Nalaka Gunewardene has an interesting article on smart cities published on Scidev.net, in which he makes the case our cities today are just too unhealthy and that the advent of smart cities will “enhance feedback loops within the complex systems” and if processed properly, result in a “the steady flow of data” that “can vastly improve the design of ‘hard’ physical environment and the provision of ‘soft’ services to citizens”.
Though it is true that evidence based decision making around policymaking is better than partisan, expedient or downright myopic decisions around urban development, the case of Sri Lanka alone suggests it is rarely the case that principled and well-reasoned foundations of governance trump more parochial gains. A cogent example – the ‘beautification’ project of the Urban Development Authority, which falls under the Ministry of Defence, is a case study in how not to go about urban rejuvenation.
Nalaka’s central thesis is essentially an article of faith – that the growing use of big data in urban policymaking, and policymaking around the development of urban spaces – will take into account concerns around privacy and the protection of the most vulnerable groups in cities. There is some evidence to suggest this could well be the case, as the research around mobile network produced big data (essentially, the call, text and mobile data usage patterns of consumers) in the service of urban planning suggests.
It is however dangerous to assume this is a given, or will be the path big data use in socio-political systems featuring chronic corruption, political instability and poor legislative oversight will sustain. Add to this mix the profit orientation of big corporate bodies, and consumer protection is often secondary, at best, to market imperatives or contra-constitutional state led directives. All telecoms companies in Sri Lanka have a sordid history of pandering to the extra judicial whims of the Ministry of Defence. The country has no comprehensive legislation around privacy protection. The UDA functioned as a supine extension of the MoD’s illiberal vision, wherein the very notion of rights based development was forgotten.
It is in this context that big data can well exacerbate authoritarianism. In my keynote address at MIT in April 2014 during the inaugural Build Peace conference, I looked at how the Internet of Things (IoT), a founding block of any ‘smart city’, could impact individual rights.
I asked the question, when does intelligence turn into surveillance? I asked how one could even remotely maintain control over privacy within ecosystem of competing owners, location sensors, proxy indicators, sentient nodes, ambient observation, pervasive automation.
The fact is, the normative assumptions around the use and utility of big data are largely anchored to a democratic framework that animates the production, usage, storage and analysis of data. Even randomised and anonymised, big data can provide insights into geo-fenced communities and specific income groups that are then hostage to the nature of government and timbre of governance prevalent at the time. While under the current political dispensation, there is arguably more interest in revisiting and reforming some of the excesses of the past – including around urban development – big data’s growth, especially when the vital barriers between telcos and state surveillance are rendered porous and invisible, can lead to policies and practices that put an even greater stress on democratic rights and civil liberties. The danger here is in the formulation of policies detrimental to communities based on big data no one individual or community has control over. Just as much the potential of big data, as Nalaka’s article makes clear, “can provide comprehensive insights for urban planning and other public purposes that do not marginalise the poor”, it is also the case that increasingly, this same data under different circumstances can in fact place them at greater risk.
Necessary protections needs to be envisioned and this requires new thinking. How can complex systems evolve to protect each constituent node? Is this technically feasible even if desirable? How can the best use cases of big data (around smarter cities) be encouraged whilst retaining ultimate control of data in the hands of informed consumers? How can telcos, if they increasingly see their role as facilitating progressive policy making, protect the privacy of customers when the data in the aggregate can also put them in harms way, no matter to what degree individual records are scrambled? How can even the most progressive cross-domain analytics be conducted in a transparent, consultative framework that informs all citizens – esp. those who are at the margins of society – around the use of what ultimately they are the producers and benefactors of? Is it enough to have complex privacy regulations, or does the advent of big data also require – for data voluntarily produced – sustained consumer education, around what uses for this data lie beyond the immediate aim of communication?
As Nalaka avers, big data can help citizens make fast and informed choices. While true, there is rich potential of big data, in the service of a few, to undermine this choice or worse, to give a compelling illusion of choice, when in fact there is little or none. Nalaka’s article deals with a familiar narrative of big data in the service of a more effective and efficient city. My concern is around, inter alia, the IoT, sensor data and the increasing sophistication of analytics around big data leading to even more opaque choices around how we see and govern ourselves, especially when dealing with entities for whom a rights based approach remains peripheral or unnecessary.