The Ambitious Venture of Sidewalk Labs:
It was one of the biggest urban innovation proposals in Canada’s history: a cutting-edge waterfront demonstration project that was promoted as a playground of heated sidewalks, self-cleaning trash cans and sustainable structures, all contained in a sensor-laden terrain.
The site of this digital wonderland was Quayside, a 12-acre parcel on the east side of Toronto’s waterfront. It’s a name that has almost become lost in the shadow of the data giant that proposed this urban Eden: Sidewalk Labs.
Three years after the first request for proposals that set all those innovative wheels in motion, the head of Waterfront Toronto, the public steward of Quayside, has been replaced; some prominent Waterfront board members have quit; hundreds of articles have appeared; thousands of hours have been spent in consultation with more than 20,000 Torontonians; a revised proposal was walked back; and the two words — Sidewalk Labs — can still stir excitement or agitation from both the creative and the critical.
For data tech entrepreneurs, the Quayside experience is both a cautionary tale and a teaching moment, particularly with regard to how data from the public realm is collected, retained and utilized.
The Quayside enterprise began in March 2017, when Sidewalk, a division of Google parent company Alphabet, responded to the RFP from Waterfront Toronto.
Waterfront Toronto had a bold intention for a data-driven smart city project: “a globally significant demonstration project that advances a new market model for climate-positive urban developments,” that would both address “local, national and global carbon emission reduction targets” and “incorporate technology advancements that enhance efficiencies (and) improve the overall quality of life for citizens, employees, students and visitors to the area.”
Seven months after the RFP was published, Waterfront Toronto announced that New York-based Sidewalk Labs would lead the project, beginning with a one-year planning phase.
Although the proposal was wide-ranging — everything from innovative construction materials to the sensor-rich environment — public attention quickly turned to the data aspects: the harvesting, storage, ownership and use of public data. Tension developed early, with critics wondering how government data transparency would mesh with private-sector data monetization; whether the data collection by Sidewalk would create a data monopoly; whether public data privacy would be assured; and whether the dictates around planning and zoning regulations would be compatible with private enterprise.
The criticism grew in 2019 when Sidewalk released a Master Innovation and Development Plan (MIDP) that many thought exceeded the original scope of the RFP. The MIDP positioned Sidewalk, not Waterfront Toronto, as the lead developer and lead data steward; sought an additional 190 acres to make the project viable; argued for an LRT through the area; offered government a 10-per-cent share of revenues generated from tech derived from the test project; and outlined the basics of a third-party-operated urban data trust to manage data gleaned from the project, among other things.
Going Forward Step by Step – Lessons Learned:
Within months, Waterfront Toronto had rejected much of MIDP, asking Sidewalk to scale back to the original project size, sidelining the LRT proposition, removing Sidewalk’s proposed role as public steward of the land and reaffirming that Waterfront Toronto “will lead all digital governance and privacy matters related to the project, including interactions with its government stakeholders.” Effectively, Waterfront Toronto re-asserted control of the Quayside project.
By the fall of 2019, Sidewalk accepted the reduced mandate. Three years after the original RFP, Quayside is essentially fallow ground, awaiting the next steps.
Those include another vote to be held in May by Waterfront Toronto on whether to proceed, with a Dec. 31 deadline set for a decision on an implementation agreement to greenlight the project. Local, provincial and federal approvals would also be required.
What the two parties in the project discovered is that working with data generated in the public realm is unlike any data project. With publicly generated data, every citizen can be a stakeholder, with myriad and varied concerns that they each want addressed.
What can data-tech entrepreneurs and municipal innovators learn from this experience in this space?
Chris Willsher, Director of Data Platforms and Data Growth Coach for Communitech, believes that the Data Governance by Design (DGbD) framework is a key element for companies entering this space.
