The self-driving car is a red herring
nautil.us
nautil.us
"If your errors are irreversible and you can't average them out, and your application is in open domain where there will never be a shortage of out of distribution samples, you are royally screwed. That is why I'm so skeptical of self driving cars, since that is exactly the application described."
For me it's obvious the real issue are the edge cases - of course a self-driving car does fine in 99%+ of cases, but those small % of outlier cases are the ones where people potentially die. I suspect an entirely different approach is required to supplement/supplant the deep learning one and fully autonomous vehicles are still quite some way off.
I do like all of the driver assist advantages coming out of this work however! I don't think you can underestimate the value of having technology that works _with_ a driver until the point where self-driving cars become viable.
[0]:https://blog.piekniewski.info/2020/06/08/ai-the-no-bullshit-...
Road tunnels exist and have been built all around the world, these kinds of ideas have been tried and ultimately general underground road tunnels are just not economical or particularly safe.
Thunderf00t has done a series of videos debunking these claims [0,1,2]. Unfortunately Elon has a habit of frankly lying which I think does serious harm to public perception of these things.
These kind of infrastructure problems are just hard and have a ton of trade offs, there is no silver bullet.
[0]:https://www.youtube.com/watch?v=HBtL3qDvdZc [1]:https://www.youtube.com/watch?v=0ezF7NmwQZs [2]:https://www.youtube.com/watch?v=Hyy-19qyNWE
In the same vein, why underground instead of light rail?
See, open air systems like those are much easier to service. Removing a crashed car from that requires just a big crane.
Cost also vastly increases, likelihood of total failure as well. Big tunnels need big supports and big service ducts.
It seems like such a waste of time for people to try and debunk these claims from people who are very competent and have access to the best in people and technology. Do they want a world where people don't try new things and make great improvements? One should really encourage people with a track record of getting people together to do amazing things to continue to do so. Elon Musk has stated that he thought that SpaceX had a 10% chance of success. I'm glad he could look at that chance of failure and tried anyway.
Blue Origin is building a large re-usable rocket like the Starship called the New Glenn, but its first launch date keeps slipping. I would guess that Starship will be launching first but it is really cool that two groups are working on such rockets.
There are fields where physical limits were already reached, for example rocket engines specific impulse has barely improved since the 60s because of fundamental chemistry, but this is not the case for tunnels.
Provided enough funding it should be possible to dig faster, and cheaper per km.
> the safety issues of a car breaking down in a tunnel
The cars have front radar and can break. It would stop the line but I don't see how it is a safety issue. Most danger in tunnels is from bidirectional traffic that Boring avoids.
> the enormous cost with digging deeper tunnels
The same could have been said about other Elon Musk projects. Nobody is surprised that an electric car can have better performance and power and yet a diesel driven tunnel boring machine can not be improved? Tunneling is incredibly conservative and uses both diesel driven tunnel boring and diesel driven locomotives to transport out the dirt. The inherent problems of having lots of diesel engines a tunnel are quite clear and eliminating it has clear operational advantages.
> underground road tunnels are just not economical or particularly safe
What the evidence for underground one-way tunnels not being safe?
> These kind of infrastructure problems are just hard and have a ton of trade offs, there is no silver bullet.
Therefore we should try to improve it with modern technology from other fields? Its not a silver bullet, it is tacking new technology from automotive and aerospace and applying it to tunneling and mass transportation.
!? I'm not a tunnel building specialist, but a cursory search seems to indicate that tunnel boring machines commonly already are electrically powered by simply using a very long high voltage cable stretching all the way from the tunnel entrance. Whether that cable then connects to the electricity grid, or to a bunch of temporary diesel generators stationed outside of the tunnel is a different matter, which in any case doesn't bear on the tunnel boring itself.
It's unfathomable to me that people believe car tunnels are the future of anything.
Right now The Boring Company is trying to recreate an underground train in Las Vegas, and is already behind schedule and running into issues people predicted before this started:
https://www.theverge.com/2020/10/16/21519692/elon-musk-borin...
One system is cheaper because the units are mass produced far more often and the system is more flexible.
Carrying around a battery rather then having complex high power electronics along the tunnel seems like a no brainier as well.
