Hacker News Reader

Comment by onion2k | Hacker News Reader

Driverless cars might not lead to a reduction in car use but they probably will lead to a reduction in the number of cars. Rather than people owning a car they'll request one via an app, it'll turn up and drive them where they want to be, and then it'll go off to the next customer. Car utilization will go up significantly. It'll be like all cars are taxis. That will probably mean more car journeys.

But...

The manufacturing of a car is responsible for a very significant amount of the emissions it produces over its life. If we can greatly reduce the number of cars then there'll be a huge reduction in emissions simply because we won't make as many of them. That's always going to be a massive win for the environment, even if each individual car is being used more.

View on HN

I'm not optimistic that a significant number of people who need car transportation will give up personal vehicles. You would lose a lot of things:

- Spontaneity: You need to wait for the car to show up

- Independence: I would rather be able to do it myself than hire/rent out for the task.

- Personal crap storage: Everyone keeps their Very Precious Crap in their car. For certain trips, it's beneficial to have a car nearby to store things temporarily at a destination.

- Self expression: In America, a huge number of people express their character through their cars. And not just through bumper stickers. What you drive says who you are. People aren't driving absurdly big pickup trucks because they're all contractors and landscapers.

You can argue the practically of these points but that's just like, your opinion man.

I think the only way to reduce cars and car problems is by making cars totally unnecessary for daily life: Denser living with local work or WFH work. Diverse zoning so you're walking around the corner for your cup of coffee rather than a mile away.

Urban life should rightfully be the most efficient and economical approach. The suburbs are subsidized all to hell and the common sense approach is unrecognizable to most people now..

> Denser living with local work or WFH work. Diverse zoning so you're walking around the corner for your cup of coffee rather than a mile away.

None of which is going to happen in the US in the next 50+ years no matter what. We're not rebuilding the entire superstructure of the US anytime soon. We can't even agree to put together funds or routine taxes to properly maintain basic infrastructure.

Should it happen? It'd be great if the US had dozens of nice cities that were inviting to live in. It doesn't. It never will. It's not going to happen. People often don't want to live in cities in the US, because have you seen the state of most major US cities? Europeans think American cities are close to being of third world condition and it's not far off as an assessment. That's going to get worse, not better, as the US gradually decays toward being a richer version of Brazil. We're not going to have more money to spend on making things nice in the future (education, housing, infrastructure, social programs, etc), we're going to have a lot less - our skyrocketing debt and entitlement obligations guarantee that inevitable outcome, the US will rot. The political class isn't going to suddenly become amazing problem solvers, they're going to get worse too, as the culture continues to also rot (all of this is a direct product of a severe cultural decline in the US spanning many decades now).

The globally stretched superpower, pre-occupied with so many things that it shouldn't be, that now can't even maintain its bridges properly, isn't going to rationally reimagine and successfully rebuild its urban superstructures. It's obvious what the next 20-30 years will look like.

Whoa, easy there.

I would caution against painting with too broad a brush. There are many nice cities in the US and yes, many are quite livable without owning a car.

You are correct that the US has a major problem with a polarized and paralyzed political system though. But I think there's reason to be optimistic: smart solutions are also the cheapest. Effective transit, like bus lanes and signal priority, and safe biking infrastructure are phenomenally cheap compared to suburban highway and bridge construction, bordering on free. Housing affordability is, at its core, a zoning problem and we would make huge progress by simply letting people build houses and apartments.

While the Netherlands are famous for their bike lanes, which do provide many benefits, they own as much cars as any other European country.

If a particular city wants this for new development it can happen. Often starts from volunteering for the planning committees.

> I think the only way to reduce cars and car problems is by making cars totally unnecessary for daily life: Denser living with local work or WFH work. Diverse zoning so you're walking around the corner for your cup of coffee rather than a mile away.

