7 ms·

I like Jarrett Walker. He does come at the problem from a certain perspective though -- he wants to solve problems with mass transit. Just like he talks about

I like Jarrett Walker. He does come at the problem from a certain perspective though -- he wants to solve problems with mass transit. Just like he talks about there being immutable laws of geometry, there are some immutable laws of transit. Mass transit fundamentally depends on having a lot of people who want to get from point A to point B, all at the same time. Since this is never true, you have to relax the requirements. You can make people wait until there are enough of them, you can make them walk to or from a stop, you can make them detour, slow down, stop and then speed up to let others get on and off, and you can make them switch vehicles. Fundamentally those are your options given some set of vehicles, whether they're planes, trains, or automobiles. All forms of transit exist on the tradeoff surface between these possibilities. As population density goes up, it's easier to find lots of trips that are close enough in the five dimensional space of origin, destination, time. Where I perhaps differ from Jarrett is that I see no problem in having transit options that have one or two people per vehicle. The 2-5 person capacity mass transit vehicle has been a very underexplored space because someone has to own it and someone has to drive it. Services like Uberpool and Lyft Line are exploring this because smartphones make it possible, and this will only increase when you don't have to pay someone who isn't riding to operate the vehicle.

imgabe 9y ago
> Mass transit fundamentally depends on having a lot of people who want to get from point A to point B, all at the same time. Since this is never true, you have to relax the requirements. There is also induced demand. If you make it very easy to get from point a to point b. Lots of people will want to live at a and work at b or vice versa. Right now we focus most of our energy and resources on making it as easy as possible for everyone to drive their car, so that's what they do.
Arnt 9y ago
Jarrett doesn't have a problem with small-vehicle options either. He just says: In a dense big city that has to be a sideshow and therefore discussing it is a harmful distraction. (I've seen small-capacity "mass" transit, btw. Never two-person-max. But I've seen ten-person microbuses in several cities. Where I live there are a few that feed minor outlying hubs.)
- djsumdog 9y ago
  Can we say Jarrett is visually impaired and can't drive? Let's also say Jarrett is a school teacher and earns less than $40k/year. Jarrett has very limited options now. This person can live in one of the few (like three? four if you stretch?) cities in the US with decent enough transport to get around, that aren't insanely expensive to live in ... or just have to get lucky and find a house in walking distance of the school and a grocery store and a bus stop to the city. Jarrett isn't going to be able to afford a Lyft/Uber and is very limited on their ability to even be social.
  - pjc50 9y ago
    "has to be a minority option" is an arithmetic necessity of density: you can't have everyone owning cars in a high-density city because it's physically impossible to fit them on the roads.
    - Arnt 9y ago
      Right. I suppose Jarrett would say that if a place is dense enough to have more than occasional traffic jams, then it already is too dense that its transport problems can be solved by road construction.
djsumdog 9y ago
America use to have more rail at one point in time than Europe has today. It wasn't that long ago. We're talking less than 70 years. If we start building infrastructure in our cities again for rail and express buses, you will see things being to naturally be more compact. Business spring up around rail stations, and people who want to live close to one another fill in those spaces and move closer to the city. The "America is too spread out" argument is total bullshit. At the very least, make the cities have good transport and walkable areas. It can happen in a very short time frame. We lost it in a short time frame. Northern Indiana has the same population density of Scotland, so it can work if we build it. We use to have trains from Nashville to Atlanta. The famous "Chattanooga Choo-choo" doesn't even have an AmTrak anymore. Transportation is the largest factor in the ability of people to move up from poverty. This isn't some subjective statement. There are tons of objective studies to support this[1]. The idea that you need to own a car (and insurance an maintenance) in most of America is pretty insane. A car shouldn't ever be a necessity. Associating it with freedom, as the American car companies have, and buses with poverty, is ass fucking backwards. [1]: https://www.youtube.com/watch?v=-cjfTG8DbwA&t=316s https://www.youtube.com/watch?v=-cjfTG8DbwA&t=316s
- Karnickel 9y ago
  > The "America is too spread out" argument is total bullshit. In support of this statement, here is a map of "mega regions": http://www.america2050.org/sync/elements/america2050map.png http://www.america2050.org/sync/elements/america2050map.png Source: http://www.america2050.org/maps/ http://www.america2050.org/maps/ Alternative: "The Mega-Regions of North America" http://martinprosperity.org/content/the-mega-regions-of-north-america/ http://martinprosperity.org/content/the-mega-regions-of-nort... > All told, these dozen mega-regions span 243 metropolitan areas in the U.S. and Canada, more than six in ten of all U.S. metros. They have a combined population of more than 230 million people, including 215 million from the United States or 70 percent of the U.S. population. Together, they produce more than $13 trillion dollars in economic output, equivalent to three-quarters of America’s total GDP. Vast regions of the US don't even matter for the purpose of this discussion. When you only look at those regions the problem is no different than in a lot of other countries. That is also the geographical level where public transport makes sense, it's per-region (and different regions may want to/have to solve it in very different ways since their respective situations are different), not one solution for the whole country.
schiffern 9y ago
> 2-5 person capacity Musk says a Model X derived minibus has "surprisingly high" people density.[1] The standard Model X seats 7, so I'd say at least 10. I'm assuming similar volume-stretching tricks as the Model 3, (pushing the dash forward, glass roof for headroom) in a microbus package. > All forms of transit exist on the tradeoff surface between these possibilities. As population density goes up, it's easier to find lots of trips that are close enough in the five dimensional space of origin, destination, time. I think the real value of "big data" (something Jarrett dismisses as "big detail" in the article) is that we can plot every trip in that 5D space, deriving a "typical" density curve for any given day of the week. With that, the fleet management software can optimally allocate vehicles. Clustering nearby points (trips) in this 5D space should optimally combine trips to minimize detours to individual houses (by grouping in the origin/origin and destination/destination 4-planes), combine trips across town (by grouping in the origin/destination 4-plane), and dynamically adjust the wait time to combine trips to/from low traffic locations (by grouping in the time/origin and time/destination 3-planes). Really it should trade off between all of these factors. The challenge, of course, is that you don't know everyone's route ahead of time. Good prediction is important. Essentially you're dynamically adjusting the bus schedule and route in response to spatial and temporal demand patterns. And with 10 passengers and no aisle[2], the maximum passenger density is similar to a bus (the "geometry" problem) So it seems like this can be more efficient than even a fixed bus schedule. It seems like a decently large (100k+) multi-agent whole city transit simulation could settle the matter one way or the other. Who's plugging autonomous taxi routing into these? This[3][4] looks promising. Accounts for induced demand. More publications on his website.