1. Mom, dad, and kid all need to go to work/school at roughly the same time. If they all have a 15-20 minute commute (well below average for many areas) and the destinations for all three aren't in the same immediate vicinity, we're right back to the family needing 2-3 cars.
2. Eh, no. Go to the mall or a shopping center parking lot and walk around looking in people's cars. The average person's car is somewhere between messy and a pigsty. IMO, this will only get worse with SDCs, since people will be able to do more eating/dressing/preparing/relaxing in their car without the need to drive. Not to mention the number of people who simply don't want random strangers using their car. Uber or Lyft would have to offer a significant fee before I would even consider it.
3. Vehicle owners are used to convenience. If I live in 20 minutes outside the city, then I probably have to wait 20-30 minutes for a ride to arrive before I even start my trip. Massively inconvenient, and not worth it.
SDCs will definitely lead to change in American society, but I have to wonder if it won't completely backfire on some of these expectations, and actually lead to American society sprawling out even more into the suburban and rural areas. WFH is becoming a viable option for more and more jobs, and longer commutes will be much more acceptable to most people when the time in their car isn't simply lost time.
That's okay, once automation really gets going, Mom & Dad will be out of work, solving that problem.
This bleeds over into decisions that affects other aspects such as telecommuting, paternity leave, etc.
Or it will become a poverty trap.
Not according to studies I've seen, including the one cited in my original comment [1]. They map commute patterns, urban, suburban and rural, and take simultaneity and fleet access into account.
In Ann Arbor, Michigan a "shared fleet [was predicted to] provide almost instantaneous access [fractions of one minute] to a vehicle with a fleet of only 15% of the number of privately owned vehicles that would have been used for these trips." In a rural study conducted in Germany, maximum wait was estimated at 5 minutes with fleet capacity at around 20%.
[1] http://sustainablemobility.ei.columbia.edu/files/2012/12/Tra...
It only accounts for trips internal to the city which already have low wait times with existing taxi cabs, and does not simulate the waits for trips traversing into/out of the city.
Give people an economic incentive for keeping a neat car, then you're going to see neat cars. This is clearly true for the current Uber and Lyft fleets.
A friend with an incredibly smart 7 year old still has a messy car. Her daughter, despite the ability to read and understand things well above her age, is a 7 year old. The car is a mess.
When it comes to these pie in the sky utopian scenarios, I wish people on HN would step outside and actual observe every day people.
Right, and that's a choice. It has nothing to do with "she's a 7 year old" and everything to do with "she's allowed to make a mess in the car".
I was never allowed to eat in the car beyond the odd sweet, and I wouldn't have thrown wrappers or anything else on the floor on pain of death... I find it mind boggling that the child is blamed for the state of the car, when realistically it is just a parenting choice. There are 7 year olds with exceptional table manners even in this modern day, and there are 7 year olds I'm embarrassed to eat at the same table as.
I'm fine with people choosing to have messy cars (mine oscillates between spotless and a dump over ~3 month periods) - but it's a choice and other choices can be made.
Maybe it is "choice", but there are a lot of people out there where the "choice" around their vehicle is closer to that of it is lived in, may have books and papers, clothing, toys, etc. (maybe food as well) that will not conform to this scenario.
Walk through a grocery store parking lot sometime and observe the state of vehicles as people get in/out, etc. my bet is on a sufficiently large pool would be in the condition where utopian sharing doesn't make sense.
That said, if there was a mind shift because of an overall benefit, it could happen. But, that is a long way off.
Even without the wage-cost of drivers, individual taxi service may still be too expensive to match the lower demand.
No one likes buses, because buses are slow. They're slow because they make lots of stops that are irrelevant to your trip. As someone that regularly rode the school bus on a route that took an hour, compared to a 10 minute direct route. Fuck buses.
Yes taxis, exist, but not many. The only people I knew that ever took them were people that lost their licenses.
What you've actually described is a shuttle service. They exist everywhere. They're not popular because they're slow. Last time I took a shared shuttle from the airport it took an additional 45 minutes to get home than driving direct. They have more convenient pick up schedules and locations, but the transit times are almost as bad.