100%! Everything comes down to building right. This is medium-density, mixed-use urban development. That means single family homes, small apartments (no more than 3 stories), gyms, places of worship, restaurants, shops, and stores all mixed together. 15 minute walk to anything you would need day-to-day. Then you have your personal car parked on the street or behind your house in a small garage for anything you need to do that requires a large vehicle or spontaneity or any of the other benefits that owning a car brings. But we get to avoid almost all of the damage.

"How do I do this with kids etc. etc." the same way people all over the world do it.

> Personal crap storage

That is a precise descriptor for my trunk. I believe you are correct with your assessment. Might depend if the "driver" would need a license. If so I believe people will still spend money on cars.

I don't think denser living is attractive either for that matter. As a student I would have loved it, otherwise nature is quite soothing. I am amazed that it is the "green" parties that want to solve housing crisis with more urbanisation. I don't get how suburbs are subsidized. High density housing will be subsidized and only affordable by giants. You will not own anything here.

I live in a high density house in a city. At some point I want out of here.

Examples of common and "hidden" car centric subsidies:

- Roads are basically free to use, transit users must pay a fare. Gas tax and other fees do not come close to the listed or actual cost (e.g. missing DOT salaries, equipment) of road maintenance.

- Number and magnitude of Federal and state grants for roads vs transit/bike capital projects

- Parking minimums in zoning code

- Public land allocated to highways and roads vs transit/bike/pedestrian

- Public land allocated to free or massively discounted parking vs bike racks, bus stops, etc

- In winter time, streets are plowed for free by the town but sidewalks must be cleared by private individuals. Bike lanes typically not cleared at all or are occupied with parked cars.

>Spontaneity: You need to wait for the car to show up

Imagine not having to walk to your parking spot or trying to find parking at your destination though. It will actually be _more convenient_ to carshare than to own private.

Nah, we'll just mandate that every business provide 4 parking spots per expected customer instead /s. If people _need_ to drive somewhere, they will want their own car and they will demand that others cater to their choice.

I still see making cars unnecessary and redundant as the key to healthier and economically resilient cities. If you can live a happy and full life without a car, it becomes a financial mistake to own one or to continue spending billions supporting their use.

Many European cities are easy to live in without a car. I did it for years.

Once you get past the acceptance of the premise, you'll find people objecting to adopting that lifestyle. For many Americans, the nearest city is not a desirable place to live so they don't know what they're missing. For the remainder that want to at least enable a car-free life, the best path forward is a matter of debate.

Sadly what I think goes unrecognized is that the suburbs have forced their car-dependent lifestyle on cities for so long that trying to reverse all the bad history urban highways and transit defunding is exceptionally tough.

Imagine you are a family with young children. You might live in a city right now - I live near Minneapolis where they've got a brand-new (like, less than 5 years old for most of it) lightrail system. It's shiny, it's fast, it's efficient, what's not to love?

Assaults. A 15-year-old girl was sexually assaulted 4 months ago. Less than a month ago someone got attacked with a knife. Less than a year ago, three teens were arrested for repeated racially-based attacks. In 2020 (because we don't have numbers for 2021 yet), there were 51 aggravated assaults and 239 "other assaults" despite 59% lower ridership from COVID. And it's not just assaults - drug use around (or sometimes even on) the lightrail is a significant problem (276 reported offenses 2020). Theft is yet another problem.

I'm not bringing my family (or even myself) anywhere near the lightrail. I'd prefer to drive just for their safety and mine. I know it's worse for the environment but their safety is my first concern.

And such is the problem with public transit. It's easy to think of it as reducing load on the environment while forgetting that lax law enforcement turns it into a somewhat scary place to be with family members.

> I'm not bringing my family (or even myself) anywhere near the lightrail. I'd prefer to drive just for their safety and mine. I know it's worse for the environment but their safety is my first concern.

It's not just worse for the environment but also much more dangerous to drive a car compared to the incidents you cited. You think you are in control because you're behind the wheel, but you're not. You have no more control over other drivers than you do someone attacking someone on this light rail. It's like if you ask someone today (2019) if America is more dangerous than before they'd say yes because that's what they hear on the news, but the statistics show otherwise. Similarly here but with car crashes actually causing quite a lot of death and destruction.

