Georeactor Blog

I've been bad about keeping up reading or finishing books, so this could hopefully reboot things. As a (former?) civic tech person, I've wanted to read Sideways since it came out a few years ago. It's a behind-the-scenes look into Google's failed smart city project in Toronto.
I'll try to connect this story to Waymo, Hudson Yards, Parks & Rec, and Amazon.

Sideways: The City Google Couldn't Buy (Josh O'Kane, 2022)

Google's Sidewalk Labs was slated to build a smart city on a few blocks of repurposed land on the Toronto waterfront. O'Kane was reporting on Sidewalk throughout the process and was a thorn in their side. This book was part pandemic passion project, part retrospective after Sidewalk pulled out in 2020.
I'd followed news stories and figured that Toronto residents got riled up by smart city surveillance, or that Google didn't study urbanism… but instead this was fundamentally an office politics / mismanagement story.

Daniel Doctoroff was picked to lead Sidewalk after his management of large projects for Mike Bloomberg (both as mayor and at his company). Selling Sidewalk would be comparable to his role developing NYC's bid for the Olympics. He would take on the startup CEO mentality of 'betting the company' on a pitch for Larry Page to make a mega investment. When that failed, Waterfront Toronto happened to contact Doctoroff at the right time to become the new focus.
Although Waterfront's RFP revolved around a moderate land parcel (known as Quayside), executives on the Sidewalk side saw this as only a formality, where a successful pilot program assured a role in a future development (Port Lands, ~40x larger). This guaranteed conflicts throughout their partnership. Another recurring theme is culture shock, where Doctoroff's approach is seen as too bombastic and quick to yell (which I don't think Canadian culture is immune from?) but it resurfaced stereotypes about Americans and New Yorkers.

Sidewalk struggled to find the correct scale for their plans. Page created multiple passion projects within the new Alphabet structure which would have to compete for his attention. The Sidewalk team prepared a book with experts on urbanism, open data, and modern privacy (de-identification, and likely differential privacy though it's not mentioned). This was interesting because most criticisms frame Silicon Valley as ignorant to precedent (like DOGE) and this team would name projects after Jane Jacobs.
Page and Google X (wow, I forgot there was a Google X) expected radical ideas, so every aspect of the plan had to be hyped up, including serious research into a weather dome. This separated Sidewalk's plans from their urbanism research.
Even after becoming more grounded to appeal to Toronto, investing in engineers, sensor networks, and new transit modalities would not be profitable at Quayside scale. Sidewalk could not last unless they secured more land in Toronto, or resold the technology in NYC and Singapore. In the final months, Sidewalk's financial backing fell into question when Brin and Page left Google, and Toronto would not grant Sidewalk any additional land even for a Google office. Then Covid forced all sides to reassess the value of real estate and financing the project. Doctoroff insisted that it was the pandemic that killed the Quayside project.

Sidewalk's rivals were more receptive to the book project, so O'Kane has good insight into them. Jim Balsillie, the former RIM / Blackberry CEO, had already built up a government network by founding the Centre for International Governance Innovation and a School of International Affairs. He would argue that Toronto should support the local Canadian tech industry or a startup campus. Maybe I was too poisoned by the Blackberry movie, but is it healthy and realistic to keep FAANG companies from building a hub office in Toronto / Canada?
The Mississaugas of the Credit First Nation were one of many organizations which had meetings with Sidewalk during development, but ultimately found none of their insights and requests made it into master plans, except for the land acknowledgement.
Current Ontario premier Tom Ford supported ideas to build fun attractions like a ferris wheel, which O'Kane dismisses offhand. I know that the Ford brothers have been a pain to Canadian leftists and urbanists, but it's a reminder that this land should be available to the average resident and not someone moving into a brand-new deluxe apartment or a Google office.

Privacy experts have an interesting dialogue with Sidewalk: getting hired and designing an "Urban Data Trust" governance model, shifting to consulting roles, and ultimately becoming public critics (from Nov 2020, post-Toronto project). Like the Mississaugas, the concern is that Sidewalk reached out to pay off and co-opt privacy advocacy. I don't get the sense that the experts were particularly stunned and activated against Sidewalk's plans, more that they saw the project become vague, fragile, and unpopular, so they left it to fade and fail.
I do believe that de-identified 'urban data' was a real objective for Sidewalk, since civic tech volunteers would also hype up open data at the time. Maybe they thought, there are a few different layers to the data trust, so no other public or private entity would scope this out and test it in the way that Sidewalk was able to. Even if they succeeded, I don't know how revolutionary this would be for daily life, because data in aggregate cannot make recommendations for individual people.

