This week I’m excited to share a special audio edition of the Urban Technology at University of Michigan newsletter. Back in 2020 we recorded a conversation with Benjamin de la Peña to discuss his work in mobility and the role data can play in transportation systems. Since this conversation, he has taken on a role at the Shared Use Mobility Center in Chicago, where he’s the CEO leading their excellent work to make transportation more sustainable and humane. So, go ahead and press play to listen to our conversation, or keep scrolling if you prefer reading the transcript. (And if you are a regular reader, please let us know what you think about this podcast experiment!)
💬 Hello! This is the newsletter of the Urban Technology program at University of Michigan, in which we explore the ways that technology can be harnessed to nurture and improve urban life. If you’re new here, try this short video of current students describing urban technology in their own words.
Benjamin “Benjie” de la Peña: Hi, I'm Benjie De La Peña. My full first name is Benjamin (“Ben-hah-min”). People have been saying it as Benjamin (“Ben-jah-min”) and I've been taking that, but to honor my own heritage, it really should be Benjamin because I'm Filipino. So I go by Benjie. I am an urban planner by training, but a design geek and transportation geek in so many ways. I'm particularly enamored with informality in cities. I write a newsletter called Makeshift Mobility where I track innovations in informal transportation.
I tend to think in terms of systems, or at least asking, what are the things that are driving some sort of behavior? And then finding out what could change there.
When we recorded this discussion, Benjie had recently left his position as Director of Innovation for Seattle’s Department of Transportation and had established a consultancy called Agile City Partners. His focus there was helping cities become more agile and helping them learn to pivot.
Benjie: To pivot is simply to change direction, right? Usually a U-turn, or to turn on a dime. Every city, everywhere will be looking at the impact of COVID-19 and the shutdowns and knowing they're not going to have the same revenue levels from taxes because businesses have closed or, even if they open back up, they will have lost six months worth of revenue. They're looking at the next year, saying, “Oh, we don't have as many resources, so what are we going to cut back on? What are we going to continue doing? What can we invest in?” And they need to respond very, very quickly to that. It can scale from the level of a project — Is this a project we still want to do? If we still want to do it, what are the most essential things? — or to the scale of a city budget or a national budget and say, okay, we need to pivot. That's why we think cities need to be more agile and less concerned about negotiating everything upfront, but instead keeping really close to the people they're serving and improvising so that they deliver what people actually need.
Bryan: Got it. So in hearing this description, it has to be uncomfortable, or at least a “new” feeling, for cities to be responsive and agile rather than planning everything up front.
Benjie: We call it engagement, right? The way you do engagement in cities is: we have a project, we have X number of meetings, this many people attended, we gave a presentation. Agile says conversations are more important than the details of the project. Engagement is a ladder, which means you move people into conversations. Having that on the forefront of asking a city asking itself, or a city department asking itself: how many conversations do we have currently, and what is coming out of those conversations? Not how many consultative meetings have we had, but how many conversations are we fostering?
Bryan: During your time at Seattle DOT you led the development of the Transportation Information Infrastructure Plan, which was released in draft form in 2019, and you said it will “forever be in draft”. What did that entail?
Benjie: Information infrastructure about mobility has emerged and is rapidly growing. Departments of transportation and cities have plans or policies for individual modes, and then maybe layer some requirements or information. They don't have a plan for the information itself. I like to say city agencies manage information as tools rather than as assets. What I mean by that is, if a tool is convenient, I'll use it. And if you have one particular tool, you may have five screwdrivers laying around the house wherever you need one. But what happens with information is that every screwdriver, or every data set, is now different from one other, and you don't actually know the cost of losing that information or gaining new information. So we need to start managing it as an asset, and then we need to be intentional about it. In terms of information technology most cities only have upgrade plans: we’re using this version of this database, what will we buy next? Rather than having a plan that says: this is what the information system should do, and these are the values that will guide it.
