Data Co-ops for “Smart Cities” Now!
Getting from Paranoia to Collective Power
Words: Trebor Scholz
Data grabbing does not only happen on social media, now also toilet seats (!), toothbrushes, smoke detectors, toasters, and fridges can pass on whatever they can find out about you. And what can we do about all those corporations running off with our data?
The American author and scholar Shoshana Zuboff, in her recent analysis of surveillance capitalism, correctly points out that demanding that Google should start protecting our privacy is “like asking Henry Ford to make each Model T by hand or asking a giraffe to shorten its neck.” It contradicts their innermost business logic.
What are near-future possibilities of refusal when it comes to these data systems? And better yet, what are some workable alternatives? How do we get from paranoia to collective power?
Over the past years, there is much more dynamism in the American antitrust debate. In 2017, a young jurist at Yale Law School, Lina Khan, made a persuasive argument that Amazon should not get a pass on anticompetitive behavior simply because it keeps consumer prices down. More recently, Senator Elizabeth Warren demanded to break up Big Tech companies. But realistically, how much political will is there to implement and enforce such efforts against monopsonies that have billions in their war chests? Will any of this still happen in our lifetimes?
By now you know that those who own the data, collected on our social media profiles or fitness trackers can and do steer our consumptive behavior. Our behavior has become their product. We need answers: Who owns and governs the digital infrastructure in our cities? What types of data are being collected? How are they analyzed and to whom are they sold? What happened to informed consent?
Data is in the eye of the beholder: perspectives, values, and interests of those who collect the data may differ from our prerogatives. Autonomous vehicles may not recognize and therefore kill pedestrians with uncommon disabilities crossing the street. Computer programs used in hospitals may not detect melanoma on black and brown skin. Black families, for example, may not receive mortgages because they live in a postal code that is associated with a low-income neighborhood. But algorithmic inclusion is not a reliable way to a fairer society either. As the technologist, teacher, and activist Nabil Hassein writes: “Liberation of Black folks and all oppressed peoples will never be achieved by inclusion in systems controlled by a capitalist elite which benefits from the perpetuation of racism and related oppressions. The struggle for liberation is not a struggle for diversity and inclusion — it is a struggle for decolonization, reparations, and self-determination. We can realize those aspirations only in a socialist world.” You can’t fully control what you don’t own: We need broad-based ownership and meaningful say in our workplaces, not merely inclusion in the apparati of surveillance.
“Smart cities” are beachheads of this discussion; our future is irrevocably linked with cities. (Yes, cities have always been “smart” but let’s discuss that in another piece.) By 2050, close to 70% of the world population will live in urban areas. From Shenzhen and Toronto to New York, embedded sensors are monitoring the flow of crowds and manage traffic. In Baltimore, Police used facial recognition technology to identify participants in the protests following the murder of Freddie Gray. Predictive policing programs in the UK and the United States are also fueled by facial recognition data. City leaders work hard to be recognized as digital innovators, holding up pretty visions. Time-strapped parents are asked to imagine their children being safely chauffeured to soccer practice (or Sanskrit classes) in autonomous vehicles. Big tech companies may put on a theater of democracy, asking for broad community input without genuinely seeking community participation and guidance. Many of these “smart city projects” are public–private initiatives, which puts local government into the compromising position of being the potential regulator of a project that they are also participating in.
I’m asking again: What are strategies that can move us from paranoia — as justified as that may be — to collective power? How could residents steer the discourse toward data ownership and democratic governance of digital infrastructure?
One powerful step toward collective power in “smart cities” is a data cooperative of residents that monitors the data that are captured within a given city or community. The proposal at its core is straight forward: Allow all residents including caregivers, condo dwellers, sex workers, technologists, undocumented workers, and even commuters, to govern the data that they generate in the city.
But let’s not be starry-eyed; most of the digital infrastructure in “smart cities” is likely to be provided by Big Tech, not predominantly cooperatives.
How would this work? In Toronto, Alphabet’s SideWalkLab proposed a “civic data trust” that would manage and govern data captured in relation to everything from transportation, social care, to education. Who would be given authority about such data trust? So far, “public actors,” government, or academia seem to be favored candidates to run those trusts. I argue that resident cooperatives should be considered as guardians of “urban data.” Why?
You don’t have to be a distinguished scholar to notice that “smart city” initiatives are routinely met with intense distrust by locals. (If in doubt, a brief Internet search of any of these projects will provide ample evidence.) Data cooperatives could help Big Tech to productively work with local communities. After all, co-ops are a familiar democratic entity to be reckoned with. One in three Americans is a member of a co-op. 52 million people in Germany. Over 13 million in Australia. Upwards of 11 million in Brazil.
The cooperative model offers an alternative to public or public-private data governance. Consider the dysfunctional nature of numerous governments, not only in the United States or the Brexiteering UK. As I am writing this, the Feds in the U.S. is mounting assaults on any workers making the federal government quite literally the last place where I would trust with democratic data governance. The so-called Janus Decision, for instance, means that public sector unions can no longer legally collect “agency fees” from non-union members. That is just one example. Surely, this differs from country-to-country and city-to-city but also municipal authorities are not always a guarantee that minority-dominated neighborhoods (or native peoples) will be fairly represented. Minorities may not be equitably represented in City Hall, for example. The co-op structure with its 1-worker, 1-vote principle is a safer bet when it comes to the representation of local interests. A cooperative rather than a government agency can more equitably represent that minority.
Review processes led by data cooperatives could be inspired by academic human subjects review committees in universities that collectively think through and monitor the possible long-term effects of data collection.
How then would this work? The data cooperative could ensure that all raw data are homomorphically encrypted, for example, ensuring that infrastructure-providing Big Tech receives data models that they can still monetize but without the possibility to ever re-identify individual users.
The data cooperative decides which data should be harvested and which data capture would not generate community benefit. In the health data sector, there are various examples of data cooperatives including MIDATA.coop, TheGoodData, Holland Health Data cooperative, and GISC (a grower data co-op for farmers) and others. MIDATA allows members of the cooperative to collect and cooperatively own their health data. Patients who see their doctor or go to a hospital will make those data available for free but when approached by a for-profit research institute, members of the co-op may decide to sell the data with the gains being given to public research.
This, to be sure, can only be the beginning of a debate that needs to be fleshed out.
While the primary infrastructure providers may be Big Tech, policy facilitators should also make room for cooperative digital infrastructure, for platform cooperatives. The model is shovel-ready. Platform co-ops can thrive in sectors such as childcare, food delivery, short-term rental, data entry, home repair, education, and transportation.
Now, you might say: Sure, sounds good but how will data co-ops ever scale? There are historical precedents that show us that it can be done. Rural electric cooperatives in the United States demonstrate that the scaling of cooperative ownership of infrastructure is feasible. In the United States, electrification was executed under monopoly rule, which left many communities quite literally in the dark. This is where cooperatives stepped in and electrified large parts of the country. Today, the National Rural Electric Corporation Association in the United States It has 40 million members in 47 federal states. It represents the interests of 900 electric cooperatives to legislators. In the U.S., electricity cooperatives own 42% of America’s electric distribution lines.
Without any doubts, also the data co-op model comes with opportunities for failure. How would residents who have never met in person feel sufficiently responsible for one another to jointly make meaningful decisions? What if co-op members want to delete or port their data to another data co-op?
Many of these questions are unanswered but collectively we can start to think about how data trusts can be owned and governed in a way that doesn’t hand over the keys to your city to American Big Tech. Without bold, long-term benefits for the community, rooted in immutable data governance structures, “smart city” projects are bound to fail.