Willsher says that data governance is often viewed through the lens of the technical aspects of data management: architecture, systems, and data quality, for example. Willsher suggests that the addition of other parameters — ethical data use, regulatory oversight and data residency (from point of collection to use) — need to be factored in. This proactive approach — “by design” — connects the data users to the data sources in an ethical way that builds confidence in how the data will be used.
It’s that proactive nature of embedding that “by-design” thinking up front, “that is significantly important for organizations to consider,” says Willsher.
DGbD, he believes, should be the new approach for companies entering the public data space. The entrepreneurial ethos of moving fast and breaking things has to be moderated to ensure an ethical approach: “…. how do we carefully consider all the aspects that are tied to this project, and most significantly, how is the data going to be used from this?”
Willsher says that the intentions of the Quayside project were good: encouraging innovation, bringing forward new ideas and solutions.
The challenge for Sidewalk, Willsher says, was “they didn’t make it publicly known that they had really thought through the data piece of it.”
He says that there wasn’t enough done in terms of communicating their intentions clearly with stakeholder groups: “I think that by using a broader view of data governance, by thinking here’s the privacy regulation piece, but here’s the ethical piece as well, you would have allowed them to build much greater trust within the community and with the stakeholders.”
Kristina Verner, Vice-President, Innovation, Sustainability and Prosperity for Waterfront Toronto, is not quite as critical of Sidewalk, pointing out that the project was evolving amidst a moment when data governance was under a critical lens.
The Sidewalk Labs proposal was unveiled only months before the Cambridge Analytica scandal and the grilling of Facebook CEO Mark Zuckerberg before U.S. legislators. Interest in personal data collection and its use was top-of-mind in the daily news cycle.
While Verner says that Sidewalk was always “a bit hyper-transparent about what the intent was and what they would actually be trying to accomplish,” she agrees that the initial Sidewalk proposal was missing the depth of detail about what data was going to be collected in the sensor-rich Quayside and how that data would be stored, managed and used. As well, many critics were concerned that a private company was framing the discourse about public data. This led to the intense media and public scrutiny.
Verner says that this scrutiny means that companies wanting to explore the opportunities for public data monetization “now know the questions people are asking,” and civic leaders know they have to connect with the public to understand “when you talk about enhancing municipal service delivery, what are the tradeoffs that people are actually willing to entertain….”
Verner said, “We’ve helped to frame the ‘no-go zones’ very well.”
There were other lessons for Waterfront Toronto, and for any other public agency looking at public-private partnerships involving public data:
- Staff up. Expect increased media and citizen inquiries about data use proposals.
- Align the goals. Verner says the goals of Sidewalk did not fully mesh with the goals of Waterfront, which, she felt, degraded public confidence in the proposal.
- The public partner needs to lead.
- Citizen stakeholders want to see their interests at the forefront of such public data projects.
- The regulatory landscape needs to be clarified. “Policy modernization around privacy, data governance and intellectual property has long been needed, but right now, it is happening with the spotlight on it in large part because of this project,” Verner says.
Jacqueline Lu, Director of Digital Integration for Sidewalk, has a somewhat different view. She says that Data Governance by Design is actually part of the DNA of Sidewalk Labs: “Responsible data use guidelines and the development of the assessment process has really been part of our internal thinking and muscle from the very beginning.”
Lu says that digital solutions were just part of the Sidewalk response to the original RFP. Sidewalk purposefully considered data tech as a result of “a pretty long and extensive process . . . That’s really anchored in our responsible data use guidelines and our assessment process, so it’s basically a living process that we apply from the very start of designing and conceptualizing any sort of digital innovation.”
Lu says that key to Sidewalk’s relationship with Waterfront and the public, has been listening.
“I think we have always approached the project from kind of a listening stance.” It began with the first roundtable and continued through other public consultations: “… We really just tried to take that listening stance and bring what we heard into the proposal for Waterfront Toronto.”