Las Vegas is literally a first of its kind system, being a few month behind is not surprising or concerning. And the problems have to do with specific fire codes, not any inherent flaw in the system.
Secondly, even if you're right (and you may well be) - imagine that a common edge case was a small child jumping out in front of the car. Would we want to accept a vehicle that consistently ran them over? The raw numbers might be better but that doesn't fully paint a picture of what those numbers would look like.
I also think the 'will randomly kill people, but rarely' aspect of self-driving cars will be majorly off-putting on a very human level.
And finally there's a perception issue here - the self-driving hype is around the idea that the car is effectively being driven by a far better driver than a human, the second it turns out it does really crazy things sometimes (doesn't even have to involve fatalities necessarily) is also going to put people off I think.
I personally prefer the idea of driver assists (responsibly developed so drivers aren't just distracted) aiding driver safety while we work on a supplemental self-driving solution, or perhaps as discussed in the article create some infrastructure along roads to make things more certain.
I am far from being against self-driving cars, I just feel the timeline is longer than currently hyped.
An autonomous vehicle, in comparison, is not relatable at all. Some equation delivered some wrong result, maybe it was something in the training set, so that's why the car suddenly drove into you. Bad luck. If you're lucky, you get some damage payments from the car company, if you happen to have enough money for a very good lawyer.
Sure, as a technical person, I understand some of the possible errors, but it still feels like a complex system that one cannot relate too with hardly understandable edge case behaviour. That's probably why we're so much more lenient with humans.
Except, depending upon your perspective, it really isn't most places. In the US it's about one fatality per 100 million miles driven and that includes driving under difficult conditions like snowstorms. People drive a lot so that still adds up to a lot of deaths per year. But the rate is not really all that high.
https://en.wikipedia.org/wiki/Transportation_safety_in_the_U...
Overall, I don't really disagree on net though.
> OTOH, I doubt a self-driving vehicle will be all that better at dealing with a deer jumping out in front of a car, unexpected road debris, or black ice all that much better than an alert human driver. (And possibly worse given that a human driver may pick up on certain cues that an automated system might not.)
I'd expect the opposite actually. Humans are incredibly slow reacting in the best of circumstances and they often drive distracted. They also suck at reacting even when they do which is why we now have automatic systems to reinforce braking in presumed emergencies as humans don't consistently apply full brakes in emergency situations. Collision avoidance in a few milliseconds with reliable reactions should make quite a few of those scenarios much better.
I think it boils down to "because we can and still have a viable commercial airline industry." (And general aviation is much riskier than commercial airlines.) Whereas were we to vastly increase the training, maintenance schedules, monitoring of driver alertness, and requirements to have all the latest safety features, you're probably effectively eliminating driving for most people who can't just hire a professional driver/car.
For many things, we accept that we could make things safer if we wanted to. But were we to do so, we'd effectively be banning the activity or at least severely restricting it. (Which is probably one reason why we don't really have flying cars.)
Why do you think this? It makes sense to me intuitively that people would be willing to accept more death causes by other people than more opaque and unfamiliar autonomous systems, but I haven't seen any hard data on this and am curious. Is this mentioned somewhere in the Wikipedia article that you linked or did you just share that to demonstrate the other number you mention?
So, planes might be around 500x better per pax-mile, but per pax-journey they're just about 5x better.
Incidentally, the fatality rate for self driving cars should be also around 5x better (per vehicle mile) than human drivers for wide spread acceptance, according to an 2018 study.
https://www.sciencedaily.com/releases/2018/05/180530132959.h...
The study showing 5x is quite neat, thanks. I expect this to be a floor for the difference though. Once some of the deaths with self-driving cars are shown to be things that human drivers would easily avoid the requirements will be higher, even if in total 5x better has been achieved. For widespread acceptance I'd bet on 10x as a minimum and 100x at the extreme, depending on what types of accidents are in there that shape the perception.
On the freeway you minimize a lot of those wildcards like bikes and pedestrians, everyone is going about the same speed, and there are only a few places you can enter and exit.
We don't need to jump all the way to self driving in SF, and have people point out all the things that can go wrong, since theres a huge gain just from being able to do it on the freeway.
