As Chuck says, "We do a disservice to our communities when we treat streets as if they were roads, when we ignore the complex environments streets are meant to create and treat them as if they were simple throughput models. Streets need to be designed block by block. Those designs need to be responsive and adaptable."
Yes, it's time for SWFL to rethink how it designs streets.
Strong Towns Blog
Engineers should not design streets
by Charles Marohn
Last Friday I was participating in the 5th Annual Mayor's Bike Ride in Duluth following a week spent sharing the Strong Towns message on the Iron Range. The friendly woman riding next to me asked me what could be done to to better educate engineers so they would start to build streets that were about more than simply about moving cars. My answer rejected the premise of the question: We should not be asking engineers to design streets.
A quick review for those of you that are new here (which might be up to half the audience -- amazing). Roads and streets are two separate things. The function of a road is to connect productive places. You can think of a road as a refinement of the railroad -- a road on rails -- where people board in one place, depart in another and there is a high speed connection between the two.
In contrast, the function of a street is to serve as a platform for building wealth. On a street, we're attempting to grow the complex ecosystem that produces community wealth. In these environments, people (outside of their automobile) are the indicator species of success. So, in short, with a street we're trying to create environments where humans, and human interaction, flourish.
Engineers are well-suited to constructing roads. Road environments are quite simple and, thus, lend themselves well to things like design manuals and uniform guidelines. There are only so many variables and the relationship they have to each other is fairly straightforward. In the United States, we have tested, refined and codified an engineering approach to roads that is pretty amazing and, in terms of engineering, the envy of the world.
There are two primary variables for designing a road: design speed and projected traffic volume. From those two numbers, we can derive the number of lanes, lane width, shoulder width, the width of clear zones and the allowable horizontal and vertical curvature. From those factors, we can specify all the pavement markings and signage that are necessary. We can then monitor things like the Level of Service, the 85th percentile speed and traffic counts to optimize how the road functions over time. Engineers are really good at this.
Engineers are not good at building streets nor, I would argue, can the typical engineer readily become good at it. Streets that produce wealth for a community are complex environments. They do not lend themselves well to rote standards or even design guidelines. There are numerous variables at play that interact with each other, forming feedback loops and changing in ways that are impossible to predict.
Consider just one variable: the future of the adjacent land. The operative component of building wealth on a street is building. Who owns the property? What are they going to do with it? What is their capacity? Will they stick with it? Will they find the love of their life and move across the country? Each property has a near infinite set of complexities to it that change and respond to change, each of which is far more important to the wealth capacity of the street than, for example, lane width.
If we're trying to create an ecosystem that results in our indicator species (people) showing up in greater and greater numbers, we can't just focus on one or two variables. It can't be just design speed and volume. The natural ecosystem equivalent would be an observation that productive forests have trees and so we hire our forest engineers to go out and plant rows and rows of the optimum tree. It's obvious that, absent other flora and fauna, insects and bacteria, sunlight and rain and a myriad of other variables, the trees we are planting just aren't enough to get the ecosystem we're after.
If we're trying to create a natural ecosystem, we first have to recognize the environment we're in. A desert ecosystem will be far different than a northern forest. We then need to seed the basic elements, but we don't direct them day-to-day; we nurture them as they grow. If we know what we're after -- if we know our indicator species of success -- if we see the experiment getting way off track, we can intervene in small ways to nudge it back on course. We can introduce small changes and see how the system responds. Over time, our natural ecosystem will show us how it wants to grow.
We do a disservice to our communities when we treat streets as if they were roads, when we ignore the complex environments streets are meant to create and treat them as if they were simple throughput models. Streets need to be designed block by block. Those designs need to be responsive and adaptable.
Understanding that 99%+ of all streets that will exist a decade from now already exist today, what we're really talking about here in North America isn't building new streets but making good use of existing streets. The way we do this -- the way we design block by block in ways that are responsive and adaptable -- is to try things and see what works. Our tools are not traffic counters and code books but paint, cones and straw bales. Before we make any change permanent, we test it -- and possibly other variations -- first to see what works.
So if this isn't the job of an engineer -- and it's not -- who should design streets? The answer is as simple as it is radical: everyone. Building a productive street is a collective endeavor that involves the people who live on it, those who own property on it, those who traverse it as well as the myriad of professionals who have expertise they can lend to the discussion.
Put your least technical person on staff in charge of your next street. Empower them to meet with people, observe how people use the street and then experiment, in a low cost way, with different alternatives. Keep experimenting until you start to see your indicator species show up (outside of their cars, of course). Now you have a design you can hand over to your engineer to specify the technical stuff -- pavement thickness, paint specs, etc... -- and get the project built.
Engineers are highly competent at building roads. When you are trying to move automobiles quickly from one place to another, put your engineering in charge and do what they recommend. When you are trying to build a street -- when you are trying to make your city wealthier and more prosperous -- make your engineer one small voice in a larger chorus of people whose words and, especially, whose actions dictate what your design should be.
See Chuck Marohn's TED Talk on "The Difference between Streets and Roads"