> And such is the problem with public transit. It's easy to think of it as reducing load on the environment while forgetting that lax law enforcement turns it into a somewhat scary place to be with family members.

It's mind-boggling how so many countries can do it but we can't in the US. How is it that China, Japan, Korea, Germany, Italy, the United Kingdom can do this but we can't? Are Americans objectively worse people?

> Are Americans objectively worse people?

In some particular ways - yes. We also just don’t have all the support networks that those other countries have. I mean - we still don’t even offer universal healthcare and it looks like we won’t for at least another 10+ years. We don’t care for the common person at all. We cater to the rich.

These other countries don't have calls to defund the police for one

Ok, but what about car jacking? Have you done the statistics to figure out which is safer? Also what about risk of health via car crash?

https://www.startribune.com/a-breakdown-of-2021-carjackings-...

Carjacking around here tends to take the fake-accident tactic. This is where the criminal bumps you enough to think an accident occurred, you pull over to exchange information for insurance, and then the criminal pulls out a gun.

I have a few notes.

1. Even though this is incredibly scary and stressful, the percentage of people actually shot in carjackings is almost unheard of. If I had a family, it would be a better trade to lose my car than to have one of them sexually assaulted on a lightrail, or have myself get a knife somewhere (both of which have happened within the past few months).

2. Carjackings can be (partially) avoided. The police have sent out warnings in the city that if someone bumps you and you suspect a carjacking, go to the nearest police station or at least drive to somewhere with a lot of cameras up close and witnesses (gas station?) before exchanging information. This means that carjacking still has some preemptive measures I personally can take, whereas riding the lightrail, stuff just mostly happens.

3. The map of carjackings in Minneapolis has far more people living in that area than people riding the lightrail. The lightrail, by number of people traveling on it and by size / concentration of people within a given area, is much more dangerous.

Now do hit and runs, drunk driving, and road rage shootings, and ask if you should take your family near a car.

(That's too ignore that accidents also happen more often than cars)

You reference an incident 4 months ago; but in 4 months, likely 300 people have died in auto incidents in the state.

The local authorities in California keep telling people to drive less and bike more, but then they allowed the bike trails to be turned into homeless camps. I know several people who were harassed or assaulted while cycling through those areas, so now they've gone back to driving for safety.

Clear example of public property and tragedy of the commons.

These are significant and general problems with society right now but perhaps orthogonal to transportation. Being able to drive past the homeless encampment might be convenient(?) right now but we've graduated from ignoring our crumbling, unaffordable car infrastructure to ignoring the crumbling, unaffordable healthcare and housing.

I don't understand how that is related to my argument at all. I'm saying that the lightrail is objectively not as safe or predictable as driving with my family, and for that reason, the reason to use the lightrail because it's better for the environment is of less importance to me.

Some suburban people in wealthy neighborhoods would say it's too dangerous to ever visit a city. It's unpredictable due to crime.

You've drawn the line at visiting the city but using a car, and maybe you are right, but I don't think listing incidents without context really makes your case.

'I know it's worse for the environment ...'

Considering all-in light rail costs per passenger, your assumption may not be correct.

>Rather than people owning a car they'll request one via an app, it'll turn up and drive them where they want to be

Which will mean turning up to the suburbs at 8am to drive downtown by 9am. Without a drastic, permanent change in working hours & commuting (or permanent WFH) the minimum number of cars will barely change.

The key is to build cities and surroundings around people and not around cars. Of course people will use cars if it's the best mode of transportation. But, as shown by many cities around the globe, if you make public transit or biking the best modes of transportation, people will not use cars to commute.

I think this is something people who have never lived in such cities inherently don't understand. Public transport and bike will always be a better mode of transport for most use cases (not for all, of course) if implemented properly - self-driving cars or not. And no, the solution is also not a hyper loop. Just good old buses, trains, trams, subways and bikes.

