“What incredible inconvenience are we trying to solve?”
40,000 deaths in the United States every year.
Hundreds of thousands of costly accidents
We can’t get great transportation even in the New York City area. We have 0 miles of high-speed rail, while China built 25,000
it’s great that you want to start building some mass transit, but it’ll take a century if we start now.
Let’s start by having trains take you to the airport directly…
In the meantime, getting self driving cars that are all electric will be a big win. Perhaps car ownership will drop.
I mean it only took China 15 years. But then again Mike Johnson is speaker so.. well, it's not going well.
IMO the main risk around self-driving cars is that it will perpetuate terrible city design. Hopefully the tide will turn towards sane urban policy soon.
China certainly didn't spend a century building theirs.
> 40,000 deaths in the United States every year. Hundreds of thousands of costly accidents
Somehow, after a decade of hype, it doesn’t seem like driverless cars are the most direct path to solving that problem.
> Let’s start by having trains take you to the airport directly
I feel like that exists in a lot of places? Not sure what you mean.
I think "a lot of places" is doing some heavy lifting. There are a few cities in the US that have a train that connects to the airport but not many.
Driverless cars are significantly safer than human drivers.
https://arxiv.org/ftp/arxiv/papers/2309/2309.01206.pdf https://arxiv.org/ftp/arxiv/papers/2309/2309.01206.pdf
China has 25k miles of bullet trains and China accounted for a third of the cars sold in 2022. A third. This is a country that limits imports and even limits the number of drivers in some cities. So it’s silly to suggest public transit can replace cars.
ETA: Replace all cars.
So as long as we will have cars, we might as well make them safer.
> getting self driving cars that are all electric will be a big win.
I will say that this is still far away and the goals are orthogonal. We have electric cars now, the technology is proven. But it will take a generation of cars to replace the existing ICE vehicles. If and when self-driving vehicles are available it will take a generation to replace existing ones.
> So it’s silly to suggest public transit can replace cars
Ew. This is so flawed.
What is china’s car ownership per capita? Trips taken per day?
https://en.wikipedia.org/wiki/List_of_countries_by_vehicles_per_capita https://en.wikipedia.org/wiki/List_of_countries_by_vehicles_...
China has 2 cars for every 10 people (#99 in the world), while the United States has 9 cars for every 10 people.
While China buys a lot of cars (duh, they have a population of over a billion), they would buy even more if they had the dismal public transit of the US.
Of course public transit replaces cars.
Another fun fact: you would need 15 lanes of highway to handle the same amount of passenger load as a two-track subway tunnel. Manhattan as we know it would be physically impossible to exist if public transit “couldn’t replace cars.”
It won’t take a century if there’s political will. China built a network larger than the rest of the world combined in less than 20 years. Yes, you heard that right, in 2007 there was only one high speed passenger rail service in the entire country.
I don’t like being a defeatist, but being anti-public transit, infrastructure investments, and inability to build public infrastructure is very well ingrained into American culture at this point. Countries like China and India call pull it off because government makes the decision for people. Would be a political suicide for any level of government to get up and say “we will take over the land as we see it useful and build useful shit”. And I don’t even want to talk how in North America, our easiest path would be to hire people who can build that stuff very well (engineers from the said countries), which also won’t sell well with the public.
Admittedly I live in Canada, and in my city we have a fairly functional subway and bus system. But nowhere near as good as Tokyo, Hong Kong, Taiwan and etc. It’s such a bummer every time I visit those countries and see how they’re ran, and then travel somewhere like California or Washington to see the opposite when it comes to transit. I know not everything is the best, but I also don’t know how we can fix this problem when majority of people are against the fix.
Oh well, sorry for the rant.
>Countries like China and India call pull it off because government makes the decision for people
India is a democracy, as is Japan where public transit is fantastic. Many European countries also have very good transit. The problem isn't the type of government, it's the people and culture political will. Americans just don't want public transit enough to build and live in cities where it would work well. Not driving everywhere in a car is unthinkable to most Americans.