The cycle of poverty is strong in these rural locations because of the lack of public transportation and the cost to maintain a vehicle to hold a job. There are plenty of low-income people moving to common destinations (city center, plants, etc.) that would use an on-demand shared shuttle service if it had at-door pickup.
Having worked a couple summers on a factory line, we would set up carpools and pick people up along the way to work each day. Don't be a dismissive dick just because you personally didn't like your school bus ride.
Because? Really? Because someone has to buy a car? I think your feelings are in the way of your cognitive ability.
>of low-income people moving to common destinations I'm sure there are plenty of low-income people moving to the suburbs/rural areas because they are being priced out of high density centers.
Lets not forget, 5% of the population uses the bus, that's including major urban centers like NYC. Nationally, nobody uses the bus, its a joke throw back to a time before cars. Fuck buses.
I rarely bother to even use their (world class) MRT I love the buses so much.
An advanced city is not a place where the poor move about in cars, rather it’s where even the rich use public transportation
Also, I would add that subways are probably a better way forward than busses for a modern city, or at least some sort of connected underground transport system.
https://en.wikipedia.org/wiki/Driving_in_Singapore
http://www.bloomberg.com/news/articles/2012-06-04/singapore-...
https://www.google.com/search?q=singapore+size
I also agree that Subways are awesome - and Singapore has an awesome system for such a young country, and they are going to be adding a new one each year through 2024 - That's a new line, not just a new station. [1]
The nice thing about bus stops - is they are everywhere, and the buses come by very frequently, and they all have GPS locators on them - so you can time your arrival to the stop to be when the bus arrives (which you can look up on your smart phone). I've never been a fan of buses before Singapore - but it's great to see what can be done in a country where cars are not the default option.
[1] http://www.lta.gov.sg/apps/news/page.aspx?c=2&id=80d3169f-dc...
55% of commuters in New York city use public transportation, 34% in SF, 26% in Chicago, 40% in D.C., etc. Don't be a dick about cognitive ability when you're quoting a 5% number that includes people who don't even have access to buses.
Smaller buses able to optimise routes to make fewer stops will help. Split the current big bus into 3-4 smaller ones and the people going close to the full distance might not stop for the first 80% of stops. That would cut out a lot of time.
Obviously rural areas will be the last to see these changes for non-commercial traffic.
On the other hand, I've used airport busses between transit hubs in cities where everyone takes the subway and it was really quite nice. Better if I ever figure out luggage delivery, of course.
> 1. Mom, dad, and kid all need to go to work/school at roughly the same time.
Start times (school and work) are just arbitrary numbers. We changed our school start time to help with parents dropping off kids to multiple schools (primary and secondary schools) and then on to work (9:30 start is unusual but fine)
#1, work/school from home, optimized schedules, etc
#2, you could get in/off the "car" in your own little capsule...
#3, those fleets of cars would park at the most efficient spots possible by using big data so the wait times could be reduced significantly (or the wait could be eliminated completely if the schedule is known ahead).
2. User feedback could be a useful deterrent against messing up someone else's SDC, in the same way that both drivers and passengers can leave feedback with Uber today.
3. True, but one can adapt to this.
Or maybe i'm just hopelessly optimistic... :)
My conclusion is that car sharing services are quite expensive. I can do no more than 3 or 4 short rentals a month before the cost start to resemble that of owning a used car. I do save money, but only because I drive a tiny fraction of what I drove when I did own a car.
Of course, car sharing is not as efficient as SDCs that can move around and maximize usage, but even SDCs would mostly end up parked at night and off-peak hours. Coupled with your point #4, I'm afraid it won't be long before driving the current average yearly mileage using those services become massively, prohibitively expensive.
I'm not sure how that would be a gain for anyone but the environment.
If I was using an on-call car service that I could ping 10 minutes before I'm ready to go each day, I can do that without triggering my concerns.
*Note however: I currently commute by bus since I moved to Seattle, and I make up for hating being ready by an exact time by instead choosing a bus route that has a bus every 15 minutes.
For 10x, you would absolutely have to do something to address peak time behavior. For example, if Mom, Dad, and Junior all have to be somewhere at the same time, you are still gonna need 3 vehicles. In aggregate, total fleet demand will be dictated by the marginal demand at the peak of rush hours.
