Comment by esotericn | Hacker News Reader

Comment by esotericn | Hacker News Reader

Parent Full thread

The 'electric' part of the article here is fluff. Electric cars already exist and will take over the market within a similar time frame to any proposed self-driving car takeover (it's primarily a matter of manufacturing capacity, which affects this company too, and waiting for older cars to die out at this point).

Aside from that, there are three main ways I can think of that a self-driving car could reduce pollution.

a) More people fit in the same car and an intelligent mechanism can pair riders. This exists today in the form of things like Uber Pool. Without a driver you might get one additional passenger in a car.

That's the main mechanism. In the best case it can reduce energy use per passenger by around 20-25% - in a realistic case it would be more like 10% or less.

b) Cars can drive closer together and with more intelligent traffic lights. I think this is incredibly overblown - pedestrians exist and cars are already pretty close together in cities.

I doubt this will make much difference especially given that electric cars use ~zero energy at idle.

c) Over a long time-frame, cities are redesigned such that parking spaces don't exist and so things are closer together.

This is both a multi-decade (possibly even a century) change, and likely to just result in people commuting further - the 1 hour sweet spot isn't going anywhere, all people won't stack in cities unless they're forced, if we're being charitable, say, 50% of people actually prefer that lifestyle.

It also works for taxis.

Basically, the whole thing is nonsense.

What I see as being more likely if self-driving takes off is crackpots sitting at desks in the back of their van, or chilling on a sofa watching movies, and doing multi-hour commutes, offsetting the savings.

I've not even mentioned here the fact that a significant amount of people just like to own their own car and have the funds to do so. Particularly if you're in a rural area, the idea of some SV startup self driving to your ranch is absurd, you're almost certain to either be priced out or just directly geofenced out.

> a) More people fit in the same car and an intelligent mechanism can pair riders. This exists today in the form of things like Uber Pool. Without a driver you might get one additional passenger in a car. That's the main mechanism. In the best case it can reduce energy use per passenger by around 20-25% - in a realistic case it would be more like 10% or less.

And it would reduce the number of cars on the road, the number of mechanics shops needed to service those cars, the number of insurance companies, most of the infrastructure around licensing, etc. Lots of knock-on effects that would also impact emissions, and a fleet of self-driving cars permit a lot more vertical integration, which reduces TCO.

A fleet of self-driving cars will simply be more cost effective and reduce emissions. Car ownership becoming a luxury is probably inevitable. It will become a collector's hobby, like old sports cars are today.

> A fleet of self-driving cars will simply be more cost effective and reduce emissions.

How? I drive from point A to point B in my car. I'm there for a while, then drive from point B back to point A. Meanwhile, someone else drives from point C to point D, is there for a while, then drives back to point C. Total trip legs: A->B, B->A, C->D, D->C.

Now a self-driving car takes over. It drives me from A to B, goes to C, drives the other person to D, gets me at B, takes me back to A, goes back to D, picks up the other person, and takes them back to C. Total trip legs: A->B, B->C, C->D, D->B, B->A, A->D, D->C.

There is no world in which the second set of trips emits less than the first, given equivalent vehicles. It's simply not possible.

(Note that this is different from the have-everything-delivered scenario. The old way was you drive from A to the store at S, then drive back home to A. Someone else drives from B to S, then back to B. Delivery drives from S to A to B to S, which can be more efficient.)

> There is no world in which the second set of trips emits less than the first, given equivalent vehicles. It's simply not possible.

Firstly, you completely neglect to consider ride sharing, where multiple people are heading in roughly the same direction and so a portion of the trip reduces from 2 legs to 1, ie. the simplest case from your examples would be A->C->B->D(wait)->B->C->A. The more people that use the service, the more likely this is to happen, and the more emissions are reduced. It's the exact same aggregation argument supporting the use of transit.

Secondly, you're only focusing on specific local effects rather than systemic effects. The number of cars on the road would be dramatically lower [1], so we're manufacturing, maintaining, licensing and insuring fewer cars, all of which counts towards reducing emissions.

Then the marginal reduction in convenience means some people would change their behaviour, ie. travel less and/or plan their travel better so they get more done in a single trip.

Frankly, it seems clear that there is no world in which people owning their cars would be superior or even equivalent to a fleet of self-driving cars when it comes to environmental impact, traffic or efficiency.

[1] https://news.ycombinator.com/item?id=21783531

Only if self driving cars are legally required and expensive. If self driving cars are not legally required (ie because they are that much safer), or they are cheap people will continue to buy and own cars. The convenience of a personal car with your golf clubs in the trunk is worth it.

Don't forget that the self driving world you imagine is worse for the environment: all those cars going empty to a distant parking lot until you are done with work eat up a lot of energy.

If you want to solve environmental problems from cars invest in public transit. The biggest problem in modern cities is neighborhood designs that make public transit too slow. (you need to have a short walk to the stop, and then the vehicle not make many twists and turns along the way to where you want to be). Density is useful for transit, but suburbs already have the minimum density required, they lack the ability for transit to be fast though.

> If self driving cars are not legally required (ie because they are that much safer), or they are cheap people will continue to buy and own cars.

