4 ms·

I think there's an interesting take on this - we are on the verge of a huge technological shift with the advent of self-driving cars. Should we hold off on mas

• jlangenauer 8y ago
  No they won't, for the same reason that cities without public transport or minimal public transport have horrible traffic: the sheer density of people a bus, tram or light rail line can move far exceeds private vehicles with one or two people in them.
  • dantheman 8y ago
    My question is can there be public self-driving buses, cars, mini-buses -- that stop less are more direct and communicate to manage traffic?
    • lvh 8y ago
      Is there a metric where they’re not worse at what you’re suggesting than trains?
      • dantheman 8y ago
        Cost and flexibility. Buses are cheaper and more responsive to changing traffic patterns than trains. They are not as nice though. If for instance people in a given community want to start traveling to a different part of a city they either have to create new rail routes or people have to take long inconvenient trips.
        • CaptainZapp 8y ago
          > They are not as nice though. Ha! Apparently you haven't used Northern in the UK. Nice! (chuckle) I don't think it's really about nicety, but about capacity. A subway train can carry up to 1'000 passengers, depending on length and construction. You can have frequencies of a train every 90 seconds (example: line C in Prague, during rush hour). So you can transport up to 40'000 people per hour in each direction. Buses, while certainly much better than private vehicles, can never match that capacity. Alas, they're much cheaper and certainly more flexible.
        • sulam 8y ago
          The flip side is that they’re on the same roads as cars, which means exposed to the same congestion. And to the degree they get some people out of cars, they get other people into them.
        • ocschwar 8y ago
          > Buses are cheaper and more responsive to changing traffic patterns than trains True, but not very useful. The #1 Bus line in Boston takes a route that a trolley once took. And before that it was a horse drawn railcar. And before that it was a 10 seat omnibus carriage route. Same route for a mighty long time.
      • ghaff 8y ago
        For better or worse, trains have fixed routes and aren't very dynamic. I actually find it a bit surprising we don't see more of a formalized mini-bus system in the US given that it was pretty popular in some other countries even before GPS and apps. I guess UberPool serves a somewhat similar purpose. Perhaps there's not a lot of overlap between people who take Uber a lot and people who willingly step into a bus of any kind.
        • richardthered 8y ago
          In the US, the bus is generally seen as a "poor-person's" transport. Trains are considered middle-class transport.
      • rohit2412 8y ago
        Going doorstep to doorstep and probably lower time for the overall trip.
    • cimmanom 8y ago
      They're far enough away that it doesn't make sense to put off purchasing normal buses (which could probably be retrofitted with self-driving tech anyway). And certainly that doesn't obviate the value of things like planning and beginning to carry people on routes; setting up stops (signs or shelters); and setting up bus lanes. And rail transit is a whole other beast. Do you know how much higher a volume of people can be carried by a subway than by even the most efficient bus route running buses bumper-to-bumper (let alone private vehicles)? (The unwillingness of politicians to acknowledge this is one of the reasons the L train shutdown in Brooklyn next year is going to be such a shit show.)
  • ufo 8y ago
    Those pictures where they show car, bus, and bicycle density side by side really drive this point home: http://www.3fidi.com/wp-content/uploads/2012/11/60_people_bikes_cars_1_bus_greece.jpg http://www.3fidi.com/wp-content/uploads/2012/11/60_people_bi... http://www.citymetric.com/sites/default/files/images/car%20bike.jpg http://www.citymetric.com/sites/default/files/images/car%20b... And rail mass transit is even denser!
    • brianwawok 8y ago
      That picture doesn't make sense, because cars don't drive that close together, nor do bikes ride that close together. Show 60 cars at 45 mph vs 60 bikers at 15 mph for a better density view.
      • a_imho 8y ago
        nor do bikes ride that close together In my experiece they absolutely do, but it is rare to see a big pack together.
      • throwaway76543 8y ago
        You'll also want to take into account the 300% increase in speed. Transportation isn't about static density it's about throughput. Objects moving 3x faster can be 1/3 as dense and have the same overall throughput.
        • jcranmer 8y ago
          Higher speed actually translates into less throughput. Cars take up more than just their physical space; there is also a gap that is needed to stop safely in time--and this gap grows according to the square of velocity. (You can argue that people don't usually leave sufficient space to stop, but I will point out that self-driving cars, which prompted this discussion, are going to be following those basic safety rules)
          • throwaway76543 8y ago
  • ghaff 8y ago
    If autonomous vehicles arrive in the form that most people imagine, there will be a surge of traffic like you've never seen. Make something cheaper/easier and you get more of it. Furthermore, unless cities actually ban cars driving around without a passenger (which seems entirely possible), you'll have vast numbers of cars circling the block or heading out to the suburbs to park which their owners do their thing.