> I think this is something people who have never lived in such cities inherently don't understand.

I agree public transit is a solution, but I think the issue is with supply rather than demand.

The fact that the market value of land within walking distance of good public transit options or bike facilities is significantly more expensive is evidence that most people clearly do appreciate how valuable it is.

Unless more cities and towns build better public transit and bike facilities then a growing population of people will keep bidding up the existing supply and turn it into a luxury good that few can afford.

Yes exactly. But it's also a bit of a catch-22, since more people would need to elect government officials that care about this stuff, but they don't get elected because people don't know that it would improve their lives etc.

It's hard to tell people "just take the bus" when the nearest suitable home to the bus stop starts at $1M. Alt transit is making slow but measurable progress but the housing situation is the real priority if you want people out of cars.

Unless you're a believer in urban living, suburban costs are hard to beat, even when factoring in car ownership costs.

That's why I'm saying in "cities and surroundings". You can absolutely build great public transportation infrastructure in rural areas.

For example, in Switzerland public transportation is mostly public, and you can go to every nook and corner of the country reliably. As a stark contrast, in Ireland, a lot of the "public" transport is privatized and there transport options suck in rural areas.

Now, I don't want to make this about private vs. public (even though I absolutely think public transport should be public), but just to show that it's just a matter of how you invest money and set incentives to build this infrastructure.

This very much depends on where you live. In SF or NYC, sure, housing is very expensive but that's not the entire country and even there it's exaggerating to say you can't live near transit for less than $1M. We definitely need more attention to urban housing availability and improving transit considerably (i.e. not treating it as a last resort for the poor / disabled) but there's also a lot of inertia and assumption driving those comparisons which leads people to think it's cheaper to live further out even when they end up spending the same amount or more (at least hopefully for a preferred lifestyle choice such as a large lawn).

Without a drastic, permanent change in working hours & commuting (or permanent WFH)...

That's already happening.

> That's already happening.

Because of a worldwide pandemic. There's no guarantee that these changes will last.

It was happening long before Covid. Covid just accelerated it.

True, but there are also backlashes, especially when something is forced.

And too many remote work enthusiast assume that the steady state for this is majority remote work. I don't buy it. I'd be surprised if we go over 25%, maaaaybe 40% best case scenario.

It could also be a minivan or bus, which can be shared with workers from the same area.

Nothing is more enticing than a shared car with other people who are incentivized to push the cleaning and maintenance onto somebody else, while maximizing their own use of the vehicle.

It'll be like a public bathroom but with wheels.

Just like rental cars are public bathrooms on wheels?

Rental cars have someone incentivized to clean and maintain them, namely the rental car company.

something like this already exists and carries even more workers from the same area -- trains

Trains don't pick you up at your front door.

> The manufacturing of a car is responsible for a very significant amount of the emissions it produces over its life.

Nope.

The embedded energy in manufacturing a gasoline car is maybe 1/8th or 1/10th its total lifetime emissions. This is why EV's are so much better cradle-to-grave than gasoline cars -- yes, EV's require extra energy to manufacture the batteries relative to a gasoline car, but even so, that extra energy cost is more than offset within the first few months of electricity-powered driving, even if your local grid is primarily coal powered. By far most of a car's lifetime emissions are from operating it, not manufacturing it.

See chart on page 3 of this Union of Concerned Scientists report: https://www.ucsusa.org/sites/default/files/attach/2015/11/Cl...

Does anyone make a driverless car that isn’t a hybrid or an EV?

EV's are better than fossil cars, but they too have most of their emissions in operation rather than manufacturing.

So the significant question isn't "will driverless lead to fewer emissions because of fewer cars", the significant question is "will (electric) driverless lead to fewer emissions because of fewer miles driven?"