EDIT: to add, I would still predict a massive move of car ownership toward corporations, and their primary domicile to large parking lots in outer suburbs, only to awake during rush hours. These cars will have a much lower return on investment, but probably enough to justify their costs.
The better pooling effects of subscription or on-demand fleets mitigate this problem in most scenarios, but it doesn't mitigate it during peak times. Peak vehicle utilization will be what drives total vehicle fleet size.
As it turns out, coordinating rides is pretty difficult, even with sophisticated software doing a lot of the work for you. Maybe if either of those companies had huge market shares, then maybe the problem gets easier. I'm not counting on it.
Labour costs are a significant part of any such operation, where I live at least.
Many, many people already do this on public transport. What we're talking about is general door-to-door public transport.
Car-pooling is awkward to arrange, particularly for short joined parts of the journey (e.g. picking someone up part way and dropping them off a little later), and it's extremely hard to arrange anything that overlaps such that the first passenger needs to get out before the last passenger (currently would involve handing over the keys to your car).
Total fantasy. Virtually nobody uses public transport. 5% of commuters use public transport. Relative to the car-pooling, who sees double the volume of users as public transport, 10%.
The question is: if that changes, what might people choose to do? The example of large cities like NYC at least teaches us that the American preference for solo car travel is not fixed.
[0] http://ns.umich.edu/new/releases/21923-hitchin-a-ride-fewer-...
http://www.ons.gov.uk/ons/rel/census/2011-census-analysis/di...
http://www.ons.gov.uk/ons/rel/census/2011-census-analysis/di...
These are not insignificant figures.
If you say "people don't want to do X" when millions of people in my country already do a more extreme form of X, then I think you're wrong.
This will be much smoother if you just open an app like Uber and it tells you the next pickup time for where you want to head.
Those people might use a vehicle subscription, because the cost of private rides would still be lower than the costs of private vehicle ownership (depreciation, maintenance, insurance). Many people would pay for the privilege of a private ride (like they do with Uber/Lyft now).
If car subscriptions services succeed and are nearly ubiquitous, the personal-freedom symbolism of car ownership will degrade.
Then, if people want to own a car as a status symbol, they will probably be better served by a more expensive one anyway, the same way people buy other status symbols (rose-gold iPhones, etc). Ironically, most of those status-symbol cars will still have an autonomous mode.
Some people pay more to use a first-class compartment.
Shared-car design could also be drastically different. I don't want to share a Honda Accord with random people, but I wouldn't mind sharing a road-train that had private compartments for each set of passengers...
car: high flexibility, high cost
public transport: low flexibility (last mile, connections, fixed time schedule), lower cost
self-driving vans: medium flexibility, possibly low cost.
10x is admittedly bullish. I expect 4 to 7x within 20 years, i.e. 3 to 4 product cycles. But 10x is more reasonable than it seems at face value.
Let's use Manhattan to illustrate. Another HN user (wadenick) kindly provided [1] some references I'd like to point to. Car ownership across the United States stands at 81% of households [2]. In Manhattan it's 23% [3]. Manhattan has roughly fewer than 1/3 the number of cars as the rest of America. This is partly why there are only 39 gas stations in Manhattan (down from 60 in 2004) [4]. Repair shops, too, are fewer and further in between. Many are owned by taxi operators for their fleets' use only.
Uber pays 75% of its earnings to its drivers. How much cheaper could a driverless be? 50% less? How would that tilt our 23% figure? All it takes is half of Manhattan's car-owning households to switch for Manhattan to hit a 7-fold reduction in car ownership from present-day America.
Everywhere isn't Manhattan. Density matters. Less-dense locales will need to accept some combination of higher wait times, higher prices or lower rates of fleet reduction. That said, the scaling economics are nearly identical. Fleets will be cheaper than individually-owned self-driving cars because they can more efficiently use their down time. This, in turn, makes individually car ownership more expensive (by degrading retail-servicing infrastructure and supply chains). That, in turn, makes fleet usage more desirable and so on.
The degree to which this happens may vary. But at its lower bound, I'd posit 4x and at its higher bound over 10. Given that income and urban dwelling are correlated, and continue to increase in their correlation, that bodes badly for the edge-case rural families demanding high degrees of simultaneity at a price even remotely approaching what it costs to maintain a small garage of cars today.