So why hasn't this happened yet?
I believe this is because what we have here is not a technical problem, but primarily a regulatory and legal problem. If the political leadership of the NHTSA had the will to make highway self-driving happen, they could publish a draft safety and liability standard.
The technical problem here is that while shuttle and robotaxi systems can go slow enough as to make fatal accidents rare, as you said, the most consumer benefit potential is for a highway self-driving system that would not require active monitoring like Autopilot does.
The political/legal question is: when fatal accidents occur, who has the legal liability? If it's the manufacturer, then why would they accept the risk? If it's the driver, then who would actually go to sleep and let their robot potentially kill people? Especially given the problems that Tesla autopilot has with stationary targets in the road.
It's possible that the key innovation required here is not anything technical, but an insurance product that would allow the manufacturer to assume the liability, but be insured for damages. This insurance product would involve a deep technical evaluation of the self-driving system to get actuarial-level probabilities of mortality.
On the NHTSA website, there are many >50 self-driving test sites, but only 3 which are classified as "highway". The most common vehicle type is "shuttle". A full self-driving airport shuttle, with purpose-built road markings, makes a lot of sense as a goal achievable with today's technology, and is far more likely to kill people due to its controlled built environment and slow speeds. https://www.nhtsa.gov/automated-vehicles-safety/av-test-init...
[1] https://www.linkedin.com/pulse/driverless-car-stopped-lawyer...
> In the case of vaccines, consumers
> cannot sue the manufacturers using
> regular courts but instead there is
> a separate system where the standard
> for liability is much higher.
And you wonder why there is no political will behind this?
But if it can get to half the fatality rate of human drivers, that'd be a huge win. And unlike human drivers, every fatality involving an SDC can become a fleet-wide software improvement to, ideally, prevent that failure from ever happening again. That's not something we can do with people; nobody but traffic engineers and wonks read NHTSA reports to find out how accidents could have been avoided.
Why is that every time the safety of self driving cars is brought up, people appear to ignore that human drivers are not 100% perfect. There are on average of 104 deaths on the road each day in the US. if we can get that number to 50 people, that's a net positive.
Self driving cars don't need to be perfect, just better than us.
So the question shouldn't be "are SDC perfect yet?" but "are they better than us?"
Would you get in a vehicle if you could be charged with manslaughter for killing someone through absolutely no fault of your own?
Conversely, can imagine society collectively throwing up their hands in the event of a death caused by an autonomous vehicles and going "Them's the breaks. It's all for the good of society."
Also it is not at all obvious that self driving cars will do better than humans. Perhaps they will, but it isn't just about numbers. As I alluded to in another comment, imagine if they had 50% of the human death rate but it was all small children and occurred seemingly randomly. Do you think that would be acceptable and a better outcome? These things are less black and white than they might seem.
Another factor is that with deep learning it is fully a black box. It is very difficult to understand WHY a particular course of action was taken, whereas in human cases you can usually determine this. The why might also be limited, e.g. 'human child was mislabelled as stop sign' or similar, which just causes further questions.
The fact that you have scenarios the car will definitely not know how to handle combined with a deep level of inscrutability makes it difficult to have confidence in issues being resolved or the technology improving on a safety basis.
The only way to remotely look at this meaningfully is in aggregate. Deaths and losses per million trip-miles.
This is where the momentum will come from. There will be a day where machines reach and surpass the human level, and we'll see a huge momentum towards penalty-boxing human drivers. I fully expect to call Progressive in 2040 and the hologram Flo will tell me that it costs four times as much to insure me if I have the audacity to retain a steering wheel.
I also think there's a snowball factor. Self-driving cars are going to be way more consistent and predictable than human drivers. That only amplifies as more and more vehicles on the road are self-driving. Once you get to 90% self-driving, I could imagine things like increasing speed limits and dynamically organizing convoys for efficiency.
I'm not in the US, and while the public transportation system here is far from perfect, I've managed to get by this far without a car and don't plan on buying one anytime soon. I much prefer a future where this is improved compared to one I have to fork tens of thousands of USD for a chromatic death machine.