[5] [1] https://twitter.com/elonmusk/status/759153112993017856 https://twitter.com/elonmusk/status/759153112993017856 [2] "Traffic congestion would improve due to increased passenger areal density by eliminating the center aisle and putting seats where there are currently entryways, and matching acceleration and braking to other vehicles, thus avoiding the inertial impedance to smooth traffic flow of traditional heavy buses." https://www.tesla.com/blog/master-plan-part-deux https://www.tesla.com/blog/master-plan-part-deux [3] https://www.ethz.ch/content/dam/ethz/special-interest/baug/ivt/ivt-dam/vpl/reports/1201-1300/ab1238.pdf https://www.ethz.ch/content/dam/ethz/special-interest/baug/i... [4] https://github.com/sebhoerl https://github.com/sebhoerl [5] http://www.ivt.ethz.ch/en/people/person-detail.html?persid=224756 http://www.ivt.ethz.ch/en/people/person-detail.html?persid=2...
barrkel 9y ago
He criticizes people arguing for smaller cars closer together as being a linear solution to an exponential problem, but mass transit is also a linear solution to the same problem. Build better mass transit, and houses near the stations become more attractive and people move there, which increases usage, and so on. And looking at it from the other end, ever smaller cars jammed ever closer together at the limit tends to the same density as mass transit. (We're nowhere near that now, but I can see us getting much, much closer in an automated driving future.) The key differences end up being routing flexibility, packet size and latency. Mass transit is worse in all three (less flexible and more waiting time are given with mass transit; I'm taking packet size to be a measurement forced space sharing, so big is bad). Its only real advantage is being at the limit for density, a density cars will likely never be able to match. But because it's still linear, it's only a constant factor improvement. (Personally, I travel around London by motorcycle and scooter because public transport is ludicrously inefficient; it multiplies journey time by 2x to 5x. Ironically, it works best if you're elite, i.e. live inside the circle line; but then you might as well cycle.)
- arethuza 9y ago
  It takes me 35 minutes to take the train from Fife to central Edinburgh - absolutely no chance of doing it by car in that time in the morning. If I had to drive into the center of Edinburgh in the morning I'd probably give it well over an hour and that would ignore the hassle and costs of parking once I get here. Edit: And with the train I can sit and sip coffee and try and spot seals and herons and enjoy the splendid views....
  - barrkel 9y ago
    Fife isn't city living; that's not relevant to the article. Mind you, with your self-driving vehicle, you'll be able to sip coffee and spot herons. And I believe packetized self-driving vehicles grouping together into convoys ought to be able to exceed 200mph in the future. That future isn't today, but the article is explicitly forward looking, about geometry that doesn't change. I think the article is ideologically blind to how mass transit has similar supply and demand problems (e.g. when I take the train home, I often have to skip multiple departures due to overcrowding), and it drastically underestimates the potential for technology change.
twelvechairs 9y ago
There's way bigger problems for 'the 2-5 person capcity mass transit vehicle' (by which you basically mean 'car') than people owning and driving the vehicle. Time is completely out of your hands (am I first or fifth in? What route will it take? How long are we going to wait for this person to show up?). Its not solving infrastructure issues massively (cars still need roads - you might make it attractive by having a lane for 3 or more people vechiles though). Its not really a system suited for dense walkable centres (see European cities) - and we are going further that way. And whats the objective benefit vs a train or tram? You are still in a space with strangers. There are places in the world where labour costs are so low that this doesnt figure. There are also those where shared rides are common (e.g. taxis that run 'routes') - I'm not sure in any of these places they are considered a better solution than public transport at all. Of course they have a place, and I think Walker acknowledges this too, but its not really a game changer like building proper public transport can be.
- improbable22 9y ago
  Re "the 2-5 person capacity mass transit vehicle": Lots of cities have enough density to support good dense tram / train / bus networks in the centre, during the day. But quite a lot of the people who would use that, also have to go further out sometimes, or need to get home later at night sometimes. If mass transit can't do this too, then they need to make a whole different plan... often meaning a car, and a house further out to afford parking. This seems like the opening for smaller vehicles. They can perhaps fill in the gaps between lines, and the long waits between night busses, in a way that couldn't be done before, essentially because human taxi drivers are too expensive. It's possible this could help make proper public transport viable in lots more places than it is now.
- improbable22 9y ago
  Or perhaps to put that another way: In a city with great public transport, lots of the person-miles are done on full trains, running back to back all rush hour, in a way that nothing else can really replace. But for the system to be useful, it has to maintain reasonable frequencies until midnight, out to a lot of points. This results in a lot of vehicle-miles being done almost empty. But they are really essential. It's the second class which maybe small automated vehicles could replace. Lots of cities only cover the first class, and wonder why so many rush-hour commuters don't use it...
rst 9y ago
Incorrect. New York has lots of people trying to get from point A to point B pretty much all the time, and other American systems with half-decent mass transit (Boston, DC) also see those systems get heavy use, to the point that shutdown of a line is a major disruption. Which is also the universal experience in Europe; Paris has added several mass transit lines in recent years, and has had no problems keeping them full.
robbiemitchell 9y ago
I often think about smaller vehicles. Would it be possible and better to have a rail system with lots of small pods slinging around instead of bus-sized containers? Each station would need to allow for the pods to pull over without blocking the flow of continuing traffic.