[1] https://news.ycombinator.com/item?id=10837732
[2] http://web.archive.org/web/20140209114811/http://data.worldb...
[3] http://www.nycedc.com/blog-entry/new-yorkers-and-cars
[4] http://www.wnyc.org/story/say-goodbye-manhattans-gas-station...
Remember that from that 75% is supposed to be things like insurance policies and such, and I suspect that currently drivers aren't bothering to properly insure their cars (using it for a business rather than personal use). In a driverless car case, someone is going to have to pay for that insurance, whether the manufacturer, operator or rider, that cost is going to come from someplace, and will probably be significant. I'm not sure that you could achieve 50% savings, and maybe not even 25%.
"All it takes is half of Manhattan's car-owning households to switch for Manhattan to hit a 7-fold reduction in car ownership from present-day America." But what does that prove? Manhattan isn't a great barometer of anything. Even if they did drop to 14% or so, that's not really moving the needle much either way, is it?
The fleet provider (e.g. Uber) would surely self-insure. Insurance for self driving cars will be extraordinarily cheap (at least an order of magnitude less than driver based insurance).
It's taken that long just to see ABS become standard equipment across 90% of vehicles.
And this just completely ignores the ~5 to 7 year product cycle. There's no way that in just one or two generations of the Taurus we have a completely self-driving car capable of operating in all normal driving conditions, and that it's displaced the 4 or 5 previous generations of vehicles commonly on the road without the tech.
Maybe it will be 90% for <30 year-olds?
1 - I think the tech will be there.
2 - Once the tech is there, there will be incredibly fast adoption for a simple reason: price.
First, insurance rates. Right now, you get discounts for eg the sensors in your volvo that warn about vehicles next to you if you turn your blinkers on. Autonomous driving will be like that but on steroids. Insurers will rapidly price in cost falls from autonomous driving.
A second huge cost increase to non-autonomous driving will come from juries. Right now, as a society, we accept car accidents as part of the cost of a modern society because there is no alternative to heavy use of vehicles. Humans driving cars means you will inevitably have accidents. Even if you kill someone in an accident, as long as there isn't negligence on your part, you generally will have nothing happen to you besides increased insurance costs. The second there is an alternative to a human driving, I expect civil juries to start returning large awards and laws to change to include jail sentences. Not to mention coming down hard on things like drunk or impaired driving. eg in WI you can basically get a couple DUIs before the state gets serious about penalties. That will change in a hurry.
Before we didn't have tar roads with road signs and traffic lights, etc. It would be infinitely more feasible to make roads more clever than trying to cram human-like driving intelligence into car AI.
Then... we wait. How long until Idaho has instrumented cities?
What about construction areas? What about parades? Even with Vehicle to infrastructure communication you will still need AI.
ABS is a sideshow, not a revolution. As soon as self-driven vehicles are market-ready, we'll see a massive car park upgrade within one car-generation, which I guess is 7 years. Droves of people will ditch their old cars just for the purpose of upgrading.
That's glossing over the Nokia video phone, the Treo, and all the Windows handsets.
> ABS is a sideshow, not a revolution.
I have no idea what you're trying to say.
> Droves of people will ditch their old cars just for the purpose of upgrading.
That seems like something that would require evidence. Or at least a theory to back it up. My 10 mile round-trip is definitely not so onerous that I'd look to replace my vehicle for something that isn't going to make it any shorter or more pleasant.
Here's some context: Nissan has been working on the Titan XD for longer than that. That's just a pretty basic 3/4-ton truck without a lot of bells and whistles with an outsourced engine.
Tesla's Model 3 will have been in development almost that long by the time it's released, and as far as anyone knows, has no new tech.
You're talking about tech that doesn't even exist yet.
But yeah, as someone who grew up driving old boats without it, I would never buy a car without ABS.
Smartphones cannot kill multiple people on a highway if the software malfunctions. What happens if I am going 75mph and my self driving car crashes and has to restart? What if one the sensors malfunctions and thinks it needs to swerve to avoid someone merging into my lane?
With smartphones you arent putting the lives of whole families in your hand.
How can you even compare those things?!?
A car easily costs 100 times what a phone does, and is literally being trusted with your life and the life of people around you.