Benjie and I talked about the technical aspects of working with transportation data inside city hall, and he described the generational differences between the public sector, which usually relies on commercial, often very costly, geographic informations systems, and the types of geospatial analysis software used by startups like the Lime Scooters and Lyft’s of the world, which are more often using open source options or coding their own tools in Python.
Benjie: In the Seattle DOT we probably had two people who knew Python, or R. And so we couldn't communicate. We would ask for data, and we'd have to translate it.
Bryan: Can you give me a concrete example of the types of information that would be useful?
Benjie: So here's one example: we need to know how many bikes or kick scooters pass a particular segment of road. So for instance, with pedestrians, what we do now is count people at intersections. Why do we count them? So we know how long we're going to keep the crossing light on. But we don't count what people are doing in the space. So if we realize that a lot of them are looking for places to sit, then maybe the intervention isn't just “how long do we keep the crossing light on,” but maybe, “we need more seating there!” So, again, driven by asking: what are you trying to do? What is the goal? And what is the function? The easy thing to do is to say, well, there are the technology providers, why don't we just get all of that data, and it's easy. So we're going to require that route data. But if you step back and ask, what is the function we're trying to meet? We're trying to make sure that we have a count of these particular routes, and how many people are there. I don't need to know where they come from or where they go in particular, but in a particular segment, in terms of traffic management, I need to count. So maybe rather than trying to grab data that's available, wherever it's available, we say, “actually, we need counters there, and they should count people, people on bikes, and people on micromobility.” That's the approach coming from a function and knowing what goal you're trying to fulfill. That's the information you want to get, not whatever information is available and worrying about the privacy issues afterwards. It's that whole thing of saying, if this is the function we're trying to meet, what are the possible information sources? And what's the strategy to get there? The other is that if we are going to interact and communicate with the emerging information, what do we need to do? Do we have an API, or a strategy for API's?—
An API is an application programming interface, which is basically a way for computers to talk to each other more easily and exchange important information.
Benjie: —and which ones are we going to put first? Why are we putting those first? And how will they interact with everything else? How does information flow within our databases?
Bryan: And so once you have those numbers, you can translate them into Maps or other tools to communicate the needs to, say, the mayor or the Director of the Department of Transportation?
Benjie: Right. Or whoever it is. The decider could be the council, which needs to make a decision and you make a case for it.
Bryan: Okay, let me see if I have this straight. Cities have tons of data, and they’re getting better at managing and using that data, and maybe even using it to make key decisions. But the speed of legacy cities is not the same as the speed of data. What do cities need to do to take full advantage of information?
Benjie: Seattle has one of the best parking data sets in the world — I think the last time they released it, it was in the order of 3 billion cells, and it's constantly updated. Ideally, that should be in a system where the city can say, all right, we're changing the parking rules on this curb today, and we can actually do that. What we actually do is we have a parking vendor, we give them the data set on a on a quarterly basis, and it takes them about six months to propagate it out into their system. You cannot change the rules overnight; at the very best, imagining you could actually sign off today, it will change six months from now. So the system is not responsive to existing conditions. But it could be if you had designed those systems and said, okay, this needs to be an API that pushes out what today’s regulation are for every street. And then upstream of that, how do you make those decisions? And who needs to know that, and what kind of information needs to be exchanged? All of that, I think, is part of managing information as an asset, rather than a set of tools and data sets.
In addition to Benjie’s work in places like Seattle, he’s a proponent (and dare I say “fanboy”) of informal transit, which are more commonly found in countries that are not as rich as the U.S. But there’s something we can learn from them, and perhaps a lot we can learn from informal transit in other countries.
Benjie: These are the tuk-tuks, the auto-rickshaws, the jeepneys, the matatus. Nearly every city in the Global South has them, and dare I say some cities also in the Global North. These are privately-provided public transportation. Very often they have some loose permitting to them, and it's very adaptive, so cities struggle with them because they have imagined modern transportation to be the fixed rail, the BRT, the subway. As it grows, every city has some imagination of its future, maybe in actual plans, and the renderings will show the disappearance of these things—no more auto-rickshaws, no more tuk-tuks, just shiny malls with autonomous vehicles going around, which is really shocking when you think about how in many cities, the informal sector and informal economy provides and contributes as much as 50 to 70% of the actual economy.
