Given the lively discussion we had the last time I posted about automated vehicles, I thought it a worthy topic to continue.
Over on CityLab is a fascinating article by Judith Donath about how transportation could end up being subsidized by people who want to move you to a given spot. The idea is that, like Facebook, we become the product and the automated vehicles become the method. So where we go could be dependent on who pays for the ride. That is, you may ask for a ride that takes you to Fenway Park, but if it’s paid for by a real estate agent, it may first drive you slowly past a home that’s for sale. Or, the free ride may only be available to those who pre-qualify for a mortgage.
This paragraph makes this future seem downright dystopian:
Autonomous cars will be part of an ecosystem of intelligent agents and personal-data vendors. The information they are able to base your route on—and how they present an itinerary to you—will not be limited to where you say you want to go, but on all the data they have about you. Note that companies with immense personal-data collections, including Amazon, Baidu, Google, and Uber, are in the race to develop autonomous cars. (Uber has recently launched UberEats, a delivery service that collects data about customer habits for participating restaurants.) The same system that one day provides your ride may have access to, if not control over, your calendar, contacts, medical records, and holiday shopping lists.
The vision here isn’t all that far-fetched given how some of our most popular Internet sites operate today, in that we get things for “free” and often don’t fully understand the true costs. Just think about how many sites exist not just to attract us, but to collect and sell our information.
I still cling to the belief that the streets belong to us, the citizens of the city, and it’s up to us to make sure they’re used appropriately. For argument’s sake, let’s assume this version of the future is true. Are we comfortable with advertisers using the streets without paying for them? How would we want the city compensated? Will consumers tolerate municipal intervention or will “free” prevail?
Interesting (if spooky) ideas but I’m more than a little dubious. I don’t doubt that some small bits and pieces of that scenario might come to pass but I doubt that it will happen on a very large scale.
The on-line world of today thrives on “free” stuff paid for by advertisers because the incremental cost of reaching each of those potential consumers is infinitesimally small – i.e. one more set of eyeball on my ad is nearly free. Facebook, Google, etc are getting rich by aggregating those very cheap eyeballs by the millions.
If someone wanted to subsidize my trip downtown in an autonomous car, they’d pay dearly for it. The hard costs of that trip (use of a very expensive vehicle for say half hour) and the associated misc costs (fuel, network, etc) would be thousands of times ($10, $15?) of reaching me via today’s advertising.
It sounds like a very specialized niche rather than generally free transport.
The future might bring anything, but we do have the present to help us think about it.
If it is such a slam dunk good idea, what’s keeping advertisers from offering discounted or free Uber or Lyft rides today? Yes, the hourly costs of autonomous vehicles will be less without those pesky drivers, but the costs aren’t so high now that multiple advertisers couldn’t start the experiment today. There aren’t even ads inside private ride sharing vehicles, which would be low hanging fruit.
If such a world did come into existence, it would be by choice. It isn’t like the majority of people in the US outside of NYC view conventional private car ownership as a crazy luxury. Any new model of transit would have to offer value without being too weird.
And costs to the city? Drivers don’t pay their way now. If ride sharing started to use more than a minute number of road miles, taxing them would be far simpler politically than increasing the gas tax (which itself will be replaced as the private fleet electrifies).
I actually know Judith, the author. We were students together at the Media Lab. Smart person. Not omniscient.
first we need the city to approve autonomous car testing on our roads, we should welcome any autonomous startup to HQ in Newton.
I can easily envision the local Mall offering Free roundtrip rides in shared autonomous vehicles. Companies will also offer it as a perk to many employees. One thing we know about technology is that it often gets used in creative ways that the original inventors had not conceived.
It’s a brave new world and we are getting closer to and moving faster toward the age of The Jetsons than most people are aware of.
The collision avoidance and AI managed traffic systems being developed for ground based autonomous vehicles will also be applied to flying cars – the technology allows for (fairly) safe and efficiently managed, heavily trafficked lanes in the sky as well as on the ground. In the air, travel lanes can be stacked on top of one another, instead of just spread out horizontally. This is the solution that will help alleviate or almost eliminate most ground based traffic jams.
Not all of us will live to see this world, but it is almost a certainty that it happens within two generations. Of course humanity could also destroy itself first.
On my way home on the Mass Pike in my 2017 Ford Fusion Hybrid ( it has a lot of driver assist features ) an alert popped up on my dashboard: “Front crash avoidance disabled: Sensor blocked.” Probably some snow/ice had melted while I was in the office, and was now frozen over the sensor. The LIDAR that provides the car with blind spot warning often warns me it’s disabled during a strong rain. One of the few groups of people that seem to be skeptical of driverless cars are software engineers like myself. I have nearly 30 years of software development, the last 15 or so working on FDA regulated medical devices; advanced image processing, computer vision, assisted surgery, patient positioning robotics, and the like. Will there ever be an autonomous car that can navigate Boston area streets? I doubt it. The problem is extremely complex and given how many times a week I hear someone tell me “ something’s wrong with my computer- it’s slow/crashing/“acting weird” – I’m amazed at how many people think an autonomous car is just around the corner. I’d go out on a limb and say there won’t be one for 50 years, and perhaps never.
https://spectrum.ieee.org/transportation/self-driving/the-big-problem-with-selfdriving-cars-is-people
I get especially worked up about the prediction that they will be safer. There is no real world evidence to back up this claim. Let me restate that; a fully autonomous car does not yet exist, therefore there is no data about their relative safety.