I’ve ridden Chinese HSR trains…and I still needed to take a taxi to get to Beijing South and whatever was on the other end of my trip.
There are over 30 US cities with trains to the airport. Another half dozen on the way. https://en.wikipedia.org/wiki/List_of_airport_rail_link_systems https://en.wikipedia.org/wiki/List_of_airport_rail_link_syst...
US also has almost 350 miles of high speed rail, not 0. The Northeast is where it’s all concentrated with Acela being the fastest.
If you live in a major city in the States it’s often way easier and faster to take rail. In Seattle for instance you can beat a driver to the airport every time during traffic. Hell even the LAX station in Los Angeles of all places will be available end of 2024 and that’s the most notorious car centric city in this country. The subway in LA keeps seeing ridership growth. I don’t even rent a car when I visit anymore. Everywhere I want to go is serviced.
Lastly NYC is one of the most serviced by transit areas in the world? Are you living in an alternate universe?
Mass transit is the solution. It’s vastly cheaper to build, moves more people, and far safer. I’ve got a highway interchange by my house that’s taken a decade to finish while they’ve ran rail and built four stations in that time. That’s a perfectly fine pace given our slow building system with environmental review and the like.
[flagged]
That list doesn't include cities where the local train (subway or street rail) system connects to the airport, which seems odd. In my experience it's way easier getting to the into the city using local rail compared to dealing with parking, rental cars, traffic etc.
The only downside is almost none of these systems are in operation 24/7, with the exceptions being NYC and Chicago. If you have a very early or late arrival / departure, you just can't use them at all without waiting several hours at the airport. I wish more cities could add reduced frequency night service, something like maybe 1 train every hour or hour and a half.
As a somewhat cherry-picked example, if someone in the southern part of my city wants to take a train into the city center, the last good route with only one transfer leaves at 10 pm. At 10:40pm Google maps suggests a route with 3 transfers that takes almost twice as long (and includes the commuter rail which is a separate, more expensive ticket). The last possible option is at 11:40 and takes 3 times as long as the normal route. If we want people to use trains and subways, the service hours have to be greatly improved.
> Let’s start by having trains take you to the airport directly…
I thought the US had legalised those now, so they'll happen as new systems come online.
> In the meantime, getting self driving cars that are all electric will be a big win.
Why? They're still going to take up as much road space (maybe even more, since they'll be moving around empty half the time). They'll still be spewing particulate pollution around our city centres - maybe more since they'll be heavier. They'll be consuming vast amounts of rare earths that could deliver much more benefit if used for electric bikes.
Self-driving electric cars will deliver at best a marginal safety improvement and a marginal reduction in whole-life emissions. There are much better ways to improve cities, transport, and lives: road-use pricing, bicycle priority at junctions, better parking laws, building more transit...
> road-use pricing, bicycle priority at junctions, better parking laws, building more transit...
Excellent suggestions. Road-use pricing could easily be added as a rule to self-driving fleets. Self-driving cars could always yield to pedestrians and cyclists. There will be a glut of empty parking spaces when self-driving cars are common. Self-driving cars are transit.
> Self-driving cars could always yield to pedestrians and cyclists.
They could, but they will be unpopular with users if so.
> There will be a glut of empty parking spaces when self-driving cars are common.
The self-driving cars still have to go somewhere. If they're not parked then they're going to be taking up space on the roads instead, which might be worse.
> Self-driving cars are transit.
Not really. In terms of space use they're still cars. https://www.cyclestreets.net/location/21715/cyclestreets21715.jpg https://www.cyclestreets.net/location/21715/cyclestreets2171...
Yes, but those deaths could also be "solved" with mass transit.
Yeah, but let's be realistic. The US car dependency is something that to change would both require major reconstructions of multiple cities, and a cultural shift that is just not happening. It is easier to solve fully self driving cars, for every single weather and situation, then it is to convince Americans that maybe a train to their neighborhood is a good idea.
