Engineer Scotty left a long and thoughtful comment on my recent post about vehicles for roads, streets and country roads. The whole thing is worth reading, especially for short-term recommendations, but I want to focus on this part:
But we still have need for intra-city mobility of the sort that cannot be accommodated with streets alone. Even if SOVs were banned, trucks and busses still need some way to get around to serve medium-distant trips. Cities need arterials of some sort to function.
I think that when he wrote that, Scotty hadn't yet read my subsequent post where I argue that trucks and buses will always kill and intimidate, and the only way to transport large loads and large groups of people without disrupting the street is by train.
The problem with urban "arterials" of any kind is that they don't work the way everybody seems to think they do. As I argued back in 2008, the nucleus of a city is a transportation pivot of some kind: a crossroads, an intermodal transfer point like a port or a railroad station, the entrance to a pass or a valley, or even just a rest stop.
It is the very fact of having to slow down, stop or rest that makes a city what it is. If the travelers and drivers don't stop, they won't spend money, and there is no reason for businesses to set up there. If there are no businesses, there are no jobs and thus no reason for anyone else to live there.
Bypasses kill cities in two ways. They can simply remove any reason to slow down or stop at all, so that every customer just goes flying by, and all the commerce goes to another city. Or they can move the crossroads away from the old downtown to a new place on the outskirts, which becomes the new downtown (if you can call it that).
Yesterday, the Urbanophile tweeted links to a Detroit Free Press article about the vast open spaces in that city, and a Chicagoist post about similar developments on a smaller scale in Chicago's South Side. My reply was simple: both places have been bypassed. They're no longer at any kind of pivot point on the way to anywhere. They no longer have access to anything useful. So why would anyone want to live or make anything there?
Of course, Scotty isn't talking about a bypass, but some kind of moderate arterial where through traffic is prioritized over local side traffic. This is the same problem, just to a lesser degree. The easier you make it for traffic to pass through the city - the more you facilitate mobility - the less commerce you have. In cities, mobility is the enemy of commerce.
I'm simplifying, you understand. A city can't be a complete bottleneck to the point where people make their own bypasses, for example by going through other cities. The people who sell things to travelers, well they have to get those things somehow. Sometimes you need to go outside easy walking distance to another neighborhood, and you don't want that to be too difficult.
That kind of wider access can be provided by transit. The ideal urban transit is the subway, because it's mostly invisible, but you can get a lot with elevateds and "green tracks" trolleys. Bicycles, of course, are a great way of providing medium-range mobility.
In the past the push for bypasses came from two directions: the shippers and travelers who didn't want to be slowed down, and the neighbors who didn't want traffic. But the reason the neighbors didn't want traffic was because the traffic was dangerous, noisy and polluting. Cities like Detroit and neighborhoods like Bronzeville will never come back until through traffic can cross them at a safe, human scale.
The key is to balance mobility and access with commerce, comfort and safety. Speed can be mitigated by the use of rails instead of tires, and lightweight vehicles like bicycles and pushcarts. Separate faster rail traffic from slower foot traffic if you must, but if you make your city too easy to go through or around, then you risk making it irrelevant.
6 comments:
There are great advantages for being more transparent to traffic - it makes it easy for secondary centers to form right around the city, which makes the entire region work better. Think of how the RER made traffic through Paris fast.
I think the importance is to not separate roads and streets. Rather, infrastructure should be shared when reasonably possible. On transit, this means the same trains that get people between suburbs should also serve the city, and vice versa. This is what you get with RER-style systems.
The concept of slowing down traffic may have worked when there were no cars, but today, forget about it. As John Adams notes, today's traffic restraint policies do nothing to slow down traffic in growing exurbs and edge cities, but instead make slow it down in the cities, where it's already fairly slow.
Good article, though I've got a few comments and questions:
1) I had seen your other post. On the safety of rail vs busses or trucks: Rail has the advantage of you know where it's going to be (barring a derailment), and rather than coming as a collection of smaller vehicles separated by smaller gaps, it comes as larger vehicles (trains) separated by larger gaps, giving better crossing opportunities. Busses and trucks have the advantage of being able to maneuver more easily, and if one limits oneself to busses/trucks operated by professional drivers, are capable of safe operation. I worry about cars more than I worry about busses or trucks, even though the latter is more likely to cause a fatal injury if it hits you. How autonomous vehicles will change this calculus will be an interesting question.