Bryan: Often these informal transportation networks that you’re describing develop in areas that have been excluded in some way from the formal economy.
Benjie: Usually there's a history of exclusion, not providing transportation to communities where people are a different color, or they are the native population and the colonial group doesn't want to provide services. But people need to move, so services emerge, and then the government comes belatedly to try and regulate it, usually from a traffic standpoint, not a provision of services standpoint.
Bryan: I see, so the regulation is to reduce traffic accidents and deaths, which is like trying to pour a little water on a fire, but they’re missing the point that people are using informal transit because they really need to get places for work, school, or what have you. What can formal cities learn from informal transit?
Benjie: I'm really excited about it because they're highly agile and very responsive, although their networks have downsides. They all run the same way. They have similar vehicles. So on a systems level, I'm excited by how we can transform them by keeping the good things about it, and eliminating the precariousness of livelihoods, the dangerous driving, and all of that. So I'm really excited by the the innovation coming out of this space but also worried that, for more orderly cities, we will kill this kind of innovation, and in terms of inclusion, lose out on the people who work and find a living in there, as well as the many, many people who use it. One more example: in the last last decade when New Delhi completed its metro lines, the Chief Minister of New Delhi said, now we can get rid of the auto-rickshaw. But 80% of the city uses auto-rickshaws, and 50% of the city could not afford the cost of a metro ticket. So there are inclusionary implications to informal transportation.
Bryan: One of the things that's been a focus for me in the development of our program at Michigan is bringing questions of “urban technology” or “smart cities” or these other buzzwords back to the fundamentals of, what are people trying to do in cities?
Benjie: I totally agree. I just came from a call with a consultative group in Vancouver, and one of the questions to me was, “if you had a billion dollars, what would you do in transportation?” And I said, “No, no, don't ever tell that to a planner, because then they build something big. That doesn't work.” I thought of Jaime Lerner, right? “If you want to get to innovation, take off one zero; if you want to get to sustainable, take off two.” One of the things I said was, spend $100,000 on just understanding the user and their experience if you want to improve transportation. Start with the people who are actually using it and see how you can improve it for them, rather than just having a kind of 40,000-foot view of “we need to connect this part to that part.” It’s that frame of: what is already there? How do we make it better for the people who are actually using it now?
Benjie and I talked about Manila as an example of a megacity that thrives because of its informal transit, including a network of jeepneys, which are boisterously painted trucks. I asked him whether a city like Manila should even aspire to formalize its transit, or if there’s another path.
Benjie: There are several layers to that. The first is that transportation decisions of the elite are usually not about transportation, they're about traffic. So you build the light rail line on the heaviest artery because you think it'll solve traffic, and it doesn't, it provides a service that doesn't quite connect for the people who are already using it. What it also means is that instead of trying to say, “Okay, we're going to eliminate all of you [informal] guys, and we'll fund this bus service or start new services.” And those will never have the same coverage as the existing informal transportation systems. Transformation would mean looking at the system and asking, “Why do the drivers behave this way? Why is it unsafe?” Well, it's because every person they see on the road is fare. So maybe we need to change the business model. Maybe instead of drivers earning their income only from fares, maybe the government should pay them to run those routes. And then, kind of the lesson from bus rapid transit, we can have a performance service contract that says you need to serve this route, and therefore you will get paid this amount for how many kilometers you run, and if there are infractions, then that's points against you or reason for cancellation. Then it meets their needs of getting a steady income and meets the needs of the system so that you get you get predictable, on-time, better stops, because they're not trying to chase down passengers and not dwelling where they think they'll get passengers. So you're dealing with that. The other thing is figuring out—you know, usually jeepneys and matatus, they are celebrations in terms of paint and art and there's a rebelliousness to them. What formality does is sometimes say, “Oh, well, we'll eliminate that because we have these nicer buses all painted the same color.” And you lose that local expression. Transforming means, how do we use that and enhance that so it celebrates our localities and local artists?