It's no question that electric is more efficient on emissions per mile, but if the number of trip-miles increases because driverless cars eg are constantly circling the downtown waiting for passengers, or they make more low-occupancy trips possible so they induce demand of low-occupancy miles, well it could be worse emissions wise!

This model works well if you're a single person traveling alone and have relatively few things you need to transport with you. In other words, the current "hail a taxi" model.

It becomes less workable when you need to travel as a family with kids, car seats, strollers, other equipment. Having a dedicated vehicle that belongs to you and is always at the place where you left it is immensely useful in these circumstances.

Yep. And as with many things there will be a mix of solutions for different needs.

When everyone needs a car to get to work at the same time and you can't be sure thst you will get one when you request it - how can you plan your trip to work or anywhere, if getting a car is random?

Maybe we could come up with some sort of 'Metacar' system where people walk, bike, or take a short ride to a central pickup point and then all ride in a 'Metacar' with passengers headed in the same general direction.

I wonder if the economics work out so that it would be affordable to bore tunnels under cities for these 'Metacars'. That would make it much easier to have them be driverless, and they could proceed directly towards their destination at higher speed. Even if that isn't possible, we could designate some lanes on bigger roads as 'Metacar' lanes so that they can move efficiently even in rush hour.

In all seriousness: I just realized that I haven't had a job in 5 years that required me to commute in a car. COVID helped, but the other half of it was working in places that take density and non-car transit seriously. I know not everyone can do that.

Your ‘Metacar’ system is literally describing public transit systems that already exist.

Most people drive to work alone. If everyone is using a rideshare app lots of those people will be using the carpool feature. If every car has 4 people in it you only need a quarter of the cars.

The only thing worse than a 45 minutes long commute trip twice a day alone in the car is a bit longer commute trip twice a day with other people.

You prebook your car in advance. This is certainly not a hard problem.

I do not see how that solves the general problem.

Either there are as many cars available for prebooking as their are rush-hour commuters who want them or we have to ration the supply using shortages and/or surge pricing. In the former case we haven't appreciably decreased the number of cars. In the latter case I anticipate that anyone who can afford to do so will continue to own a private vehicle to ensure they can get to work on time, so we also haven't appreciably decreased the number of cars.

It solves the problem that you can plan your trip. The complaint was that you couldn’t plan your trip.

Not everyone go to work at the same time, so every car will be available for several trips. And what would be wrong with prices that varies with demand? Isn’t that how market economy is supposed to work? It will still be cheaper than owning your own car.

It is theoretically possible, sure.

I'm trying to imagine a nontrivial number of people who currently own a private vehicle selling that vehicle to rely on prebooking both of their daily commuting trips with a flexibly-priced mode of transportation to save money.

As a datapoint, I know many people who regularly take car services like Uber, but I don't know of anyone who actually sold their car to rely exclusively on a car service to commute. Perhaps if driverless cars halve the price of car service it might start to happen?

I'm not saying it is not theoretically possible that driverless cars will reduce car ownership, I am saying that I am skeptical of the weak arguments and minimal evidence provided thus far for such a claim.

There’s not much evidence in either direction before they are a reality.

pre-booked cars allow an app to arrange a de-facto floating car pool. There will be a shared fee that will depend on the length of the trip optimized segments. Ride alone types will pay more.

When everyone needs a car to get to work at the same time

They don't.

You're going to have to expand your pithy take there. Rush hour still exists. Bumper to bumper traffic still exists. There's an awful lot of people who apparently need a lot of cars all at the same time.

All those people, or more accurately the companies that employ them, choose to have the same operating hours, the same inflexible time policies, and the same insistence on going to a physical location to do work. Some of them won't have a choice of course - you can't really stack shelves or manufacture bits of metal remotely.

For many of the rest travelling, commuting, and working in an office is a choice. It is not a need. The fact rush hour traffic exist shows that businesses are inflexible and unwilling to trust their workers; it doesn't show a daily commute is a requirement.

