To make better streets, we must first measure them

Excerpts from the Foreground panel discussion: To make better streets, we must first measure them. 

I was interviewed along with Nicole Kalms and Libby Gallagher by Andrew Mackenzie. I have only included my quotes here.


The English mathematician and biostatistician Karl Pearson once said, “That which is measured improves. That which is measured and reported improves exponentially.” The question is, how do you measure a street?

This article is part of Foreground’s The Street special series

The following conversation is an edited transcript from Foreground’s ‘The Street: design for people’ held at the Powerhouse Museum, Sydney.

Andrew Mackenzie: David, in your recent Foreground article you argue that we need to refit footpaths to meet current and future needs. This includes the co-location of pedestrians and robot delivery services. When it comes to the street there is a corresponding debate about when and if autonomous vehicles will ever be a thing. What are your thoughts – will we all be using AVs next year, in 50 years, or not at all?

David Levinson: Between next year and 50 years! My official timeline, and you can bank this, is that AVs will be available in the market in the United States and some other countries by 2020 or 2021. If you’re wealthy and you want to buy one, you will be able to. By 2030 autonomous mode will be standard on new cars. There might still be a steering wheel, but you will be able to push it into autonomous mode. And by 2040 human-driven vehicles would be prohibited.

AM: Presumably regulation for those AVs will need to keep up!

DL: Well, timing and sequencing will be critical, but frankly, until we get close to mostly autonomous vehicles, it doesn’t make a lot of sense changing how we use road space, because we still have to allow human drivers in between them. It then becomes a question of where it be fully autonomous first. Full autonomy will hit freeways, then campuses and parking lots, before it hits complicated environments such as the city streets. But remember, there are autonomous vehicles on the road right now in certain test sites around the US. It’s really a question of when, not if.

Once that happens we’ll see a lot of efficiencies. AVs can follow more closely behind each other, and function in a narrower lane. Most car lanes are currently 3.2 metres wide –  much wider than cars, because humans are not very good drivers. Cars driven by robots will take up less space, allowing us to claw back a lot of street space for other uses. Parking spaces will also be smaller because AVs don’t need space for the door to open, so you’ll claw back space there too.

AM: The the reason I wanted to start this panel discussion with AVs, is because in different ways the question of regulation and standardisation was one underlying feature of each presentation today. Arguably automation, whether on the road or the pavement, carries some risks, not least because it represents a net increase in  regulation. Yet informality, or the deregulation of the street, also has supporters.


AM. Leading on from the question of how streets are controlled or regulated, we have seen in recent years a marked increase in community engagement, when it comes to planning decisions that will impact our streets and cities. Libby’s work is a good example. But this becomes much more challenging when dealing with billions of dollars, or major street infrastructure. David, are there opportunities to improve how people might participate in city-making at the large scale.

DL Definitely. My impression here in NSW, is that there is an amazing lack of transparency in decision-making, when it comes to major infrastructure. The government spends billions of dollars on a project, whether the project is good or not, and they don’t release the business case until after they’ve already made the decision. It’s considered cabinet-in-confidence, even though someone’s clearly going to either leak it, or the Freedom of Information Act will cause it to be revealed. Why wasn’t it revealed along the way?

That said, the advantage of this system is that Australia is actually building things now in a way that the United States can no longer do, because Americans have grid-locked themselves with public participation processes and legal challenges to infrastructure. I’ll give you an example from my first job out of college. I worked for the Montgomery County Planning Department, which was in a very high income suburb in Washington D.C. where a lot of government bureaucrats lived, and understood how government worked. There was a proposal to put a light rail line there, between two suburbs, on an existing freight line and right of way. They stopped running the freight in 1984 and in 1989 they said, “maybe you should make this into a light rail”. In the interim, they turned it into a trail, while the proposal was studied. They finally started construction in 2017, because it was tied up in the courts for 28 years, while the legal challenges ran their course.

I believe that there is a happy medium somewhere between no public participation in large billion dollar projects, and the grid-locking of project development. We should probably try to find out what that happy medium is. We are too far to one extreme of opacity here in New South Wales, and too far to the other extreme in the United States. The response to the freeway revolts in the 1960s, and the environmental impact regulations of the 1970s has greatly slowed down America’s ability to accomplish infrastructure there now.