Bryan: How is technology a part of making that possible?
Benjie: Simple. Let's talk about fares, right. Currently it's a cash-based system, so for the drivers, you want cash on hand. But in many places, like in Nairobi, governments move to a cashless fare collection system. Mostly they want it so that they can tax correctly, and also sometimes so that they can start reaching out to the unbanked. What technology can do, if we think about it and frame it correctly, is that it's a way for people to pay without having to transact in cash, particularly during COVID times. But then you have to consider, how does that system get money or get the driver paid on the same day so that they're not waiting a week? So how does technology solve that problem, and what does FinTech do to be able to solve that problem? At the same time, if you're going to run these cashless fare collection systems, they're going to be networked, right? Because you need to know where they are and where the vehicles are—you have a vehicle location system—how do you use that to manage the system itself, and know what's happening in the system? Now, all the cities are actually blind when it comes to forms of transportation until they spend on these expensive vehicle location systems. But what if you could get all that on the right platforms and with the right API's, so that the managers now have a handle on it? Could you take that and move away from the one permit, one route, you can only run that route until you renew your permit system, to having a much more demand-driven system where we say “today, we know that this particular area need is underserved, and so we will actually pay you drivers to run that,” or “today there's a particular emergency there. We need more people coming out at this particular time.” Can we shift the movement of money through technology so that you put out the right incentives?
Bryan: Benjie, who are the different types of people, or let's say the skillsets, that you'd need to have as part of a dream team to take on that transformation you're describing?
Benjie: First, you'll have to engage the passengers themselves, right? So a lot of user experience studies to see what their pain points are. You have to engage the drivers and the operators—what are their pain points? In Nairobi, Google tried cashless fare collection more than a decade ago, and it failed because the drivers and the conductors needed cash, and the card payment system wasn't working for them. So you have to engage them and figure out what they need — When do they need it? What are their pain points? — and then you have to engage system designers to ask, okay, what is the information being used? How will it flow? What information needs to be out open in the public? As a user, can I figure out the GPS if I needed to? Or if I don't have a phone? What is the system that you need to install as government in places so that I know how long it is before a driver gets here, or a jeepney gets here, a bus gets here? And then as you move up, maybe there's a coordination system that allows a jeepney, when it arrives, to say to the bus, “delay yourself a minute, so that we can transfer correctly.”
Your question is about the skillset. One is understanding API's and data sets and the exchange of information. So you need people who can do that, and you also need people who understand the social realm. What injustice is going on here? Or are we hiding some sort of injustice, so that improving the service one way could mean furthering that injustice? Or creating new ones or excluding people? I think you need people to think about what the goal of the system is — What are we trying to achieve? — and the technologies themselves to design all of these systems or design how they interact.
Bryan: As a final question, this is something that I've been asking pretty much every smart person I know. I'm curious: what does the phrase “urban technology” mean to you? How would you define it, in your own words?
Benjie: When you say urban technology, I would like it to be the technology that allows us to live in large groups better, and to manage deciding and engaging with each other better. And it's not about efficiency, but about the facilitation of relationships—in terms of transportation, it’s about managing transportation demand, or sharing loads with your neighbors, or just finding beauty in the city.
Thanks for reading this special edition of the Urban Technology at University of Michigan newsletter. If you enjoyed this experimental podcast, please let us know by liking the episode on Substack, subscribing, or leaving a comment.
We also want to know what stories you are curious about when it comes to urban technology. Let us know, and we’ll track ‘em down.
From leafy Ann Arbor, I’m Bryan Boyer. Production and mixing by Alex Trajano. Charlie Keenan on editing. Until next time!
🏃🎧🏃🎙🏃
Urban Technology at University of Michigan week 112