imgabe 9y ago

> Mass transit fundamentally depends on having a lot of people who want to get from point A to point B, all at the same time. Since this is never true, you have to relax the requirements. There is also induced demand. If you make it very easy to get from point a to point b. Lots of people will want to live at a and work at b or vice versa. Right now we focus most of our energy and resources on making it as easy as possible for everyone to drive their car, so that's what they do.

Arnt 9y ago

Jarrett doesn't have a problem with small-vehicle options either. He just says: In a dense big city that has to be a sideshow and therefore discussing it is a harmful distraction. (I've seen small-capacity "mass" transit, btw. Never two-person-max. But I've seen ten-person microbuses in several cities. Where I live there are a few that feed minor outlying hubs.)

djsumdog 9y ago
Can we say Jarrett is visually impaired and can't drive? Let's also say Jarrett is a school teacher and earns less than $40k/year. Jarrett has very limited options now. This person can live in one of the few (like three? four if you stretch?) cities in the US with decent enough transport to get around, that aren't insanely expensive to live in ... or just have to get lucky and find a house in walking distance of the school and a grocery store and a bus stop to the city. Jarrett isn't going to be able to afford a Lyft/Uber and is very limited on their ability to even be social.
- pjc50 9y ago
  "has to be a minority option" is an arithmetic necessity of density: you can't have everyone owning cars in a high-density city because it's physically impossible to fit them on the roads.
  - Arnt 9y ago
    Right. I suppose Jarrett would say that if a place is dense enough to have more than occasional traffic jams, then it already is too dense that its transport problems can be solved by road construction.