They don't need to and they don't all need to get there at the same time, but rush hour exists. Even if there would be cars available even during rush hour, people will expect rush hour to have less availability and will err on the side of not getting fired.

> Rather than people owning a car they'll request one via an app, it'll turn up and drive them where they want to be, and then it'll go off to the next customer. Car utilization will go up significantly. It'll be like all cars are taxis.

Traffic is quite bursty: there is a lot of generally unidirectional demand for trips during the morning rush hour, and then generally unidirectional demand in the opposite direction during the evening rush hour, and much less demand for anything in between.

You have to take into account peak usage and depreciation.

Cars are typically only discarded once they hit a certain distance. No matter if the car is a taxi or a personal car, it will be discarded after the same number of miles.

You also need to buy enough cars for the worst possible scenario, which is the peakiest of peaks.

Unlike a personal car, robotaxis will also be replaced as soon as it economically efficient to do so, whereas cars are typically eventually driven past the point at which they are economically viable by someone.

I do not believe there will be a significant decrease in car manifacture.

The only way it could be is if instead of robotaxis, we built robobuses that carry 5-6 people on average.

> No matter if the car is a taxi or a personal car, it will be discarded after the same number of miles.

Is this actually true? My personal anecdata would suggest that most passenger vehicles are not driven nearly the number of miles that a taxi or commercial vehicle are driven.

A quick google suggests passenger vehicles are typically salvaged at 100-150k miles whereas an average taxi might do 500k or more before retirement.

The average car in the US seems to be 12 : https://www.wsj.com/articles/average-u-s-vehicle-age-hits-re...

And the average yearly mileage in the US is around 13.5k, yielding an average mileage of 173k miles. It is incredibly unlikely that the average car is salvaged at 100-150k miles. 200k+ miles is more like it. This number will also increase as vehicles produced recently have much higher durability, while the 16 year old tail-end is dragging down personal cars.

I can't find a great source, but it seems for example USF only expects 85k miles out of a car, for example: https://www.google.com/url?q=https://www.usf.edu/administrat...

Taxis are an exception with around 210k average mileage, but that's because they are individually owned.

Rental cars seem to be retired at around 60k miles : https://www.wsj.com/articles/rental-cars-with-higher-mileage...

It doesn't make sense commercially to operate a vehicle long after its warranty ends. You risk losing revenue when the car breaks down, and you can expect upwards of 1000$/year in repairs and increased insurance costs as an extra. It makes more sense to sell the car shortly after the warranty ends, recoup 50% of the costs, and buy a new one.

Rentals are somewhat of an odd model, at least pre covid. By buying less popular models/excess capacity they buy at slightly above build cost, and the manufacture or dealers buy the cars from the rental company at it's depreciated value.

https://www.autorentalnews.com/155877/car-rental-qa-what-are...

As I wrote below, a 2006 NHTSA report found the average passenger vehicle was scrapped after 152,137 miles. Of course, that's 15 years old.

The taxi number seems pretty low. Do you have a source for that?

That's actually a lot more than 15 years old. The vehicles that were being scrapped 15 years ago at 152k miles were themselves 15 year old, so we're really talking about the lifetime of a 30 year old car.

You're way off. Average passenger vehicles in the US are scrapped closer to 200K miles. Of course there's a lot of variation by state because vehicles driven in winter on salted roads rust out much faster.

That number has been increasing every year, so the OP may simply be remembering an earlier number. We didn't use to keep cars around for so long but they have gotten more reliable (and rust-proofing has made big advances). It will soon be 250K.

The latest comprehensive figure I could find was a 2006 NHTSA report which found the average passenger vehicle was scrapped at 152,137 miles. It may have gone up, but I haven't found a more recently number.

2006 was a while ago, so that sounds about right.

Here's an article from 2012: https://www.nytimes.com/2012/03/18/automobiles/as-cars-are-k...

Here's an article from 2018: "a typical passenger car should last 200,000 or more miles" https://www.aarp.org/auto/trends-lifestyle/info-2018/how-lon...