    • throwaway76543 8y ago
      No one is yet sure whether traffic will increase or decrease. As you describe, people may buy self driving cars and use the self-driving feature to drive them more often. Wealthy folks and people in rural areas are almost certain to take this route. Others predict that people won't buy cars at all and will instead rent them from ride services, who in turn may promote ride sharing on an unprecedented scale. Poor folks, for whom a car is a major expense, are almost certain to take this route, as well as city dwellers. Personally, I think we may see traffic increase in some areas and substantially decrease in others. I suspect the largest source of traffic -- the commute -- will shrink considerably as networks of reasonably dense self-driving vanpools and small busses pop up.
      • sandworm101 8y ago
        You have five or six revolutions happening all at once. Self-driving isnt going to revolutionize car ownership. People own cars for irrational reasons (fashion etc). Nor will self-drive suddenly make carpooling cool. >> reasonably dense self-driving vanpools and small busses pop up. Utopia much?
        • ghaff 8y ago
          Alternatives probably will decrease ownership at the margins. For example, I know a couple in SF who don't own a car and that wouldn't have been nearly as practical before Zipcar and Uber. However, it's not like paying a driver $100/day to drive a minibus around to pick people up and drop them off is likely the reason that isn't a widespread service today. So someone would need to explain to me why self-driving changes that equation.
• dexterdog 8y ago
  Self-driving buses are almost always going to be a cheaper option than SDCs assuming they are not grossly over-provisioned.
• imgabe 8y ago
  Even with self driving cars, they cannot move any where near the quantity of people in a given space that a train can. It's not going to be a solution for dense cities. Besides, if we are just on the cusp of having self-driving cars we should be able to deploy self-driving trains right now and reap all the benefits of autonomous transport immediately.
  • transfire 8y ago
    > self-driving trains Should have had these for a while now.
  • sbisson 8y ago
    We are. In London the DLR is the most obvious driverless network, but various parts of the Tube have been essentially driverless since the 1980s (initially on the Victoria Line, then on the Central after its resignalling, and on the Jubilee extension).
  • fs111 8y ago
    Those exist in many cities. The train to/from the Airport in Copenhagen for instance has no drivers.
  • Zxian 8y ago
    Many cities use automated light rail or subway systems. Vancouver, Canada has has such a system in place since late 1985. https://en.wikipedia.org/wiki/List_of_automated_urban_metro_subway_systems https://en.wikipedia.org/wiki/List_of_automated_urban_metro_...
  • masklinn 8y ago
    > Besides, if we are just on the cusp of having self-driving cars we should be able to deploy self-driving trains right now and reap all the benefits of autonomous transport immediately. Self-driving (GoA4) trains are close to 40 years old: https://en.wikipedia.org/wiki/Kobe_New_Transit https://en.wikipedia.org/wiki/Kobe_New_Transit https://en.wikipedia.org/wiki/Lille_Metro https://en.wikipedia.org/wiki/Lille_Metro
• paulie_a 8y ago
  Driverless cars are a pipe dream decades away.
  • JustSomeNobody 8y ago
    Indeed. It's a way to funnel money around. Companies doing self driving cars will get it 90% there, then struggle with the last 10% and eventually funding will dry up as investors lose interest and move on to something with faster returns.
  • marcoperaza 8y ago
    So are some of these public transportation projects. Seattle has started funding train lines that will finally open as late as 2040.
    • irq11 8y ago
      That’s especially true when rich ideologues attempt to defund or block every project that comes along. In Seattle’s case for ST1, at least half of the problem was that every plan was litigated to near-death by opponents whose only goal was to cynically create delay. Couple this with idiotic fantasy projects (like monorails that go nowhere - the self-driving car of the mid-aughts!), and you have a self-fulfilling prophecy.
    • paulie_a 8y ago
      I don't see how that is in way relevant to self driving cars. Is Seattle funding self driving vehicles also?
• lvh 8y ago
  An army of self driving cars doesn’t get you the density of a train when space is already at a premium, which is the case in many urban centers. Even the source acknowledges this; they just downplay how many centers it’s true for. It’s an interesting case of doublethink to suggest that the programs are unwelcome but also lead to gentrification. Apparently a lot of people do like transit :)
• codingdave 8y ago
  Self-driving cars may help with traffic, but not with pollution or environmental issues. New mass transit projects should try to take advantage of newer tech, but self driving vehicles and mass transit solve different problems, so I see no reason not to pursue both.
  • mcguire 8y ago
    Self driving cars would seem not to help with traffic. When I'm sitting at home or in my cube, my car is not out cruising the streets.
• protonimitate 8y ago
  As much as I want self driving technology to help ease the pains of public transit in major cities, I have a feeling were quite a bit further from that than we are being told. Public transit also helps out those who cannot afford alternative forms on transit. I don't see self-driving tech replacing transportation for low-income people, unless we move away from personal vehicle ownership.
• conception 8y ago
  I think your analogy how but you have what's the container so and what's the cargo shop mixed. Self driving cars will allow us to increase car densities and certainly lot road deaths but they will never be as efficient in transporting large numbers of people. It's just a matter of space. You can fit just so many more people in the space. This tweet's gif shows this pretty well https://twitter.com/adamjcs/status/529431691534667777?s=21 https://twitter.com/adamjcs/status/529431691534667777?s=21. It's a bandwidth problem.
