Monday, February 13, 2012

Locational privacy: beyond privacy as property and secrecy?

In April last year, it was revealed that every time an iPhone user synchronises their device with a networked computer, Apple downloaded files which contain fairly detailed locational information about the movement of its user. 

Soon after that, it was revealed that TomTom GPS devices also send locational information about their user's movements back to the company which provides the navigational services.
Map of iPhone movements, produced by Alasdair Allan and Pete Warden with their iPhone Tracker application (discussed below)

Predictably, in both cases the corporations involved have denied any sinister intentions, and have tried to re-assure folks that the data from individuals is de-identified and used only to improve services for consumers of their products. Equally as predictably, such re-assurances have not satisfied critics, who argue that these corporations have breached the privacy of their customers by keeping locational information without their explicit consent.

This question is most often asked, and answered, through the concept of privacy. And if the extent of newspaper coverage and opinion on such episodes is anything to go by, the issue of locational privacy is beginning to generate some overdue scrutiny. While most readers of these articles are assumed to be aware that their activities in 'cyberspace' might leave behind digital shadows, the fact that their movements through the city are also leaving such shadows is presented as something new that has emerged as a result of the increasing diffusion of location-aware mobile media devices such as smartphones and navigation devices.

Media discussions of locational privacy are generally infused with a nagging sense that something is  kinda wrong with these privacy incursions ... but there's not a clear sense of what exactly that is.
For the most part, these emerging locational privacy issues are presented as a problem of 'informed consent'. The assumption seems to be that as long as users are aware that digital data about their movements is being collected and stored and have given their consent to this, then there is no problem.

From this perspective, the only problem with corporations like Apple and Tom Tom collecting and storing locational data about individuals is that those individuals were unlikely to be aware that it was happening, and so could not make an informed choice. By contrast, when people broadcast their location by 'checking-in' to places via locative media applications like Foursquare or Facebook Places, there is no problem because they are doing so knowingly -- they have made an informed choice.

[There are also concerns about the security of such data, even where there is informed consent. Last year's breach of security at Sony exemplifies this risk. And even where data is de-identified, there exists the potential for re-identification in many cases. But I'm gonna leave this important issue aside for now...]

The understanding of privacy that informs this presentation of the problems of location-awareness is interesting. Privacy is almost universally assumed in such stories to be something that is traded by individuals, in return for the benefits of owning and using devices such as smartphones and SatNavs. From this perspective, as long as choice is free and informed, what individuals choose to do with their privacy is entirely up to them. The argument goes something like this: no-one is being forced to buy an iPhone or a SatNav, or to sign up with Foursquare or Facebook Places etc. So, if they value their locational privacy, they should not use the gadget and/or service. Similarly, if they are worried about digital surveillance through CCTV cameras or credit cards and are not prepared to 'trade' a degree of locational privacy for a bit of security and/or convenience, they can choose not to go to places with surveillance or use credit facilities.

This way of articulating the privacy problem has an associated policy response -- to ensure that adequate consent and notification mechanisms are established for users of gadgets and applications and places that might track/store/broadcast their movements. Once such mechanisms are in place, the problem appears to be solved.

This classically liberal presentation of privacy as something to be valued and traded by an individual has at least two limitations. First, there's the important question of whether conventional consent mechanisms are actually adequate to their task in a complex digital world. At the Engaging Data Forum hosted by MIT's Senseable City Lab a couple of years ago, Solon Barocas and Helen Nissenbaum gave a great paper about some of the significant limitations of notice and consent mechanisms in the digital realm. While their paper is primarily concerned with Online Behavioural Advertising and the tracking and targeting of individuals as they roam the internet, I think many of the points they raise are pertinent to the emerging discussions of locational privacy in the city.

Second, there's the broader question about whether individual consent really is the key issue here. If I'm the kind of person who wants to maintain some locational privacy, it's easy enough for me to not buy an iPhone. I could probably even give up my beloved 7-year old Nokia mobile phone without too much inconvenience! But here in Sydney, and in many other cities, making the 'choice' not to have my movements digitally surveilled would also mean not using public transportation systems (hello CCTV). It would mean not driving my car on freeways (hello eTags and traffic cameras). It would mean not going into any shopping malls or most shopping streets (hello CCTV and EFTPOS). It would mean not going into my university library (hello again, CCTV).  In other words, the issue is not only whether my 'choice' is informed. Even if I'm informed, do I really have the option not to consent? When a 'choice' about locational privacy means that I can't access facilities and services that are actually a part of my everyday life and citizenship, then is it really a matter of individual 'choice'?

