007 cyber attacks on Estonia as his beginning point that galvanized western security concerns, so much so that he claims it served to elevate cyberwarfare to the same risk levels as terrorism itself. (Russia is usually held responsible for the Estonian attacks, mostly consisting of DDoS or distributed denial of service using botnets.) Certainly after this we see the US Cyber Command being established (June 2009) headquartered at Fort Meade, home to the NSA. (DIRNSA is also DIR USCYBERCOM.) However, the latter also has origins in US military’s pre-existing planned or operational cyber operations, and the command is actually a subunit of US StratCom. Additionally, its mission states states that it will only defend DoD computers and networks (not commercial).


Kaiser’s paper does raise some good issues, whether or not his chosen event is the “birth” of cyberwarfare. (Cyber attacks are known since at least the early 1980s, when the CIA placed a “logic bomb” that blew up a Soviet gas pipeline–see Thomas Reed, At the Abyss: An Insider’s History of the Cold War, p. 269). Operation Merlin, recounted in James Risen’s great 2006 book State of War may also count, although it is not strictly speaking a primarily cyber event (the CIA planned to leak doctored nuclear blueprints to Iran) but rather a Stuxnet precursor (Stuxnet being perhaps the most famous single cyberattack to date, and is surprisingly not mentioned by Kaiser).

One issue that doesn’t get mentioned by Kaiser deserves much more attention, and that is the geographies of encryption, which I’ve written about at length before. I was going to say more here, but this blog post is already pretty long, sorry ¯\_(ツ)_/¯

Suffice it to say then that I don’t think you can understand cyberwarfare without involving encryption, and in fact to some extent they overlap. Cyber vulnerabilities are in part encryption vulnerabilities. And cyber defenses are also in part about encryption. The “end-to-end” encryption that David Cameron is proposing to take away (or ban) would remove protections on privacy and increase security vulnerabilities, nicely illustrating that all these issues are bound together. So we really need to be talking about a material assemblage of cyberwarfare/encryption/privacy/security together.

While Kaiser is right to highlight cyberwarfare and help geographers start working on it, his article shows how far we have to go (indeed his article is not really about cyberwarfare, but more about what he calls discursive aporias of imagineering about cyberwarfare, or more plainly the ways the Estonian event was dealt with in terms of policy enactment). He is also pretty skeptical about deterring cyber attacks, which is hard to square with the fact that thousands of attacks are prevented every day. And yes of course there are fear mongers and exaggerations by security companies with things to sell you, but there are also other security experts out there and usable estimates of damage.

Still, as they say in footie, we’re off the mark and hopefully we can kick on from here.

CFP: Spatial Big Data & Everyday Life (AAG 2015)

Call for Papers: Spatial Big Data & Everyday Life
American Association of Geographers Annual Meeting
21-25 April 2015

Agnieszka Leszczynski, University of Birmingham
Jeremy Crampton, University of Kentucky
“What really matters about big data is what it does” (Executive Office of the President, 2014: 3).

Many disciplines, including the economic and social sciences and (digital) humanities, have taken up Big Data as an object and/or subject of research (see Kitchin 2014). As a significant proportion of Big Data productions are spatial in nature, they are of immediate interest to geographers (see Graham and Shelton 2013). However, engagements of Big Data in geography have to date been largely speculative and agenda-setting in scope. The recently released White House Big Data report encourages movement past deliberations over how to define the phenomenon towards identifying its material significance as Big Data are enrolled and deployed across myriad contexts – for example, how content analytics may open new possibilities for data-based discrimination. We convene this session to interrogate and unpack how Big Data figure in the spaces and practices of everyday life. In so doing, we are questioning not only what Big Data ‘do,’ but also how it is they realize particular kinds of effects and potentialities, and how the lived reality of Big Data is experienced (Crawford 2014).

We invite papers along methodological, empirical, and theoretical interventions that trace, reconceptualize, or address the everyday spatial materialities of Big Data. Specifically we are interested in how Big Data emerge within particular intersections of the surveillance, military, and industrial complexes; prefigure and produce particular kinds of spaces and subjects/subjectivities; are bound up in the regulation of both space and spatial practices (e.g., urban mobilities); underwrite intensifications of surveillance and engender new surveillance regimes; structure life opportunities as well as access to those opportunities; and/or change the conditions of/for embodiment. We intend for the range of topics and perspectives covered to be open. Other possible topics include:

• spatial Big Data & affective life
• embodied Big Data; wearable tech; quantified self
• algorithmic geographies, algorithmic subjects
• new ontologies & epistemologies of the subject
• spatial Big Data as surveillance
• Big Data and social (in)equality
• “ambient government” & spatial regulation
• spatial Big Data and urbanisms (mobilities; smart cities)
• political/knowledge economies of (spatial) Big Data

We welcome abstracts of no more than 250 words to be submitted to Agnieszka Leszczynski ( and Jeremy Crampton ( by August 29th, 2014.

Crawford K (2014) The Anxieties of Big Data. The New Inquiry.

Executive Office of the President (2014) Big Data: Seizing Opportunities, Preserving Values. The White House.

Graham M and Shelton T (2013) Guest editors, Dialogues in Human Geography 3 (Geography and the future of big data, big data and the future of geography).

Kitchin R (2014) Big Data, new epistemologies and paradigm shifts. Big Data and Society (1): In Press. DOI: 10.1177/2053951714528481.


Privacy–still alive down here!

Stephen Schulhofer on why privacy may still be alive:

  1. Privacy gets renewed prominence as a constitutional value
  2. The “nothing to hide” argument can be buried forever
  3. Legislative compromises are probably unacceptable
  4. Executive Branch safeguards fare even worse
  5. Warrantless section 702 surveillance of non-US persons abroad is now especially suspect
  6. The third party doctrine is badly shaken but still intact

Note especially the sixth point. You may remember from the Snowden files that the discussion about “metadata” turned in part on it being something the public knowingly concedes to thirds parties (ie telecoms) and that no individual warrants were required to obtain it.

This will be especially relevant to geolocational privacy and tracking (ie our locations, journeys and movements are potentially willingly given up to third parties, eg Google, which the government can then collect). Schulhofer notes the counterargument that it is “intensely personal” and thus subject to the Fourth Amendment (you may also remember the argument of “precise geolocational information”). But he warns:

At bottom, Riley shows the Court’s full appreciation of the threat posed by unrestricted government access to digital files, and this may ultimately prove to be its most important legacy. But the logic of the third-party doctrine still will have to be tackled head-on in situations where police get information directly from an intermediary like an internet service provider or a cloud computing service.

We need more of this kind of analysis from geographers–why aren’t we taking part in these legal debates? Surely we have something to say about geolocational privacy and surveillance.

Schulhofer: Pleasant Surprises – and One Disappointment – in the Supreme Court’s Cell Phone Decision

Very thorough and informative reading of the Riley decision. If you are not familiar with this ruling, it is the Supreme Court decision that searches of your phone if arrested are not permissible without a warrant. This piece gives several forceful reasons why this is a good ruling for privacy advocates (not a sentence one types too often!). Note the geolocational privacy implications this gives rise to.

Pleasant Surprises – and One Disappointment – in the Supreme Court’s Cell Phone Decision.