Taxis definitely are built for a different use profile and they go more miles. Ford taxi version of Crown Vic had an hours meter.

>No matter if the car is a taxi or a personal car, it will be discarded after the same number of miles.

What? I'd disagree with just about everything there.

In general fleet vehicles go much farther because of regular maintenance.

The other factor that must be taken in effect is will the rate of accidents that total cars decrease, if so, the average total mileage will increase.

The decreased rate of accidents is true. A newly built car can be expected to be retired after 250-300k miles. Rental car companies seem to be trying to sell their cars after around 60k miles : https://www.wsj.com/articles/rental-cars-with-higher-mileage...

I hope you are right but I am skeptical that the reduction in the number of cars will offset the increase in miles driven over the next couple decades where it matters most.

Shared-use vehicles exist (the taxis you mention, as well as car shares and rental cars) and with the exception of cities that have extremely expensive car ownership costs most people still choose to own a personal vehicle for commuting. Would you rely on the availability of a shared vehicle during rush hour to meet your commuting needs?

I have a very hard time imagining many Americans who live in a suburb or car-dependent city giving up their private vehicle no matter how cheap or reliable shared vehicles become. Perhaps if car ownership became two or three times more expensive (e.g. carbon taxes on new vehicles, no parking minimums, paying market rates for parking spaces, etc) people might switch, but I am skeptical that any of those will come to pass as long as most voters drive cars.

Driverless cars de-incentivize carpooling and incentivize idling on public roads to avoid storage / parking costs (unless states implement fee structures for occupying roads).

A taxi model means you can compete for market share by being fastest to respond, which means you can fill the roads to the brim with your vehicles to push your competitors out of the market with sheer capital expenditure. You would be further incentivized to do this through economies of scale.

You only need to look at how taxi freelancer apps increased the number of vehicle miles on the roads in all major cities and how bike rental apps filled urban areas with bikes annoyingly scattered all over the place.

Any car based transport vision of the future means more cars on the road, and consequently more car-focused infrastructure.

To really be more efficient you need to focus on passenger miles / unit of energy, which very decidedly means mass transit.

Not to forget that if the cars don’t have to be parked next to their destination, towns and cities can be built much denser, so the distance travelled will be smaller. That might even lead to people start walking more since they don’t have to cross 400 meters of parking lots to get to Starbucks any more.

No thanks. It would be nice to have the car drive itself (although I doubt we'll achieve true SAE level 4+ capability on most roads in my lifetime). But I keep a lot of useful stuff in my car. I don't want to load and unload it every time. The car also serves as a reasonably secure storage locker at my destination.

But your model might work in shithole cities like San Francisco. No one can leave stuff in their cars because they have to park on the street and auto burglaries are rampant.

> The manufacturing of a car is responsible for a very significant amount of the emissions it produces over its life.

According to the Kurzgesagt video about climate change and how to stop it, what's even more significant is the carbon emmissions of road construction. Apparently a single meter of road is equivalent to one car. If we increase car use, we need more roads and those roads need more maintenance and all that has a big carbon budget.

We simply need to move our society away from private car infrastructure. My opinion is that a well-designed urban environment built for a mix of electric trains, buses, and e-bikes, with dense housing and mixed use zoning would work really well, as shown in places like Scandinavia, Holland, etc.

And all the food and goods you need will be delivered to you via helicopter?

We need the roads for trucking, if nothing else. But ambulances and fire trucks can be pretty useful too.

Nowhere in my comment did I say we should do away with roads completely

More likely you will see people go from 2 car households to 1. I know alot of people that have 2 cars simply because one of them sits at a person place of work all day, if they could have to car return home then it would allow them to only have 1

I think it'll be more than that, but even if you're right that would still be a massive reduction in the number of cars.

But I can buy a self-driving car and sleep in it for my commute from Yosemite to San Francisco.

Reply on news.ycombinator.com