> Droves of people will ditch their old cars just for the purpose of upgrading.
Back in the real world, a car is a major purchase and the majority of people couldn't afford to drop their current car and upgrade even if they wanted to.
Also, if anything, it might be more important to incentivize people to switch to electric cars more so than self-driving ones, given the ecological impact. Anyone up for a 500% tax on gas by 2025? ;)
Kids will one day laugh at how we used to buy vehicles instead of rent them by the miles. "Auto loans?!?" The horror!
The future is not one-size-fits-all. Just because Uber execs and investors are blathering about how personal car ownership will go away (meaning they will own everything, how convenient) doesn't mean it actually will happen that way.
Think how much consumer excess would occur from 300 million people not having to buy a vehicle outright.
Those costs don't go away just because you're renting. In the same way property taxes don't disappear just because you aren't paying them directly in a leased apartment.
> Free street parking will gradually disappear and off-street parking is not going to get any cheaper.
This applies to maybe 10% of the population. And I feel like that's pretty generous. You probably haven't seen much of the US outside of the North East if you think that's even remotely plausible.
Most assuredly and it can't happen soon enough. We need to run headlong into automation, eliminating as many jobs as we can. The sooner we eliminate industries, the sooner we enter the post-labor era.
Not to be trite, but one should never underestimate the impact of lobbying from massive established corporations.
I'll even go as far as predicting that this technological advancement will be a personal (or family) aerial vehicle[1] that makes commuting as easy as vehicular travel, but perhaps safer, and definitely faster.
In other words, I think the automobile will diminish in quantities, much like the horse, not because people need them less, but because something better comes along.
1. This could be a gyrocopter-like device or a drone-like device, but will likely not be a fixed-wing "flying car", especially beyond the extent to whicih a V-22 Osprey is a fixed-wing.
Everyone can hear you, see you, and most importantly, see your body language. Go a little further with hardware power and have the computer place clothes on you. Wear a ratty tshirt in your home but wear a nice Armani suit in the meeting. Some of this tech is here already but just nowhere near as fast as it would need to be to achieve perfect, virtual telepresence.
No more garages and driveways. Neighbourhoods will look a lot nicer, and maybe this will foster community engagement? House design will change, maybe with more multi-family units.
Much narrower roads, with real estate that's now sacrificed to roadway given over to parkland or housing.
Maybe cars will move partially or fully underground, travelling in large-diameter pipes. Like a sewer. Save the surface for people and bikes.
Maybe I can own my own "car interior", a custom box that I can commute in every day. It won't have wheels or an engine, so when I want to go somewhere I'll summon an auto-auto that picks up my auto-box and ferries it around.
We conflate two separate things when we consider what cars _are_ - a way to transport ourselves and things from place to place AND an extension of our home. The idea of home extension will be more difficult to break than the idea that you have to drive everywhere.
There will not be a predominance of consumer owned self driving cars in my lifetime, despite what Tesla, etc. think. Just like how most rail transit in the US makes little sense due to lack of density, the autonomous car won't appeal to everyone due to lack of use, want of control, etc. As such, mechanics will be around for a long while.
People buy vehicles like Toyota Tacomas and will own them for 20+ years. I see that not changing with the 2016s, despite all the announcements of every major automotive company working on autonomous cars.
Maybe in major cities in the US (maybe) and maybe in parts of Europe.
But I don't see it. It's about as farcical as door to door delivery by drone with Amazon. Looks great in theory, practice (and practicality) not really once you dig into it.
This makes me think of an interesting side effect.
Will there be pressure on automotive engineering teams to significantly increase drivetrain half-life in consumer vehicles due to automated-driving?
I imagine there are trade-offs made at present that limit the drivetrain half-life simply because there isn't a great deal of pressure to increase the life of a consumer car beyond 300,000 miles ( 300,000 mi * 1 yr / ~10,000 mi = 30 yr old cars).
Semi-trucks may be a nice example where a similar engineering pressure currently exists. A mechanic friend once told me semi's can drive a million+ miles.
If you are running a fleet of cars or ride sharing yours all day when you are at work, its going to get 100k miles put on it a year. Expect a new battery ($5k-$10k) annual replacement, in addition to the other 80% of the car that isn't hybrid like suspension, electrical system, interior and body.