I am reminded when I hear this of Physicist Richard Feynman’s report on the Challenger shuttle. He threw cold water on management optimism. The same is needed here.
https://history.nasa.gov/rogersrep/v2appf.htm
Frank,
As the percentage of cars that can ‘talk to each other increase, autonomous vehicles will become 100% safe. However, I think that is 3 decades away
And here I thought this was going to be the more strIghtforward subsidised transportation Japan has. A company (say, Seibu) builds a small shopping complex and train station in the Tokyo city center (Seibu-Shunjuku station) and then buys chap land 30+ km out and builds a huge shopping and spots complex (Home of the Seibu Lions!) connects them with a rail line, and then makes even more money off of long-term ground leases near station and /or building the housing themselves.
Even with all this, no company wants to take a loss on the train itself, so – like all trains in Tokyo – they charge by distance. 10km takes 20 minutes (on the express) and costs about $2. All trains are packed from roughly 5:30 to 9, morning and night, every weekday.
Packed trains are much cheaper to operate than single passenger vehicles, and the benefit of at-cost or subsidised transport to the company is clear. I’m not clear how the dollars work out here.
Anne,
Interesting perspective. Similar thing happened in Newton. Norumbega Amusement Park in Auburndale was built by the Middlesex and Boston Street Railway to increase ridership on their streetcar line. The line opened in 1897 and closed in 1930, marking the end of streetcar service by the company. The streetcar line ran the length of Commonwealth Ave in the median from Boston College to the Charles River.
The M&B converted to all bus service in 1930. They were taken over by the MBTA and the bus routes were eventually renumbered with a “5” prefix. The 535 bus, which ran the old streetcar route from Boston College to Auburndale, was abandoned by the MBTA and not used after 1976.
“According to Wikipedia.”
Rick,
Agreed that there are no self-driving private vehicles, but Google’s Waymo has just recently started deploying test vehicles that have no drivers and are essentially fully autonomous:
https://www.theverge.com/2017/11/7/16615290/waymo-self-driving-safety-driver-chandler-autonomous
They and other manufacturers are testing on roads, weather, and hostile drivers more challenging than in Arizona.
Mike,
Good to know. And I suspect that Wikipedia is probably more accurate in the particulars than my memories from college (I minored in East Asian Studies) about how private rail works in Japan.
You suggest a more community oriented approach, Anne.
The niche for an ad-driven transportation dystopia can only exist if public transportation fails to evolve. There is no good reason that twenty years from now, we couldn’t have replaced all fixed-route fixed-schedule MBTA buses with on demand service (and with door to door convenience for at least some people). The vehicles will be autonomous, electric, and range in size depending on demand.
Running buses the way we do now is very inefficient off peak and often unpredictable at peak. Before it failed, Bridj began on demand service experiments partnering with transit in Kansas City. As an experiment, most in the industry considered it a success in spite of the fact it didn’t draw in too many riders.
A key challenge, in general, is that transportation itself is not a lucrative business. Frankly, it rarely makes a profit. Uber and Lyft, the two ride-hailing leaders, have yet to turn a profit meaning each ride is effectively subsidized by VCs. Worse, they’re clogging roads. In New York City the average speed in the urban core is now 6mph during business hours. (https://www.citylab.com/transportation/2017/12/how-to-fix-new-york-citys-unsustainable-traffic-woes/548798/).
AVs may fix some of this, but they’re not going to fix it all.
When Bridj was ascendent I sat in a meeting with the City of Newton and Bridj execs. The most eye-opening moment for me came when one of the city employees asked a simple question: can this help the elderly and disabled?
A public transportation system needs to work for everyone, regardless of their economic status or personal need. A private system is optimized to reach the people that offer the highest revenue. If we let the private systems run amok, we risk leaving a lot of people behind.
My first full-time job in Boston was working the overnight shift for a downtown-based TV station. I lived in Sharon and would take the Commuter Rail to work. The challenge was that the rail was optimized for traditional commute hours. I worked off hours. Working an hour late cost me 4 hours in commute time. I’m not alone in that, if you’re a shift worker at a restaurant and need to rely on the bus, you need to time your rides around your shift. If those don’t line up, you are just left wasting time. Now, imagine you’re paying for daycare for a child, what does that hour or two of extra time mean to your income? What about the commute itself?
Chuck, an on-demand autonomous vehicle based public transportation system would meet the needs you describe in ways never possible before. Labor costs are one thing that makes late night service to or from the suburbs prohibitively expensive.
The Ride service would also benefit from such a system. Even for those riders who need assistance, there’s no reason the person providing that assistance also has to be a professional driver.
We just have to agree that the service is important as part of the Commonwealth, taking profit out of the equation. And plenty will argue for “let the private sector handle it.”
Bridj was never in a position to offer such a service. Even their best shot fell short.
@Mike I think we agree on most things here. In general, AVs will change things for the better. I just don’t want to leave it entirely to the whims of the private sector, especially when we own the roads.