Part of the problem is the general incompetency and mismanagement of US public transit construction -- the costs are insane, even when compared to "socialist" "high-cost" union-heavy Europe.
This was a great read on why we get so little for our money: https://www.nytimes.com/2017/12/28/nyregion/new-york-subway-... - and it busts the common misconception that it's 100% because unions. (tldr: it's because politicians get rewarded for creating jobs (votes) and spending money on contractors (lobbying), and because we outsourced everything)
Right now there is a difference in quality of that time in that your attention is more free when taking public transit but self driving cars fix this issue.
A fleet of SDCs snaking down a road can "tuck in" tighter to each other because computer reaction times are better (especially if they're networked and sharing decisionmaking signals).
Of course, it's still not as space-efficient as a bus. But the bus infrastructure as a solution has been challenged by COVID-19.
The constraining factor on SDC commonality in the short-to-medium term will be maintenance. These cars require dedicated service (which is why they'll likely start as replacements for taxis, not privately-owned; people will have to be hot-rodders to want to do all the work necessary to keep the sensors clean and tuned).
There may be a threshold in the future where the sensors are hyper-rugged or self-maintaining, but it's a little ways off. The maintenance burden will keep the ghost cars down in the short- to medium-term.
Human-driver cars consume parking spaces at the destination, self-driving cars don't have this limitation. They park themselves away from high-traffic areas or they could ferry other people around.
But that means it's not just a matter of building mass transit. First you have to build taller buildings.
And even then there are going to be places where it's just not happening. It doesn't matter how you change the zoning, nobody is going to build skyscrapers in Pine Bluffs, Wyoming.
If I'm not wrong, then we have vastly different definitions of "great public transit" -- I would not be happy there without a car.
The key is to realize the cars themselves killed density.
Parking lots, parking spaces, and extra lanes all conspire to push humans and human spaces further apart. This then makes walking less feasible, cars more required, and more space required to accommodate those cars in a feedback loop.
Taller buildings are not required for density, far from it. Look at Somerville, MA, where just about nothing is higher than 3 stories, yet they fit almost 20,000 people into a square mile -- and they like it. What they don't fit is 20,000 parking spaces, and that makes all the difference.
That's less than 1/3 the density of Manhattan and about on par with SF.
>where just about nothing is higher than 3 stories
The point is that cars destroy density regardless of what your building height limits are.
During this period the federal dept of transportation was offering to pay 90% of urban freeway projects. There are a few famous examples of people negotiating a different outcome, like the light rail vs mt hood highway in Portland, Oregon. But the bulk of local politicians simply took the free money and built massive freeway infrastructure without much consideration of the future.
So no, this was not some broadly democratic choice or invisible hand of the market. It was a small number of politically powerful people making unilateral decisions using vast government funds.
How The US Government Sold Us On The Suburbs https://www.youtube.com/watch?v=HmL6xIg-EJ0
A lot of people still do. The people that could afford to move to the suburbs were also people that could afford to move out of tenements in the city, too, generally.
OTOH, it allowed them to be farther from (and outside shared facilities like schools, etc., with) the tenements and the people stuck living in them. (Which, due to economics and outright, overt discrimination in both lending and things like restrictive real estate covenant, especially meant non-Whites.)
The point is that suburban sprawl was something that was ultimately lead by a small group of people, and was done without a lot of foresight, particularly about induced demand.
Of course many people chose the new suburbs. One of the problems with this style of building is that it is indeed very appealing when it's at it's initial lowest density state. The problem is that as people move into the suburbs and congestion begins to rise, suburban sprawl has no answer. Those wide open freeways that felt so fast and convenient initially become the bane of your commute or even doing simple errands.
Then people get frustrated, and believe the only answer is still that same default idea to return to that initial utopia: just expand the freeways even more! Except as we see today, that simply doesn't work, and the pattern just repeats.
The whole context could have been different, and that would have resulted in very different decision making from consumers, particularly today.
I'll use Portland as another example, as I'm both familiar with it and it's something of an outlier on these issues historically. All the cool neighborhoods that are desirable to live in now are where the old streetcar lines used to run. The entire shape of these neighborhoods is different, with high walkability, because of the remnants of that transit system.