djsumdog 9y ago

America use to have more rail at one point in time than Europe has today. It wasn't that long ago. We're talking less than 70 years. If we start building infrastructure in our cities again for rail and express buses, you will see things being to naturally be more compact. Business spring up around rail stations, and people who want to live close to one another fill in those spaces and move closer to the city. The "America is too spread out" argument is total bullshit. At the very least, make the cities have good transport and walkable areas. It can happen in a very short time frame. We lost it in a short time frame. Northern Indiana has the same population density of Scotland, so it can work if we build it. We use to have trains from Nashville to Atlanta. The famous "Chattanooga Choo-choo" doesn't even have an AmTrak anymore. Transportation is the largest factor in the ability of people to move up from poverty. This isn't some subjective statement. There are tons of objective studies to support this[1]. The idea that you need to own a car (and insurance an maintenance) in most of America is pretty insane. A car shouldn't ever be a necessity. Associating it with freedom, as the American car companies have, and buses with poverty, is ass fucking backwards. [1]: https://www.youtube.com/watch?v=-cjfTG8DbwA&t=316s https://www.youtube.com/watch?v=-cjfTG8DbwA&t=316s

Karnickel 9y ago
> The "America is too spread out" argument is total bullshit. In support of this statement, here is a map of "mega regions": http://www.america2050.org/sync/elements/america2050map.png http://www.america2050.org/sync/elements/america2050map.png Source: http://www.america2050.org/maps/ http://www.america2050.org/maps/ Alternative: "The Mega-Regions of North America" http://martinprosperity.org/content/the-mega-regions-of-north-america/ http://martinprosperity.org/content/the-mega-regions-of-nort... > All told, these dozen mega-regions span 243 metropolitan areas in the U.S. and Canada, more than six in ten of all U.S. metros. They have a combined population of more than 230 million people, including 215 million from the United States or 70 percent of the U.S. population. Together, they produce more than $13 trillion dollars in economic output, equivalent to three-quarters of America’s total GDP. Vast regions of the US don't even matter for the purpose of this discussion. When you only look at those regions the problem is no different than in a lot of other countries. That is also the geographical level where public transport makes sense, it's per-region (and different regions may want to/have to solve it in very different ways since their respective situations are different), not one solution for the whole country.

schiffern 9y ago

> 2-5 person capacity Musk says a Model X derived minibus has "surprisingly high" people density.[1] The standard Model X seats 7, so I'd say at least 10. I'm assuming similar volume-stretching tricks as the Model 3, (pushing the dash forward, glass roof for headroom) in a microbus package. > All forms of transit exist on the tradeoff surface between these possibilities. As population density goes up, it's easier to find lots of trips that are close enough in the five dimensional space of origin, destination, time. I think the real value of "big data" (something Jarrett dismisses as "big detail" in the article) is that we can plot every trip in that 5D space, deriving a "typical" density curve for any given day of the week. With that, the fleet management software can optimally allocate vehicles. Clustering nearby points (trips) in this 5D space should optimally combine trips to minimize detours to individual houses (by grouping in the origin/origin and destination/destination 4-planes), combine trips across town (by grouping in the origin/destination 4-plane), and dynamically adjust the wait time to combine trips to/from low traffic locations (by grouping in the time/origin and time/destination 3-planes). Really it should trade off between all of these factors. The challenge, of course, is that you don't know everyone's route ahead of time. Good prediction is important. Essentially you're dynamically adjusting the bus schedule and route in response to spatial and temporal demand patterns. And with 10 passengers and no aisle[2], the maximum passenger density is similar to a bus (the "geometry" problem) So it seems like this can be more efficient than even a fixed bus schedule. It seems like a decently large (100k+) multi-agent whole city transit simulation could settle the matter one way or the other. Who's plugging autonomous taxi routing into these? This[3][4] looks promising. Accounts for induced demand. More publications on his website.[5] [1] https://twitter.com/elonmusk/status/759153112993017856 https://twitter.com/elonmusk/status/759153112993017856 [2] "Traffic congestion would improve due to increased passenger areal density by eliminating the center aisle and putting seats where there are currently entryways, and matching acceleration and braking to other vehicles, thus avoiding the inertial impedance to smooth traffic flow of traditional heavy buses." https://www.tesla.com/blog/master-plan-part-deux https://www.tesla.com/blog/master-plan-part-deux [3] https://www.ethz.ch/content/dam/ethz/special-interest/baug/ivt/ivt-dam/vpl/reports/1201-1300/ab1238.pdf https://www.ethz.ch/content/dam/ethz/special-interest/baug/i... [4] https://github.com/sebhoerl https://github.com/sebhoerl [5] http://www.ivt.ethz.ch/en/people/person-detail.html?persid=224756 http://www.ivt.ethz.ch/en/people/person-detail.html?persid=2...