They will be cheaper. No licensing, insurance, maintenance, fuel, but still almost all of the convenience of a personal vehicle requiring only a short wait time.

> Don't forget that the self driving world you imagine is worse for the environment: all those cars going empty to a distant parking lot until you are done with work eat up a lot of energy.

That's not how it would work. If parking becomes necessary because network capacity is too high and utilization too low, the car would park in its immediate vicinity. We have lots of parking everywhere to accommodate far more cars than would be needed in a world with mostly self-driving cars.

Why wouldn't their be licensing? Insurance maybe. Maintenance is a requirement - the tires will wear out if nothing else. Fuel is still a factor - it might not be liquid, but you need some sort of energy.

The advantage of a personal vehicle is 0 wait time. If you are willing to wait the bus is cheaper because you can share the ride with so many other people.

Sure we have parking everywhere. However at that point there is essentially one car per person anyway, so it may as well be your personal car so you can leave the golf clubs in the trunk.

> Why wouldn't their be licensing?

Do you need a license to hail an Uber? If that Uber became self-driving, why would you need a license? Certainly Uber cars would need to have some kind of inspection, but that's just an extension of whatever inspections exist for road-worthiness.

> Maintenance is a requirement - the tires will wear out if nothing else. Fuel is still a factor - it might not be liquid, but you need some sort of energy.

These are no longer costs born by the driver/rider, but are now aggregated by the car service. Aggregating yields considerable cost savings.

> The advantage of a personal vehicle is 0 wait time. If you are willing to wait the bus is cheaper because you can share the ride with so many other people.

A bus wait takes a minimum of 15 minutes. An Uber arrives within 5 minutes, and with self-driving cars and capacity planning, this can be brought down to some arbitrarily small number if needed. A bus also requires walking to and from stops, which also adds at least 5-10 minutes for most people. That's a big difference, not to mention these are multiplied considerably in rural areas or areas with considerable urban sprawl. And the first few minutes of waiting for an Uber I'm putting on my shoes, jacket, etc.

Walking down to your car takes at least 3 minutes in a condo. If you own a home with driveway, it will still take you 1 minute to walk to your car and start it up (and a few minutes to warm it up if it's winter).

So the advantage of ownership amounts to about 1 minute in winter climates, and maybe 2-3 minutes in warm climates. That's perfectly reasonable tradeoff considering the expected cost savings.

> However at that point there is essentially one car per person anyway

I don't know what this means. My point was that we don't need to build huge parking lots outside of the city for idle self-driving cars to which they would drive to/back when utilization is low, we already have a glut of parking available for a transition to a self-driving world, and all localized within our cities. A self-driving car service would just use all the parking scattered everywhere throughout the city and so have cars available nearby, no matter where you are.

There is still license fees on the car - aggregated, but still there.

I already aggregate the cost of maintenance across all my personal driving: there is no cost savings. (if you do your own work it is more expnesive)

A bus wait can be as little as 3 minutes max (q.5 minutes on average). If it isn't the is the fault of lack of ridership. Walking counts as exercise, like it or not your doctor wants you to get more.

Rural areas yes - but rural areas have a lot longer to wait for a shared car.

> I already aggregate the cost of maintenance across all my personal driving: there is no cost savings.

No, aggregated across all drivers, not just aggregated across your driving. As in, 1 million drivers insuring, licensing and maintaining 1 million cars costs far more than 1 million riders sharing 200,000 cars whose insurance, licensing and maintenance is aggregated by a riding sharing service like Uber, and those savings will show up in your final bill, ie. total cost of ownership of your own car will exceed using ride sharing services with self-driving cars for virtually all people.

> A bus wait can be as little as 3 minutes max

It can be as little as 0 minutes if you get there at the perfect time. That's irrelevant though. Worst case and average case wait times are what's important.

> If it isn't the is the fault of lack of ridership.

No, it's the lack of sufficient incentive and benefits to use transit. America is not Europe.

It won't be 1 million riders sharing 200,000 cars. It will be 1 million sharing 900,000 cars at best. There is a reason a single occupant car is by far the most common - for the individual it is the easiest, most convenient choice. No waiting for a shared car, leave when you want, keep your personal things in the car for latter.

The same lack of incentive and benefits to use transit applies to shared cars.

That's all conjecture. Queueing theory can plan spare capacity for any level of utilization. The higher the utilization, the lower the cost to the individual rider, ie. the more sharing that's allowed or encouraged, the lower the cost. The sharing equilibrium is almost certainly not 10%.

> The same lack of incentive and benefits to use transit applies to shared cars.

Sure, but the increased convenience of not needing a shared meetup/dropoff of transit is considerable. The biggest cause of traffic into and out of a city is the 1 driver commute from the outlying regions. If car ownership were more expensive than ride sharing, most of these people would use ride sharing because it has the convenience benefits of car ownership with the efficiency benefits of transit. The ride sharing equilibrium in this model would be upwards of 70% for rush hour, at least.

Non-rush hour times would have fewer cars on the road because the cars would park themselves when utilization falls. This is more efficient and more convenient than transit.

Reply on news.ycombinator.com