And as a practical matter, unless we run rails everywhere (such that every home and business is within 100m or so of a railspur), we'll need trucks for local deliveries. And if we do start to run rails everywhere, then trains will start to have a safety profile which more resemble that of trucks and busses--rail will lose its predictability and its relative lack of frequency. (You almost sound like you are proposing an urban environment where nearly every vehicle that has a motor runs on steel wheels).
Rail does have the decided advantage that if you decide to grade-separate it, its a lot easier and cleaner to do.
2) While portage has long been a raison d'etre for many cities, it certainly isn't necessary today. San Jose isn't really on the way to anywhere--it's primary product is intellectual, not physical, for instance. A "rest stop" isn't a good business model for any city larger than Breezewood, PA--there's too much competition in the market for traveler services. If a city can sit astride a bottleneck or a place of mode change (such as a harbor), that's a major advantage, but bulk freight movements aren't addressed by boulevards.
3) I fear you may be confusing different scales. Just because I may want travellers to stop in my city, doesn't mean I want them stopping every block from the proverbial gates to the town square. Many trips within a city, or from somewhere within a city to somewhere outside, are point-to-point. I agree that bypasses are often bad for cities, and while "boulevards" as discussed could be used to expedite through traffic (and reduce the chances it would stop), I was mainly concerned with traffic originating within a city, not with travelers or shippers trying to get through it. The boulevard is something that long predates the automobile, after all.
One other thing--the boulevard as I discuss it is a design to avoid the worst features of "stroads"--many of which fail by trying to support high speeds in an environment where it's not appropriate, and do so by restricting access. Boulevards can support volume, but aren't optimized for speed.
There are many types of "stroads" I am keen to avoid:
* As mentioned in the referenced comment, six-or-eight lane divided RIRO arterials with the occasional stoplight, when routed through urban fabric.
* Highways (or former highways) lined on either side with strip development fronting the road, often with parking lots between the buildings and the highway, and no place for pedestrians.
* One-way street grids with timed signals, designed to permit cars to travel through an urban area unimpeded.
Scotty: San Jose happens to be within an hour of the SFO airport, which is a pretty major transfer point for trans-Pacific flights, which is incredibly important for, oh say, consumer electronics businesses to be able to get to their factories in China or their business partners in Taiwan. And in some cases, every hour of travel time really does count, so being fairly close to an airport with direct flights to Taiwan makes a noticeable difference. Now, the airline routes are largely that way because SF already has a large number of people wanting to use these flights. But it's a symbiotic process between cities that create demand for transportation and the transportation infrastructure that channels business and population into cities.
Just a suggestion, but I think it might help to separate out freight and people. Cities don't necessarily benefit from maximizing the number of people whizzing about at high speed within or near their territory, but I suspect that freight would be subject to a different analysis.
In fact, I'd analogize freight to communications. We don't think about it much anymore, but communications did not always whiz about cities at light speed (and these days, communications between two points in the same city may whiz out to some faraway places before coming back, but so fast no one notices). Despite some predictions to the contrary, however, such advances in telecommunications have not in fact killed cities.
I suspect the same would be true of freight--it could whiz about arbitrarily fast without doing any particular harm to the development of cities. The issue, then, is how to decouple freight moving around cities from people moving around cities.
In the past people approaching this question or a variant thereof have imagined things like networks of large pneumatic tubes--which may well still be a good idea. Autonomous vehicles, though, might allow for more pragmatic approaches.
They already do in many contexts, in fact (autonomous vehicles move around freight in many warehouses, factories, hospitals, and so forth). But one could suggest Google-type technologies could be used to more easily incorporate that approach to freight movement into cities without the need for massive expenses in infrastructure retrofitting.
Anyway, just a thought that might have some clarifying benefits. You'd still need to sort out exactly how you want to handle people.
When is mobility the enemy of commerce? When mobility kills density.
Suppose drivers of wealth can be categorized into two categories: a) accessing people, b) accessing fixed resources. The first driver scales as how many people you can reach (density * mobility^(fractal power)) while the second goes as how much area one can cover (~ mobility^(fractal power)). The fractal power is somewhere between 1 and 2 depending on the transportation layout.
Cities derive their wealth and existence primarily from the first category. Therefore, if increasing speed in order to increase mobility by 50-70% kills more than half your density, you're losing wealth generation.
A country as a whole may derive a substantial proportion of its wealth from resources (tourism, mining, distributed manufacturing, freight, your own big back yard). Mobility is key to generating this wealth, regardless of local density.
A transportation network that balances in-city density with mobility, and promotes regional mobility is then the ideal.
(As an aside, bypasses serve a role in promoting regional wealth. While, they generate little to no wealth for the city being bypasses, the individual city should not be considered in isolation.)
Post a Comment