The fuel savings often goes into battery maint/replacement, if you plan on keeping the car for extended mileage/time.
A few times that a customer won't be happy with "oil change needed" indicator lighting up, threading on one of the tires, some random food wrappers scattered on the floor, kid's soccer equipment forgotten in the trunk or just lack of recent car wash, and the transportation provider's rating system would squeeze this family out of network.
Car ownership is 45% in NYC. Not sure why this is considered "scarce" :)
How much cheaper would a driverless Lyft or taxi be? 50%? How many more people would forego driving, in Manhattan, if ridesharing were half as expensive as it is today? If half do we're at 11.5% car ownership, a 7-fold reduction from the present-day U.S. average. If 2/3 do, we get to 10-fold.
I'm not certain that other elements of your vision work in non-metro areas.
[1] http://www.reuters.com/article/usa-cities-population-idUSL2E...
People in NYC don't own cars because public transit options are convenient, extensive (at least in Manhattan), and inexpensive.
While Uber/Lyft/etc are convenient and flexible, an unlimited MTA pass is $117/mo. It's hard to imagine Uber's cost dropping so dramatically so as to compete with that.
But maybe you'd argue that they don't need to, because Uber is replacing taxis [1]. I do think the discounted car sharing offerings like uberPOOL and Lyft Line could bring competition to public transit. It'll be interesting to see where their pricing settles in the long run and if/when either experiments with an unlimited rides offering.
1: http://fivethirtyeight.com/features/uber-is-taking-millions-...
Desktop processing apps were supposed to make office workers so productive they would only work a 4 hour work week. The reality? Workers are now so connected that they end up working 6-7 days a week.
If I can rent out my car when I'm not using it, my car is cheaper for me to operate, and more cars are available to drive people around, so they make more for hire trips when they otherwise would not have. More cars, more rides all around.
It makes no sense to own and maintain your own FLEET of self-driving cars for your own family (since it's virtually guaranteed that multiple family members will need a car at the same time, at some point in time).
In the suburban sprawl where the majority of the USA lives, we already have a nicely distributed network of storage (homes with attached garages). That would be far more efficient than having Uber and Lyft build their own duplicate storage for cars, which would almost certainly put them further away from the customers who need them and be far less efficient.
I know this is HN and we're all supposed to be just super ga-ga over the idea of one single company owning the entire future of transportation, but there are many more possible scenarios than just "everyone gives up control and ownership of how they get around".
Parsing this took me a moment - what they already do is subscribe to a fleet, and for years it has been the taxi fleet. That's shifting shape and I guess we'd both predict it will be Uber- and Tesla-like in the future (I don't know what mix of public / private ownership might look like).
You're talking like that isn't something already
> AND FINALLY, as distributed car ownership falls, retail locations for their maintenance go out of business
The number of these doesn't matter as long as they're within reasonable driving distance.
You sir are well qualified to work on Wall Street.
Insurance. As soon as human drivers are provably poorer than automated drivers, the cost to insure a meat driver goes up. It will not take long to build a very strong case for the lower risk profile once the early adopter vehicles take the road.
FIRST, instead three cars for mom, dad and kid, the average family has one car that drops mom off to work, then dad, and then the teen off to school.
SECOND, aforementioned car makes use of its down time to drive strangers around via Uber, Lyft, et cetera. The car's owners collect a fee from Uber, Lyft et al.
THIRD, economies of scale and changes in perceptions towards the utility versus cost of owning cars causes people to stop buying them. Instead, they subscribe for access to a fleet–such as Uber or Tesla's. (Note: people already do this in cities like New York, where car ownership is relatively scarce.)
AND FINALLY, as distributed car ownership falls, retail locations for their maintenance go out of business. Barring massive public subsidy for car ownership, this creates a feedback loop that increases the cost of individual car ownership until it becomes a luxury.
1 will happen later than we expect. 2 will happen sooner, 3 much later and 4 drastically sooner. In fact, I think it will ultimately be 4, not 3, that drives a rapid, self-perpetuating dominance of self-driving cars within our lifetime.
[1] http://sustainablemobility.ei.columbia.edu/files/2012/12/Tra...