The streetcar system here was deliberately deconstructed here, again, largely due to powerful special interests. The auto and tire industry worked quite hard on promoting buses as the new modern approach, a vision local politicians also helped promote.
Today Portland is reconstructing the street car system, as part of a larger overall transit plan. 100 years later we can see we wished we'd just kept and maintained the original track system, and designed our zoning around it.
That's really the core point I want to make: suburban sprawl is a mirage, that appeals initially. However it fundamentally cannot deal with congestion, which means sprawl only has three possible futures: Door one is continued cycles of expansion and induced demand where congestion just gets worse in the long term. Door two is a mode switch, to provide a meaningful transit option to relieve the pressure of congestion without freeway expansion. Door three is economic collapse, where the whole question is mooted because entire subdivisions become blighted places no one wants to live in.
A large portion of US suburbia is headed for door #3 atm over the next half century. Hopefully we can be smarter than we were half a century ago.
LA County is full of detached single family homes and undeveloped land.
If your talking about Assembly Row, that’s a “town center” development. Used to be a movie theater and a sea of parking lots. It doesn’t factor into Somerville’s style of urbanism.
But I live in Cambridge, MA and I can assure you Somerville would remain one of the densest cities in the union if every last building therein was lopped down to 3 stories by a giant lawnmower.
People have very naive explanations of why the US is so bad at density and urbanism. Canada and Australia, which are very similar to the US, have very dense cities like Montreal, Toronto, and Melbourne. Do they not have cars in Canada and Australia? If you want to find the real reason why all other developed countries have dense cities with good public transport, but the US doesn't, you need to look at what's different between the US and all other developed countries.
The U.S. doesn't want urbanism. People by and large do not want to live in dense urban areas. Some people like it, but it's not everyone's cup of tea. With the remote work revolution, it's going to become even less appealing to live in a dense urban area.
[1] https://www.bloomberg.com/news/articles/2015-04-17/a-pew-sur...).
https://www.wolframalpha.com/input/?i=population+density+tor...
Well…OK, the core of Toronto is pretty dense, but the megasprawl that is amalgamated Toronto is not that dense.
But good luck getting by in any non-major metro area in Canada without a car. Toronto, Montreal, Vancouver…maybe Ottawa…it's doable, but the second you get into the suburbs, or cottage country, it's Subaru time.
Even Vancouver—Canada’s densest major city with 5,493 people per square kilometre—ranks 13th out of 30, and is significantly less dense than San Francisco (7,171 people per square kilometre), a comparable west coast city. In Toronto, there are 4,457 people per square kilometre. In fact, Toronto’s population could triple and the city would still barely have the density of Brooklyn (14,541).
And crucially, Toronto’s population density is less than many other American cities including Philadelphia (4,512), Chicago (4,594) and Boston (5,376).
The transit systems in both cities operate across the entire metro area not just the "city" proper.
Vancouver Metro Area: 2,463,431 / 2,878km² = 856 persons per km²
San Francisco Metro Area: 4,729,484 / 9,128km² = 518 persons per km²
(Numbers from Wikipedia)
>Toronto’s population could triple and the city would still barely have the density of Brooklyn (14,541)
Brooklyn is not a city. It's a densely populated subsection of one.
>crucially, Toronto’s population density is less than many other American cities including Philadelphia (4,512), Chicago (4,594)
I'm not sure why you think Chicago and Philly are not comparable to Toronto despite being being only 1% and 3% more dense. They're effectively all the same density for the purpose of this discussion.
Remember it was founded in 1642. The older a city is, the more dense it is.
We do not have subways partly because it would be very hard to build them in LA's geology. Seismic activity as well things like natural tar pits.
Historically I don't think its right to say cars killed the density. LA was several cities that grew into each other. Part of the reason LA is so wide is because of white flight to the outskirts. Do you build transit for something like that?
What does "high" mean? I think this should be quantified, preferably by someone qualified.
Unqualified opinion: It seems to me that if we took all of the money spent on personal vehicles, time spent driving, fuel and emission costs, infrastructure maintenance, and costs to human happiness + safety, we would be able to provide a better transportation system.