That listening is reflected, she said, in the realignment of the MIDP and its 482-page Digital Innovation Appendix, and in the ongoing consultations that Sidewalk is undertaking. Asked about how Data Governance by Design could have averted some of the early friction in the project, Lu said, “A lot of the larger discourse around technology and society more broadly is a recognition that things have to change . . . what we are hearing is that people want to understand how these ideas, how these solutions, are really going to improve quality of life and really improve cities. I think that companies need to be focused on how they are actually demonstrating that that is true.”
Willsher doesn’t think that DGbD would have eliminated all of the friction that Sidewalk Labs and Waterfront Toronto encountered because there are “always going to be people who are going to resist that type of data use,” but proactive thinking — anticipating some of the public response to the proposal — would have helped.
Waterfront Toronto could have framed the original RFP more clearly in terms of data governance by asking potential bidders “how do we think through these (the regulatory and ethical issues), and how do we ensure these are built into whatever we proposed, to ensure that our key stakeholders – the residents of the city – are on board in terms of where we are going.”
Sidewalk could have acknowledged the challenges of the project and invited people to be “innovators around the governance of this project.” Sidewalk could have sought public buy-in by asking for help in thinking through the issues, or could have presented its own plan for proper data use and encourage feedback.
To not have shared that plan for data use left a vacuum into which criticism poured.
“I think their initial approach did a bit of disservice to the project . . . it will take a lot longer in terms of gaining people’s trust again, because I think there is a view that they have come out of the gate as not being a trustworthy steward of this land.”
The proactive thinking of Data Governance by Design might have gone a long way to reducing the pain points in what remains an exciting project for the urban environment.
The Smart City Ecosystem Evolves:
New players have appeared on the data governance landscape in the past several years.
Many have private- and public-sector support as they grapple with the new world of data collection and use, with both transparency and privacy. For many, Data Governance by Design is part of their mandate, even if not explicitly stated as such.
Among them are:
- The Open City Network: Executive director, Andy Best says that the Kitchener-based nonprofit has the intention of protecting democratic institutions and supporting practices that can be scaled out to any community. Best says the Network’s board – includes representation from such companies as Miovision and MappedIn; Ontario-based nonprofit Code for Canada; and municipal agencies such as Ottawa City Council; and Invest Stratford – balances public and private interests. Best believes that “It is possible to create public good, democratic good, social good and economic good here, if we do this carefully and thoughtfully.” Key to that thoughtful thinking will be “strong public policies, strong governance, strong oversight.” Open City Network: theopencity.org
- Open North: The national nonprofit based in Montreal assists the public and private sector to create values-driven, ethical and collaborative processes to data governance, management and use. Their advisory work is centred on their definition of an “Open Smart City” where “residents, civil society, academics, and the private sector collaborate with public officials to mobilize data and technologies when warranted in an ethical, accountable and transparent way to govern the city as a fair, viable and livable commons and balance economic development, social progress and environmental responsibility.” Open North believes that in an Open Smart City, “data management is the norm and custody and control over data generated by smart technologies is held and exercised in the public interest. Data governance includes sovereignty, residency, open by default, security, individual and social privacy, and grants people authority over their personal data.” Open North: opennorth.ca or view the Open North Smart Cities Guide online on their publications page.
- Standards Council of Canada, Data Governance Standardization Collaborative (the “Collaborative”): Created in 2019, the federal initiative is led by the Standards Council of Canada. Its intention is to facilitate the building of trust in institutions and agencies managing public data, through the creation and acceptance of standardization strategies for data governance. The Collaborative will look to prioritizing data governance needs; cataloguing stakeholder needs; and making recommendations for national and international standardization. One of the key deliverables is a comprehensive and consensus-based roadmap describing the current and desired Canadian standardization landscape, including recommendations to address gaps and new areas where standards and conformity assessment are needed. Canadians will be invited to comment on this roadmap in the next few months. Data Governance Standardization Collaborative: scc.ca/en/flagships/data-governance