After being reined in, the Quayside plan was mostly concept art, heated sidewalks, and wooden construction. That construction and variable traffic signals would require significant changes to regulations. Ultimately there was not a lot being offered for groundbreaking or all-powerful computers. Doctoroff and the planners even avoided saying 'smart city' - so what was left for Googlers to do?

Toronto today

Before Sidewalk officially left Toronto, several people contributed their retrospectives to the website some-thoughts.org.
Toronto has been processing their experience with this book, as well as a stage adaptation with performances in 2023 and winter 2024–25.

Recent developments in waterfront real estate led that 2024 reviewer to ponder:

To the characters in the play the rules and regulations around land sales, rezoning, the independence of the three levels of government and so on are immutable. But since last year we have learned that's far from true as Doug Ford has driven a coach and horses through them in the Green Belt, at Ontario Place and the Science Centre and has shown repeatedly that he can fist Toronto whenever he feels like it. Ironically Google's vision for the Quayside project was probably achievable but Doctoroff just didn't work out that the answer was to get Ford on board!

This spring Ford is in the news for allowing an inexperienced 'mega spa' in the Ontario Place / Villiers Island segment of the waterfront.

In December 2022, Waterfront Toronto announced an affordable housing proposal for the Quayside site. In the past week, they made their first construction notice.

In February an Archinet article touted a $680 million 'city within a city' plan for Toronto's Port Lands.

Reddit commenters remain fearful that a key foundation of any waterfront development, a new tram line, may not survive to launch in 2032–2035. The YouTuber RM Transit did a video about waterfront transit, including a couple of minutes showing the current construction and infrastructure.

I didn't mention it here, but O'Kane tracks a grassroots project which got Google to change its plans for a new office in Berlin. That building continues to be a community meeting space, and in 2024 Google renewed their contract for another 2+1/2 years: https://www.bum.berlin/ueberuns/geschichte

Could Sidewalk and Waymo have traded fortunes?

Another Google spinoff, Waymo, has become super-visible in the past ~2 years after it expanded beyond Phoenix and SF. Even though this is seen as a success, and a dunk on Tesla, Uber, and GM's projects, it comes with an incredible cost, significant human monitoring, importing a specific Chinese EV because there's no f-ing way they can use a car off the lot… which will take decades to break even.
The Waymo entity has gotten $11 billion in funding, and if you go back to the Google cars driving around in 2012, I bet we're at the lifetime cost of the Hubble Space Telescope and repair missions ($16 billion).

In another universe, you might go to a Sidewalk City's library and pick up a book titled Crash, Test, Dummy, telling how maniac boss Levandowski killed self-driving cars at both Google and Uber, leaving it for Elon Musk to fumble. In this world, what would neo-Detroit or neo-Alameda look like?

Some Google X companies have been allowed to grow slowly (Wing, the drone delivery system, is looking more real now), others died off later than you think (Glass, Loon, modular phones).
A conservative Sidewalk could maybe have made an incubator alongside a hackerspace / hacker-house, with different residential blocks for more party-centric and more crunchy/organic family living. The WeWork / WeLive guy took that trajectory, lost everything, but now has Flow.life buildings open in Saudi Arabia and Florida. Burning Man has ideas about building a more permanent / autonomous space. There are multiple co-living companies like Outsite.co, though these might be in decline among non-startup, non-party people. Ever since I left the nomad game I stopped tracking these so much. If a Sidewalk Hacker Block Party became popular and spilled out into the neighborhood with street paint, weird stackable robot cars, enviable influencers, or grilled cheese food trucks, maybe that's something?

Let's turn the dial in the other direction. The Quayside project launch gave Page a photo op with Justin Trudeau. He wants to hoverboard around Sidewalk City gesturing like Walt Disney - that's not going to happen for another share house.
In 2016 Doctoroff presents the Olympic bid / urbanist research / dome book, but Page demands something more radical. He was seasteading-pilled by 2013, and apparently said something about wheeled buildings in this meeting (before the Mortal Engines movie, but possible that he read the book?). Bikesharing was everywhere in the US, sharing economy and apps are still hot, and "Attention Is All You Need" is a year away.
In our imagined universe, Doctoroff calls an audible and presents a new vision: Earth (a mountain of rolling turf roof full of hobbit hole windows, and rock caverns at the base), Air (monorail pods), Fire (a flower-shaped solar energy concentrator / amphitheater), and Water (a marina to test modular houseboats).
Is this a 'smart' city? Sure, maybe there is 3D-printed construction, roads with sensors, a subscription-based restaurant-grade kitchen, an automat revival, a social credit score, wheeled robots that follow carrying your stuff.