The Chinese and Japanese rail networks would like a word.
I agree that most of the US has been designed explicitly around the car, but there is no need to be fatalistic about it. Most US cities (including LA) had good public transportation before WW2, and they could have it again if they wanted.
> The fundamental prerequisite for functional mass transit is a high population density.
Lot's of relatively low density places in Europe, Japan, etc. have very functional public mass transit. It's not universal, but there are good examples.
If you start with the assumption that approximately everyone has a car, you can choose build cities in ways that only support that...
In that same period, just about every large town and small city had a trolley line.
You don't need skyscrapers for mass transit. Just enough political will to keep it funded.
I don't think there's anything fundamental about density. Its about ridership and funding. Even LA could have functional mass transit if it had the will to fund 15-minute busses on a grid all day. A 25 cent tax on gas would do it. If even 2 people average rode each bus, it'd break even on traffic density as well?
Busses accelerate, decelerate, and turn differently than cars, so even if in terms of steady-state flow the two-riders = break-even argument is correct, it's not in city traffic.
Moreover, I'd like to see your work on operating and capital costs, especially accounting for road wear for which the usual approximate rule of thumb is that it varies with the fourth power of vehicle weight.
And remember, the City busses are running only every 15 minutes. So whatever effect they have on localized traffic, it's very intermittent.
But even if car people lived somewhere with the best public transport in the world, they would drive.
It’s not about technology or even reality. This conversation is a lot like trying to tell private health insurance users about single payer systems. Or private school users about federalized education. Or telling people with 4 children or 0 children how to live in a world where a thing like the sedan is designed for people with 2 children.
It is telling rich people you know better and also they are wrong, which is not an emotion you’re permitted to send.
The advantages of individual vehicles for motion around a city are diminished wait times and arbitrary point-to-point linkage. One is not constrained to traveling only on the public transit arteries plus foot-travel distance.
(The tradeoff of fossil fuel consumption is huge and should not be ignored, however).
But more dedicated bike infrastructure would be a huge help in general.
There are surely limits to everything and not everyone is into extreme sports, but even in the winter I really like cycling. Much, much more, than the virus infested public transport.
There is a saying, there is no such thing as bad weather, just bad clothing.
But most people probably consider a grey sky, where some rain drops may fall, their limit.
i was more pointing out that 2 relatively inexpensive changes (use scooters, convert parking to bike lanes) could dramatically change the balance of cars vs. public transport in those places (not everywhere). it doesn't have to require a wholesale change of our built environment to get benefits now (even for those dependent on cars).
One problem with buses is that paying the drivers is expensive. As a result buses need to be big to amortize the cost of the driver across lots of users. But big buses don't work well in low demand times of the day. The only option is to run them less frequently, which sucks for users. And big buses don't fit down small streets. And fewer larger buses means fewer routes. Self driving buses could solve all these problems.
I'd still prefer to walk or cycle, but I'm probably biased because I live in a warm, dry and flat part of the UK and don't have kids.
Buseses on wheels are more flexible than trams or whatever on tracks, and can route around a collision.
Rails in the road are a severe hazard for cyclists and somewhat hazardous for pedestrians.
Self-driving cars would enable self-driving taxis which would be superior to both private vehicles and public transport.
Imagine if there was no need to park and you could summon a vehicle of the appropriate size at will.
* It could be even worse. Imagine the morning rush - thousands of taxis have driven to peoples' homes and then driven them to their work/school. Then most will have to park until the next peak of demand in the evening. Either they park in the city, or they drive to an out-of-city parking lot, then drive back in to collect, then drive people home, then drive somewhere else to park until the morning rush. What would have been two vehicle journeys using a private car became six.
I'm glad you said this because it's a key point so many in this thread appear to miss: what works for one person or locale is not the solution for another, therefore cars aren't going away but this does not negate the legitimacy of bikes or trains either.
Public transit can't haul freight cross-country. Self-driving vehicles can.
As others have pointed out, self-driving vehicles could significantly cut the operating costs of public transit. This also seems like a major win for investing in public transit.
Freight should be hauled cross-country on rails, not on roads.