Our 'choices' about locational privacy, then, are also constrained by the wider context in which they are made. This raises important questions: what are the 'reasonable expectations of privacy' that pertain in different contexts, and how are such expectations established, upheld and modified? These questions draw attention to the public nature of privacy. As Lauren Berlant and Michael Warner put it, "There's nothing more public than privacy". Of course, they were talking about the ideological privacy of sex and sexuality, but hey, their point holds here too -- 'reasonable expectations of (locational) privacy' are normative, and inscribed in laws and other public institutions and arrangements.

Given this public dimension to locational privacy, the extent of 'reasonable expectations of privacy' ought to be open to public debate and deliberation. But our collective capacity to conduct such debates is hampered by at least two crucial factors. First, there's the question of whether enough of us understand the new technologies and practices which impinge on locational privacy. It might now be obvious to most people that their movements in (certain parts of) the city are surveilled by CCTV cameras. But how much do most people know about the kinds of locational data kept by retailers, banks, advertisers, mobile phone companies, public services, etc?

Second, attempts to initiate debates about threats to locational privacy are often shut down by the claim that "if you've got nothing to hide, you've got nothing to fear". Here, the 'public interest' is equated with security in the face of terror and other forms of risk, and that public interest is said to outweigh most privacy concerns.

Daniel Solove (a Prof of Law at George Washington University in the US) has recently tackled this logic in a book called Nothing to Hide: The False Trade Off Between Privacy and Security (you can read a shortish article excerpted from the book here). The crux of Solove's criticism of the 'nothing to hide' logic is that it reduces the value of privacy to individual secrecy from the state. To paraphrase him, the disclosure of 'bad things' to the state is just one of a range of outcomes that may ensue from the collection, storage and/or broadcast of our location and movements. And we ought to be debating threats to locational privacy with some of these other outcomes in mind.

What are some of these other outcomes of the reduction of locational privacy that we might try to debate? Well, interestingly, one of the other outcomes frequently discussed is the risk that is generated, rather than prevented, by exposure -- this is the fear that 'good' people who make their location public might be at risk of stalking, or worse forms of criminal behaviour. (The good folks who designed the website Please Rob Me sought to draw attention to the fact that many people are basically broadcasting both their home address and the times they are not home via location-aware mobile media applications, in a manner which could be quite handy for, say, burglars.)

The potential aggregation of individually-identifiable information is another concern raised by Solove that is pertinent in discussions of locational privacy. His point here is that while we may not be too troubled by any one of the different digital shadows we leave behind in the course of a daily lives (a face on a CCTV camera here, a credit card transaction there, a Foursquare check-in here, a tag on a Facebook image there, etc etc), once those traces are aggregated, they build up a much more detailed picture of our movements and activities. As such, the question of who has the capability and authority to aggregate these different bits of data becomes an important question. Should some state agencies like the police have that authority? Under what circumstances and with what controls? Should insurance companies? Should employers?

There are some further outcomes of locational data collection and storage that also warrant debate. In  cases like the Apple and Tom Tom ones that I mentioned at the start of this piece, the collection and aggregation of data about our movements through the city is commodified through its use in the design new devices and applications, and through its sale to third parties like advertisers. Here, our movements through the city are generating a kind of surplus value that is being captured for profit by private economic interests rather than any 'common good'.

In seeking to raise these broader questions, the big question here is whether 'privacy' is an adequate concept to capture the variety of concerns that might emerge from the rapid diffusion of location-aware technologies. This has been a matter of debate for folks who have been thinking about surveillance for a while -- Colin Bennett recently published an article 'In Defense of Privacy' in the journal Surveillance and Society, and there have been a number of responses in the journal (gathered together here).

I haven't quite worked through my own position on this yet. Having thought quite a bit about publicness and the city, I'm certainly now pretty interested to think about the urban dimensions of privacy. I wonder if in urban studies, we've been so concerned with critiquing various forms of 'privatisation' that we might have missed the simultaneous threats to other forms of privacy that we might value?

I'm fairly certain that I'm not down with the 'privacy is dead and we should celebrate it' crowd. IT pundit Bill Thompson gave a really interesting talk advocating this position ("The Death of Privacy and Why We Should Welcome It") at the Lift conference in 2009, the video of which is sadly no longer available -- but there's at least an abstract here and a longer summary here. His talk was deliberately provocative -- but it seems to me that to posit the notion that 'we' have all traded in out-dated expectations of individual privacy for the benefits of smart phones and social media seems to rely on the very Enlightenment ideal of possessive individualism that it claims to supplant. I'm caricaturing his position, but he basically seems to be suggesting that the adoption of these new technological wonders has been an informed and unconstrained choice by individuals who constitute a technological vanguard that gets the benefits of it all, and need to teach the rest of us how to learn to stop worrying and love the iPhone.

In trying to wrap my head around it all, I'm especially interested to think more about the particular nature of locational privacy. The very notion of 'locational privacy' challenges simplistic assumptions about the public/private distinction which hold them to be separate spheres and places, because it implies that we might have a 'reasonable expectation' of a degree of privacy concerning our movements and activites 'in public'. This issue is addressed to some extent an in interesting piece on Locational Privacy from the Electronic Frontier Foundation.