> Let’s start by having trains take you to the airport directly…
Why should trains be the only alternative to driverless cars? What is difficult about buying a few buses and start taking people to the airport, today?
They already exist. They suck.
Buses are useless if they get stuck in car traffic. They'd need dedicated lanes
Buses do take people to the airport, they are extremely inconvienent for anyone but a solo traveler with minimal bags that happen to live close to a bus line.
The only time I take a bus is when I need to take a transfer from the airport parking to the terminal, that part usually takes longer than my drive to the airport. I'd gladly pay another 10-20$ for a valet to park my car from the terminal to avoid those busses.
Nobody bought a Tesla in the last 5 years because “I want fewer highway fatalities”, they often bought it because they thought the overhyped self driving features would let them chill out on their commute.
People want self driving cars so bad because driving everywhere sucks (and sitting on public transit sucks too). Routine travel is mundane and boring.
You want to take a bite out of road deaths legislate a vehicle/phone interlink that disables everything but basic phone functions on smartphones while in the confines of a vehicle, then legislate center console infotainment displays out of existence. Miss me with the bullshit "AI drivers will usher in a new era of road safety", you can trap an AV with a fucking ring of salt on the road if you're clever about how it's arranged. :P
Would that or mandatory breathalyser to start the car do more?
> Miss me with the bullshit "AI drivers will usher in a new era of road safety", you can trap an AV with a fucking ring of salt on the road if you're clever about how it's arranged. :P
How many people die when you do that? An AV stopping cause it encounters a weird situation is failing safe, and it is the correct thing for it to do. And yeah, they normally stop if you maliciously attack them. But human drivers also stop if you blow out their tires, or hit the driver with a well aimed rock. Car safety is not judged on malicious attacks, we are not talking about autonomous tanks.
When they cause a pileup? Non-zero.
>How many people die when you do that?
Depending on how you do it and where, potentially everybody in the car.
> it’s great that you want to start building some mass transit, but it’ll take a century if we start now.
I don't buy this. Specially in cities. You have roads, buy some buses and hire and train good bus drivers. Put some poles as bus stops. Lots of countries have done it in less than a decade.
The problem is, unless the government forcefully does it, the people sue the hell out of the city. “Bus lane lawsuits nyc” google search will result in bunch of results. Also doesn’t help how it would be a political suicide for anyone to be pro-mass transit during local elections. Sometimes they even oppose bus routes to go through their neighbourhoods as well, becoming some sort of an inverted “tragedy of commons” issue.
I'll be astounded if we have autonomous vehicles in the snow in my lifetime.
More cars that suck at navigating weather is not the answer to traffic problems.
There's no evidence that self-driving cars would drive any better and plenty of evidence that they're in fact, worse. It's just wishful thinking.
Mass transit in the US is a political problem. Political problems in the US are unsolvable by design. So driverless cars it is. Meanwhile countries where political problems are not unsolvable by design (either because they are dictatorships or they don't have broken political systems) will move ahead with mass transit.
> “What incredible inconvenience are we trying to solve?”
> 40,000 deaths in the United States every year. Hundreds of thousands of costly accidents
If reducing traffic death would be a priority, we could reduce the speed limit (would also help with resource consumption and climate change). We could also have smart phones go into sleep when in motion. That's a trivial technical problem. A much bigger challenge would be to allow smart phone usage for passengers while in moving vehicles, but until that is solved, cry me a river. Using a smart phone in a moving vehicle isn't a god-given right.
While true, we kind of need those people to die every year.
More people living contributes to a greater labor pool, which reduces wages. It leaves a greater demand on housing, increasing costs and making so many peoples’ lives more difficult.
If you’re working class, an extra 40k deaths/year is making the world better for you.
No one really knows where the bottom of that number will be, or even what it will be with autonomy. Whenever people bring up this argument I feel like it’s the same as saying “there will be no human error in plane crashes with an autopilot” when we know that isn’t true.