I don't know. I think we've found that packet-like networks (cars) are more robust than the alternative (trains). Robustness gets built into rail systems by adding more tracks that sit unused most of the time so you can bypass problem sections. Otherwise, a train breaks down/derails/rails go out of alignment, etc. and not just that line fails, but the entire systems is impacted. In other words, whatever solution they start with, will be overcome by the event of a fleet of self-driving taxibots.
Still, I think it's fair to view the dismantlement of NYC's above-ground rail system as having been a mistake in many ways. It was replaced by fleets of yellow cabs, and while living conditions in buildings are better, and more light gets to the streets, the road system has become more difficult to deal with.
Perhaps the better solution is to force all non-taxibot traffic (except for delivery trucks) to park in peripheral garages in New Jersey or somewhere and force everybody to taxibot where they want to go?
hmm...on second thought, I guess this is a hard problem without a single solution?
The packets in this case have an enormous overhead. A four-car train takes up about as much space as 4 or 5 cars, but can carry 720 people compared to a best case of about 20 for the cars.
I crunched the math on this a little while back as a fun exercise:
To replace a single rush-hour NYC subway train with all single-occupancy vehicles, based on the average length of car in the US, would occupy every lane in a single direction on Market St in San Francisco from the waterfront all the way to Van Ness.
In other words, in order to get a subway train full of people into the city, but in cars instead, the lineup of cars (spread over multiple lanes!) would stretch the entirety of downtown SF. And this assumes perfect packing - i.e., bumper to bumper.
Side note: I'm quite optimistic about self-driving cars, but it never ceases to amaze me how many self-driving car proponents seem to be utopian futurists who are entirely reality-resistant. No, self-driving pod cars will not produce greater capacity than a heavy rail mass transit network, are you even serious. They may very well become a key fixture of our transportation infrastructure, but there is simply no physical way fleets of taxibots will ever rival the throughput of a NYC subway train at rush hour.
I totally agree with you on the capacity standpoint. Door-to-train and train-to-door remain stubborn issues with "circuit" style transport systems though.
I don't live in either city, but I do live in a top-10 metro with the second largest subway system in the U.S. For me to get to the train system from my house is currently a 30-45 minute drive (there's a bus, but for "reasons" it's even worse). Stop-to-stop my ride is 40 minutes. To get from the nearest station to my work is a 50 minute walk or a half hour bus ride. My commute then using the best available mass transit is a little over 2 hours or 3 hours if I assume we live in a post-surface vehicle world. Even if I move my household to one of the single bedrooms built right next to the station, it's still over an hour door to door with half of it on a surface bus or most of it walking (and now it's 2 hours again).
Ignoring all that mess, I can just drive and it's a 30-45 minute (the length of just the train leg) easy drive from my garage to my work's garage.
So I can spend 4-6 hours a day taking mass transit, or I can spend 1-1.5 hours a day in a car.
For what it's worth, I'm a huge proponent of mass transit and subways, and think the U.S. is pitifully behind in every way, I personally wish we engaged in the articles plan, and in many of the proposed mass transit and livable cities initiatives. But there are sometimes good reasons why people take cars over mass transit, even when it's as good as it gets where I live or in NYC and all of these plans are decades away from happening and changing things. The important thing is for stopgap ideas like cars not to get in the way of these kinds of improvements in they way that they have.
The car is not a "stopgap idea". It is the direct cause of the terrible design of most American cities (ie, urban sprawl). It is, in several important ways, the reason why public transit in the US is terrible.
Yeah sure, there's all kinds of ways to pack lots of people into a little space and get them marginally close to where they started from and where they need to get to. But there's still the first and last legs of that journey. Even in Paris, where no spot in the city is further than 500m from a station, means you may still have to walk a km in the worst case.
It's fine when holding nothing, now do it with a couple armfuls of groceries or whatever.
A car free city eliminates cars, which carry around big empty storage boxes in the back all day, and can go from right where you are to right where you need to go. Bikes are an "ok" solution for some segment of the population, but not for every.
A quick example, my 80+ year old father, who walks with difficulty with a cane, would not be able to live in a car free, train only city at all, and bikes are not an option for him for the same reasons. But he still insists on working every day because retirement bored him. There's lots of people like my father, old, handicapped, both, how do you design a mass transit system that can accommodate people like him?
That's why transit systems designed by healthy 20-40 something internet forum people are usually bad ideas, they optimize for these kind of best-case scenarios, but there's so many edge cases in humans that these solutions work poorly in general. Cars work well in general, but a terribly many of them in places like NYC are personal vehicles used for just a few minutes a day, adding to congestion.
There's only ~13k Taxis in NYC and ~40k other for-hire vehicles. What if we doubled those numbers, and killed off all other personal non-business vehicles entirely (the majority of vehicles in the city)? Sounds like that solves pretty much all of the problems.
Okay it doesn't solve all of the problems, you still have to deal with Taxi drivers.
Bikes can actually be better for people with mobility issues than cars. Biking is pretty low impact and uses less energy than walking, so for some people for whom walking is a challenge, biking might be possible. With safe bike infrastructure, these people can still get around without cars. As a bonus, the exercise will keep them mobile longer.
> There's lots of people like my father, old, handicapped, both, how do you design a mass transit system that can accommodate people like him?
You build an ADA compliant system. Compare the NYC subway, built long before the ADA, and the DC Metro. Except when elevators are broken (a big problem given WMATA's incompetent stewardship of the system), you can use pretty much the entire DC metro system in a wheelchair. Escalators are in every station. In Japan IIRC there were markings in the walkways of train stations, airports, and sidewalks to help the blind navigate, and we could incorporate that in the States.
Right, the system can be very easy to get around in, but it's the system to destination legs that can be impossible. This is almost always bridged with some kind of handicapped-accessible bus system in addition to the normal bus system. But then we're back to surface vehicles. Trains can never provide door-to-door service, it's a classic "last mile" problem.
I would never suggest eliminating car travel; just providing more efficient alternatives for the most common use cases. 90% or more of people are physically capable of using mass transit. Why should they take up 30 times as much space as they need to just because the other 10% need to use a less efficient means of transportation?
To duplicate the capacity of the trains, you'd need enormous amounts of parking and bridges. Someone crunched the numbers and found you'd need a parking lot the size of Manhattan and at least two layers deep to accommodate all of the commuters if they drove, as well as 48 additional 8 lane bridges.
Right, but with a self-driving taxi system you don't need a discrete car per person that's only used for transit in and out of the city. You can use fewer and reuse them over and over and over again during the course of a day. This is both a more flexible and more robust system than trains. Plus, given a grid system like Manhattan, blockages can be routed around.
People don't seem to believe me, but we're all on the internet aren't we?
Still, I think it's fair to view the dismantlement of NYC's above-ground rail system as having been a mistake in many ways. It was replaced by fleets of yellow cabs, and while living conditions in buildings are better, and more light gets to the streets, the road system has become more difficult to deal with.
Perhaps the better solution is to force all non-taxibot traffic (except for delivery trucks) to park in peripheral garages in New Jersey or somewhere and force everybody to taxibot where they want to go?
hmm...on second thought, I guess this is a hard problem without a single solution?