Does this vision survive reality?
I think about MIT's Stata Center, and NYC's 'pencil towers' and new WTC transit station, which have tried to build something non-standard and end up paying more for something which is harder to maintain. Construction is a precarious place to make everything about disrupting norms. These guys think they can make anything because of Disney World, but those buidings are carefully faked and when they made their town Celebration they mostly played it safe.
Does the vision survive Covid? Definitely the wrong time for a big investment in real estate or flashy co-living spaces. O'Kane points to Doctoroff's April 2020 op-ed in the New York Times, which repeats several familiar ideas about sustainability and wood buildings, but also suggests a pivot to medical research campuses, covid testing infrastructure, and staggered work hours.

Doctoroff's history with bringing Hudson Yards to NYC is another warning sign that development alone does not happily plug into an existing city and culture. There's a risk that Quayside or another Sidewalk city on Alameda's former navy base, would be too isolated. I recently watched videos 1 and 2 on the Baltimore Peninsula (aka Port Covington), a development on their waterfront which remains under-populated and disconnected from the city. The developers there likely intended that buffer between them and downtown, and Sidewalk had a much different agenda for transit. But with Toronto's transit arriving in the 2030s, there's bound to be a time where Google is across the highway from most of Toronto's pedestrian activity.

When I think about city innovation which has been successful post-Sidewalk I think about delivery, specifically Amazon's shift to Rivian vans and electric cargo bikes, and ultrafast grocery delivery (Gorillas / Getir) though that might be on the way out. AirTags breaking out of the IoT doldrums. There was a whole era of covid tracking functionality built into iOS and Android which I still don't know much about. Also surveillance in some cities is much more advanced, with most transit crime stories now quickly including video of where people entered and exited the system.

Parks and Recreation and techlash

By 2013–14, there was enough cultural awareness of Google as a gonzo college campus / grownup playground that we got The Internship and HBO's Silicon Valley.
More relevant to smart cities, in April 2014 the TV series Parks and Recreation depicted a town kissing up to tech company "Gryzzl" to get free WiFi for their town. Their influence and data collection became a major conflict in the 2015 season. So it maybe gives us a time capsule of thoughts around Google Fiber, drone delivery, and "you are the product" era of consumer consciousness. And it weirdly predicts the whole Sidewalk debacle!

Techlash was inevitable for any big Google move at the time:

• The book mentions Amazon's HQ2 bid process, which earned its own techlash and made Sidewalk glad to have gone a different route.
• The book mentions Surveillance Capitalism came out around the time and came to be a good term for Sidewalk taking a role in collecting data, data governance, or co-opting local government to make their proposals legal under Canadian law
• In 2018, San Francisco passed a referendum for tech companies to fund homeless services.

Techlash makes sense to the average person and their frustrations with technology. Once something is ingrained in public consciousness, it's hard to introduce counter-programming, like "Biden increased federal funding to police", "Facebook doesn't train AI on your private messages", or "23andMe went bankrupt because your DNA data isn't that valuable and has been strictly regulated since 1996".

In 2013, Amazon revealed a design for their Seattle campus with huge biospheres. Around May 2017 Amazon assured Mary's Place, a family shelter using a hotel on the property, that they would have space there "until homelessness is solved." The new shelter opened in 2020 and after this 2021 press release Amazon has been more quiet about fuinding it.
Not knowing anything concrete about this project… this is an example of something which could get a lot of techlash. But… is the takeaway Amazon good and other companies should be ashamed, or omfg this is hyper-capitalism, or something else? The story was even-handed enough, or beyond the simplistic form, and so you really don't hear so much about Amazon HQ.

I think you could have some smart real estate mega-corp, perhaps tied to a sports team, build up a new city. Without a name like Google, will there be any techlash?
But I hope that they don't lean too much on the charter city libertarian ideology so much? We are already seeing some controversy on this since Trump's vague "Freedom Cities" campaign will obviously be politicized, plus it's bringing back company towns and they're making an unspeakably abhorrent proposal to locate one at Guantanamo. omfg hyper-capitalism indeed.