barrkel 9y ago

He criticizes people arguing for smaller cars closer together as being a linear solution to an exponential problem, but mass transit is also a linear solution to the same problem. Build better mass transit, and houses near the stations become more attractive and people move there, which increases usage, and so on. And looking at it from the other end, ever smaller cars jammed ever closer together at the limit tends to the same density as mass transit. (We're nowhere near that now, but I can see us getting much, much closer in an automated driving future.) The key differences end up being routing flexibility, packet size and latency. Mass transit is worse in all three (less flexible and more waiting time are given with mass transit; I'm taking packet size to be a measurement forced space sharing, so big is bad). Its only real advantage is being at the limit for density, a density cars will likely never be able to match. But because it's still linear, it's only a constant factor improvement. (Personally, I travel around London by motorcycle and scooter because public transport is ludicrously inefficient; it multiplies journey time by 2x to 5x. Ironically, it works best if you're elite, i.e. live inside the circle line; but then you might as well cycle.)

arethuza 9y ago
It takes me 35 minutes to take the train from Fife to central Edinburgh - absolutely no chance of doing it by car in that time in the morning. If I had to drive into the center of Edinburgh in the morning I'd probably give it well over an hour and that would ignore the hassle and costs of parking once I get here. Edit: And with the train I can sit and sip coffee and try and spot seals and herons and enjoy the splendid views....
- barrkel 9y ago
  Fife isn't city living; that's not relevant to the article. Mind you, with your self-driving vehicle, you'll be able to sip coffee and spot herons. And I believe packetized self-driving vehicles grouping together into convoys ought to be able to exceed 200mph in the future. That future isn't today, but the article is explicitly forward looking, about geometry that doesn't change. I think the article is ideologically blind to how mass transit has similar supply and demand problems (e.g. when I take the train home, I often have to skip multiple departures due to overcrowding), and it drastically underestimates the potential for technology change.

twelvechairs 9y ago

There's way bigger problems for 'the 2-5 person capcity mass transit vehicle' (by which you basically mean 'car') than people owning and driving the vehicle. Time is completely out of your hands (am I first or fifth in? What route will it take? How long are we going to wait for this person to show up?). Its not solving infrastructure issues massively (cars still need roads - you might make it attractive by having a lane for 3 or more people vechiles though). Its not really a system suited for dense walkable centres (see European cities) - and we are going further that way. And whats the objective benefit vs a train or tram? You are still in a space with strangers. There are places in the world where labour costs are so low that this doesnt figure. There are also those where shared rides are common (e.g. taxis that run 'routes') - I'm not sure in any of these places they are considered a better solution than public transport at all. Of course they have a place, and I think Walker acknowledges this too, but its not really a game changer like building proper public transport can be.

improbable22 9y ago
Re "the 2-5 person capacity mass transit vehicle": Lots of cities have enough density to support good dense tram / train / bus networks in the centre, during the day. But quite a lot of the people who would use that, also have to go further out sometimes, or need to get home later at night sometimes. If mass transit can't do this too, then they need to make a whole different plan... often meaning a car, and a house further out to afford parking. This seems like the opening for smaller vehicles. They can perhaps fill in the gaps between lines, and the long waits between night busses, in a way that couldn't be done before, essentially because human taxi drivers are too expensive. It's possible this could help make proper public transport viable in lots more places than it is now.
improbable22 9y ago
Or perhaps to put that another way: In a city with great public transport, lots of the person-miles are done on full trains, running back to back all rush hour, in a way that nothing else can really replace. But for the system to be useful, it has to maintain reasonable frequencies until midnight, out to a lot of points. This results in a lot of vehicle-miles being done almost empty. But they are really essential. It's the second class which maybe small automated vehicles could replace. Lots of cities only cover the first class, and wonder why so many rush-hour commuters don't use it...

rst 9y ago

Incorrect. New York has lots of people trying to get from point A to point B pretty much all the time, and other American systems with half-decent mass transit (Boston, DC) also see those systems get heavy use, to the point that shutdown of a line is a major disruption. Which is also the universal experience in Europe; Paris has added several mass transit lines in recent years, and has had no problems keeping them full.

robbiemitchell 9y ago

I often think about smaller vehicles. Would it be possible and better to have a rail system with lots of small pods slinging around instead of bus-sized containers? Each station would need to allow for the pods to pull over without blocking the flow of continuing traffic.