Meanwhile, for those of you who have iPhones, Alasdair Allan and Pete Warden -- the researchers who initially revealed the tracking of iPhone users -- have developed an open source application called iPhone Tracker, which lets you map the information that your iPhone is recording about your movements.

And using this application, James Bridle has just published what looks to be a beautiful book of maps of his (iPhone's) movements, called Where the F**k Was I?. I love the essay that he's put up on his website about the book, which is very evocative of the invisible infrastructure supporting location-aware media devices that saturates the city.

Tuesday, February 7, 2012

The Sydney Alliance Transport Assembly

A couple of posts back, I offered up some reflections on the Founding Assembly of the Sydney Alliance, a coalition of unions, community organisations and faith groups working together to "advance the common good and achieve a fair, just and sustainable Sydney". A short couple of months after that event, I was involved in launching the Sydney Alliance's campaign on transport at the Transport Assembly in Penrith, one of the outer-western suburbs of Sydney.

The Transport Assembly was the culmination of six months of action-research by the Alliance's Transport Research Action Team. In May of this year, when the Alliance decided that Transport would be one of the three key issues on which it would work, the Research Action Team was formed. It's a loose, diverse group, comprised of people from across the member organisations of the Alliance who were prepared to commit to working on transport. There's been quite a bit of learning-as-we-go here. One of the first issues to confront the group was that it was difficult to find a time and place to meet that would work for everyone across the city -- especially given some of the very transport issues that the group is trying to address! To get around this, we ended up with two groups, one that has been meeting in the Central Business District and the other that has been meeting in Penrith.

Since May, Transport Team participants have been busy talking to people across the member organisations of the Alliance about their transport issues, as well as reading up on the latest research on transport and meeting with a range of transport experts based in Sydney and beyond. Personally, I've found this really interesting. I've never conducted research on transport before, so compared to some of the other participants in the group, I've had a lot of catching up to do. Those involved in the team include transport workers and advocates in community organisations like the Western Sydney Community Forum and the Cancer Council, folks from various unions and community groups who've had a long involvement in transport campaigns dating back as far as the anti-freeway green bans in the 1970s (and sometimes further!), bus drivers and railway staff who have intimate knowledge of how public transport works and how it might run better, and people like me who are new to transport but fired up to make a change, among many others.

Sydney Alliance Transport Assembly, Q Theatre, Penrith, November 2011

The event started with small groups of people making their way to Penrith on public transport from various parts of the city. These groups conducted some research on the trip, surveying commuters about their experiences of public transport and telling them about the Sydney Alliance. They also chaperoned some members of NSW Parliament who were attending the Assembly. On arrival at Penrith Station, there were bagpipes playing (seriously!!!), and a march to the Q Theatre in Penrith. After the surveys were handed in and some food was consumed, the formalities began...

One of the main purposes of the night was to launch and explain the Sydney Alliance's formula for public transport in Sydney:

400:15:1 SCA2.

This formula addresses the different dimensions of accessibility to public transport which have emerged as Alliance priorities through the initial phase of action research.

The 400 is for 400 metres -- meaning that everyone in Sydney should have some form of public transport within 400 metres of where they're at and where they want to go.

The 15 is for 15 minutes -- meaning that public transport services should come at least every fifteen minutes all day, across the entire network

The 1 is for 1 ticket -- meaning that if your trip requires you to change modes of transport, you should not need to buy separate tickets, as is currently the case. 1 ticket should buy you access to the network, not just a single line.

The S is for Safe. The C is for Clean. The first A is for Accessible. And the second A is for Affordable.

Adding all of this up, the formula addresses the different 'accessibility gaps' that characterise the existing public transport system (see Corinne Mulley and Rhonda Daniel's work on accessibility gaps in public transport). The Alliance formula also recognises that Sydney-siders' travel needs are complicated -- we don't simply travel from the suburbs to the city centre for work, although the system tends to be planned with that kind of trip in mind. Rather, in our day-to-day lives we travel to lots of places, for lots of different reasons, at all times of the day and night (only about 16% of trips in Sydney are commuting to work, which constitutes about 28% of the distance we travel -- a large chunk of our travel is much shorter trips for other purposes like shopping, recreation, and services like health and education. See the most recent stats on transport in Sydney here). As such, for public transport to facilitate universal access to a range of activities and services which are central to our lives, it has to provide a network which enables us to access the whole metropolitan area.