  • Firadeoclus 8y ago
    I'd argue that efficiency in a transport network has to be measured in how many other people and facilities it puts into people's realistic reach. It's a bandwidth problem only up to a point. Streets aren't all constantly clogged. If you got rid of all powered transport there would be plenty of unused capacity. But many trips would no longer be feasible. Adding public transport helps enormously, but only along popular routes. Self driving cars may not have the density (though, if used exclusively, they could well be an order of magnitude more dense than ordinary cars) but they have the ability to take the most efficient route, and take you there comfortably with significant luggage. That makes trips possible that public transport can't serve.
• justinph 8y ago
  Aside from the density issue which others mention, self driving cars are un-proven technology. I get that a lot of smart people are working on them, but it's one thing to test a fleet of even a few hundred cars, but another entirely to have cities with thousands or hundreds of thousands of self driving cars. This is going to take longer to roll out and have the kinks sorted than you think.
• johannes1234321 8y ago
  Driverless has some benefits, but mind that the driver has a secondary purpose besides steering the vehicle: He is a security contact and brings saftey in different situations. Safety for women travelling at night, safety in case of an accident to handle evacuation etc., safety in case of obstacles on the road, ... automated systems can assist quite a lot, but won't handle all.
• reallymental 8y ago
  Why should they replace buses? What about self driving buses? What about an automated subway? It's just a box on a track, with the state of all the trains in the track known, self driving would almost be overkill for subways/trams/trains. It's still public transport. There's no need to hold off the development on both ends.
  • masklinn 8y ago
    Self-driving subways have been a thing for nearly 40 years.
• ameister14 8y ago
  Economically, not everyone can afford self-driving cars. In the near future, they appear to be luxury items. Trains move more people more efficiently and cheaply. That should remain the case even as driverless cars become popular, at least for another generation - maybe two.
• WhompingWindows 8y ago
  I'd wager you will continue to see a mix of transportation modes at first, then a proliferation of non-vehicular transport for shorter/medium range trips, with greatest acceleration of change occurring in the highest person-density areas like NYC, LA, Chicago, and other big cities with their eyes on ending the scourges of public parking, local emissions, traffic accidents, and just generally the unsightliness of a car-centric world. While the ride-hailing self-driving cars will be a great boon to those not wanting to drive, saving parking and potentially preventing accidents, I do believe SDC's will come in waves starting from the SW USA and California before spreading to the rest of the country/world. In these areas, public transit will likely be less useful, since for instance in Tucson, AZ, the city is massively spread out and busses would simply incur a much greater time-cost due to their non-personalized route. Moving forward, as SDC's become more % of the fleet in urban/suburban areas (still won't be in rural areas for another 5-10 years or more), you will probably see more variety of SDC's. Instead of having a 50 person bus, you can have a 5-7 person SD shuttle/minivan (think Waymo Chrsyler Pacifica) that will take 5 people with similar routes to the same place. Once SDC is highly penetrant in terms of %, I'd expect a great prolfieration of electric scooters, electric pedal-assist bikes, regular bikes, pedestrians, and all sorts of non-vehicular transport. I myself would ride a bike to work if not for the very low-skill drivers I encounter daily in my commute. This happens to be in an urban area with very poor bike infrastructure and very large hills. SDCs + electric bikes would solve these two issues, IMO, and they are probably ~10-15 years out in my location.
  • dantheman 8y ago
    I actually think self driving technology will be most cost effective in rural areas - you can't run bus routes because there is no common traffic patterns and the there isn't enough demand. However, self driving on demand transit will be able to pick people up and take them where they need to go. You may have to wail till a vehicle is free, but that's better than super infrequent bus routes and messy connections.
• tonyedgecombe 8y ago
  Didn't the head of Waymo say we are still 30 years away from generally available autonomous cars. I think it would be a shame to put off important projects on the basis of something that may never happen. I'm sure it would keep the Koch brothers happy though.
  • mcguire 8y ago
    https://en.wikipedia.org/wiki/Vaporware?wprov=sfla1 https://en.wikipedia.org/wiki/Vaporware?wprov=sfla1 "Vaporware is often announced months or years before its purported release, with few details about its development being released. Developers have been accused of intentionally promoting vaporware to keep customers from switching to competing products that offer more features."
• Fricken 8y ago
  As excited as I am for robotaxis, nothing beats the raw people moving capacity of trains. The places where robotaxis offer the greatest benefit is in providing first and last mile services getting people to the train station and back in sprawling suburban areas where adequate transit service isn't feasible. You'd need to have some very smart, and tough public policy in place, though, to constrain robotaxis to instances where they provide maximal societal benefit. It will more likely be a free for all that makes traffic worse, not better, because that's just the way these things tend to go.
• 0x4f3759df 8y ago
  A self driving car takes up as much space as a normal car. People tend to know how much the new lightrail is going to cost but don't know how much it costs to turn a two lane road into a three lane road, or the general costs of the automobile infrastructure. When I went digging around I found each traffic light pole unit cost $150k, so a normal intersection is $600k