Freight cannot be efficiently moved between the railyard and final destinations on public transit, on foot, or on bicycles. Self-driving vehicles can do this.
I have always been in the camp that real dystopian future is tens of millions trapped in large cities instead of being free to live and travel where they want. I can never willingly place myself back into a large city.
there is no reason the concept of public transportation does not include self driving vehicles in many sizes. already this day we run separate transit vehicles for those with handicaps which make regular bus and train travel difficult or impossible to access. So why not smaller cars that can serve any one in need at any time regardless of where they are located.
one of the most common traits in any recently freed or booming economy is the personal vehicle and for good reason. it offers the means to travel where and when you want to not be trapped in one location.
Cars are extremely useful and are not going anywhere. They have problems like carbon emissions and being extremely dangerous but those will have to be solved while preserving the fundamental utility of a cars: They are individual vehicles that can go nearly anywhere and carry considerable cargo.
I agree, but I also think this is part of the problem.
Choosing to use public transportation for longer distances is a significant personal cost, while the benefit of lower costs in externalities is almost invisible.
Conversely, driving a car has a comparatively lower personal cost, essentially negative, since it increase freedom of movement. But the externalities are conveniently hidden by our economic and political systems.
It’s cheaper to pollute our planet, than enduring the inconvenience of public transport.
But the only reason that is true, is the terrible economic systems we all live under.
End it there and stop trying to make decisions for other people.
I’ll easily do that when every single western country stops subsidizing gasoline and shoveling the consequences onto younger generations.
End it there, and stop defending old people destroying the very foundation that humanity depends on.
1.) Public transportation in the US is at best, tolerable. Usually far worse. Not just in terms of coverage and utility, but in terms of cleanliness and safety. People use it because they have no other choice that is affordable. Get it to a level of quality like Tokyo or Seoul and more people would use it because they actually like it. I doubt that will happen anytime soon due to cultural issues, unless people are willing to tolerate draconian levels of enforcement.
2.) Bicycles are awesome, but you have to contend with precipitation, sweat, and theft. E-bikes help somewhat with the sweat, but then you have "traditionalists" that are trying to fight tooth and nail to ban them, and they are even more inviting theft targets. And you can still work up quite a sweat on an e-bike. Rain is still a problem.
3.) Living close to city centers is abominably expensive. People can drive in, or take public transportation in. But then go back to #1 in regards to public transportation.
I think this is very dependent on where you live, in both DC and the Bay I think public transportation is good, but expensive. To me, the problem more appears to be an American cultural aversion to intermingling the working class and the "professional" class of people.
> 2.) Bicycles are awesome, but you have to contend with precipitation, sweat, and theft.
Every mode of transportation has some downsides. You could say 'Cars are awesome, but you have to contend with about ~40k people dying in crashes every year.'
> 3.) Living close to city centers is abominably expensive
Hm, maybe if we could densify cities by removing some of these large paved spaces that people are supposed to stay away from.
Then again, many Newyorkers think the subways in those cities are horrible and unusable just because they don't operate between ~1-5AM.
That's a design choice and it's a particularly bad one that's making your life less convenient, your body less healthy, your streets less safe, and your environment more polluted. When everybody drives you get congestion, which forces people to spread out more, which increases the reliance on cars...and so it goes.
As someone who commutes primarily via bicycle, I've always been mystified by colleagues and friends who choose to drive to work in the densely populated northeastern city while cycling through center city is faster, more fun, better for your health, etc. There are obvious times when driving a car will make more sense, e.g. transporting delicate flowers, but for the average commuter it breaks down to the fact that the car is just somewhat sacred in our society. It is an important aspect of the American identity and our vision of success in this country, and so here we are, innovating around cars with technologies like self-driving rather than looking forwards to new potential paradigms of transportation.
This is just in regards to the average commuter, I'm sure that self-driving does have its place in like, moving 40' containers in low-density areas.
The idea of hopping from a subway to a bus to another bus just to get to a doctor's appointment sounds awful. And if I had to take 4 young kids along with me for the trip, double awful.