Here, our thinking broadly fits with some of the work done by folks like Gustav Nielsen, Jarrett Walker and Paul Mees on the so-called 'network effect' in public transport. As Paul Mees puts it, you want the public transport network to be similar to the road network. The road network is not planned to enable all possible car trips to be taken on a single road. Rather, the roads form a network which provides a kind of mesh that enable car drivers to get anywhere they want to go. The same should be true of the public transport network. But in Sydney, public transport tends to be planned and provided as a series of distinct routes feeding into the CBD rather than as an integrated network which enables trips across the entire city.

To illustrate the '400' and '15' parts of the formula, a group of us produced a series of maps of public transport accessibility and frequency in Sydney. First, we mapped every train station and bus routes across the Metropolitan Area, to come up with an approximate representation of which parts of the city had access to public transport within 400 metres. Then we calculated the average frequency of each bus and train route between 5am and midnight on an average weekday. (Lots of member organisations in the Alliance represent shift workers, so looking at frequencies across the whole day was vital to adequately address their transport needs.) The result was the maps below, which show the extent of public transport coverage at different average frequencies.
In this map, those who live within the yellow shaded areas have some form of public transport within 400m

In this map, those who live within the yellow shaded areas have some form of public transport within 400m that comes at least every 30 minutes during the day

In this map, those who live within the yellow shaded areas have some form of public transport within 400m that comes at least every 15 minutes during the day

As you can see, while Sydney is relatively well-covered by public transport routes, not many of those routes have adequate frequencies to provide decent public transport for people across the city.

There are no doubt those who would suggest that lower density residential suburbs in the north, south and west of the city can't sustain frequent, comprehensive public transport and that densification is the answer. But like Paul Mees, we're not convinced of this. As the Public Transport Users Association in Melbourne puts it
Any city with sufficient population density to cause traffic congestion has sufficient population to support a first-rate public transport alternative.

Of course, the point of the Alliance is not just to raise a series of issues and toss around ideas for solutions, it is to build power and take effective political action. To that end, at the conclusion of the Assembly participants were provided with maps of their region of Sydney, and reps from the Sydney Alliance regional groups were on hand outside the theatre to sign people up who were ready to start taking action to address the gaps in the network.

Reflecting back on the Assembly and the work of the Transport Research Action Team so far, I think we've got at least two big challenges as we continue to develop the campaign... (beyond public transport being, like, a big issue!!)

First, I've found it really interesting that the issues of safety and cleanliness have emerged as very significant issues both to members of the Alliance and to members of the public who we surveyed. When we first presented the idea of a formula for transport at the Founding Assembly in September, it was '400:15:1'. But when we went back to member organisations across the Alliance with this formula, many told us that it didn't effectively capture all of their major concerns. This shouldn't have surprised me, given it has also emerged from previous research on transport accessibility. But I do think it's going to be a real challenge for the Alliance to find ways to address issues like safety and cleanliness that don't resort to the typical punitive 'law-and-order' responses. This is going to be especially important given that one of the other issues on which the Alliance is working is social inclusion. Indeed, the harassment of young people by police and security guards in public spaces and on public transport has emerged as one of the key issues for the Social Inclusion Research Action Team. I'm really excited about the potential for the Alliance to come up with something progressive rather than exclusionary on this issue...

Second, finding ways to build power and take action on public transport presents a particular set of geographical challenges for the Alliance in Sydney. To build an effective campaign on public transport, it's essential that people can connect with the issue and take action 'locally', in their neighbourhoods and workplaces. And yet, we've got to ensure that local actions add up to more than the sum of their parts. After all, a big part of the problem with public transport in Sydney is the fact that local services are not effectively co-ordinated into a network that provides integrated and comprehensive access to the entire metropolitan area. Inevitably, confronting network-wide issues like route structures, ticketing and safety will push us all beyond our localities. For instance, a 'local' safety campaign to get train stations staffed at night would have budgetary implications, and change to local bus routes and timetables would mean changes to procurement contracts between the State Government and public and private bus operators. So, we absolutely need to go local and build genuine citizen participation in transport planning -- that's a crucial ingredient that's been missing from all the expert-driven plans for improving the system which are now sitting on a shelf somewhere gathering dust, never to be implemented. And yet, we also have to make sure that local action is itself co-ordinated across the city -- you could say our organising challenge is to build a campaign that is as integrated and comprehensive as the transport network we want to create...

I'll be presenting a paper reflecting on the Sydney Alliance transport research-action process at the Annual Meeting of the Association of American Geographers later this month in New York, at a session on 'Placing Justice and Struggle in Transport Studies'. It's going to be crazy cold I'm sure ... but I'm not complaining!! I just hope that the transport heads go easy on me...

Meanwhile, if you're in Sydney and you'd like to know more or get involved, further details about the on-going activities of the Transport Research Action Team can be found here.

[Note: the ideas above are the product of collective research-action by lots of people involved in the Transport Research Action Team, but it's my write-up, so blame me for any mistakes or misinterpretations!]