For the denser cities - my wife and I lived in SF for close to five years. As our family grew, and the city space grew more populous, parking spots were really rare. Getting to a pediatric appointment was misery. We knew the above scenario was our future. And it sounded completely miserable. So we moved to a spread out, suburban city.
I want to go directly to the location and start the errand as soon as possible. I hate waiting or slow buses or waiting for people to get on, get off. I just want to be able to go.
Nonetheless, our cities must eventually become crabs. So we need to have a plan for how we're going to make that transformation gradually, as the cities grow and evolve.
I’m also very happy to let my car drive for long boring stretches on long trips. I find I arrive at my destination much more relaxed. And yes, I keep my hands on the wheel and eyes watching around me. The car does the stupid stuff like keeping the car in the middle of the land and watching for other cars all around.
Public transit is high throughput (and can be low latency if you are lucky and stay on the same link) but as soon as you have to make a transfer (bus to train or bus to bus) the latency immediately spikes up. And the latency pretty much is unbearable as soon as you move to and from a non-central location (suburb to suburb) or not during peak hours.
To further use the computer network analogy, the last mile is generally also a problem. Doesn't matter if most of the network is fiber, if my house can only get a twisted pair the network is going to be bad. Same thing if the bus station or train station is 2 miles from my house; it's like it didn't exist.
The most significant is that we spend a huge amount of inner city land on parking. We could build denser without worrying about that. A fair amount of inner city traffic is people cruising around looking for parking too. Also, outlying rail or subway stops should really have a lot of development around them but often in the US they're surrounded by parking lots for people commuting in to work. Reducing the need for parking really helps in this case.
Also, people currently have to buy a car big enough for their greatest needs. You don't need a large a car when commuting in to work as when you're taking a road trip with your family but if you're thinking of buying a car you need one to do both. So hopefully we see the average car on the road being much smaller than they are now.
The experience is horrible almost every day. Between being crammed in the cars, the mentally unstable accosting you, the lack of cleanliness and the occasional assault I'm simply not willing to deal with that continuously. I put up with it because it was cheaper than owning a car in NYC at a time where I needed to make that trade off but I don't need to make that trade off any more.
This is with the MTA receiving billions of dollars and subsidies. Give us a system like Japan or even some of the light rail like I experienced in France and I can see people moving towards public transport. Continue to give us systems like the New York subway and I don't see cars going away.
This article makes largely the same point -- the way we do public transport in the US is flawed and technology can help us make it less so, but it has to have buy in from our political systems and the people. As long as institutions like the MTA have a say we'll only see mismanaged and rapidly aging systems.
Tokyo was in a league of its own (2 weeks experience). London Tube was mostly great, with some minor flaws here and there (9 months experience).
This is without even beginning to compare the reach and convenience of stations in each system.
This feels like looking at our ability to Lockdown for COVID. Even when we attempted to fully close, we sucked at it, and got the worst of both worlds (economic crush + virus still spread). At some point, you have to look in the mirror and stop trying to be something you’re not willing and/or able to be and find a new approach.
Three of the four of those are things you don't hear people complaining about with the transit system in Japan because of the culture differences in the USA.
How can that be fixed, here?
It corrodes all aspects of society when a sizable population will place themselves above all else. The ironic part is that except for a thin slice of the wealthy it doesn't even benefit the people who espouse it the most.
I'm highly individualist. I stay inside, wear a mask everywhere I go and listed to experts like Dr.Fauci on the subject of Covid. That's just one example.
There's a large amount of the population that lack empathy and are willfully ignorant. Individualism may be the cover word but what you actually mean is "hyper ignorance and selfishness"
1) Japan has scalably solved mental illness to the point that, in expectation, you will never be near someone still suffering therefrom.
2) Japanese law enforcement doesn't f--- around when it comes to train etiquette, and even so the most dangerous people don't mess with riders.
Whenever this gets suggested, there is a loud outcry from people that say they would gladly put up with the filth, unreliability, and occasional assault/rape/murder if it meant 24hour service. Usually arguments put forth is that there are always people (including lower income night shift workers) that must have the subway at wee hours of the night/morning. I'm pretty sure Seoul and Tokyo also have night shift workers and they manage to do just fine using buses or other alternate transportation. If anything Seoul and Tokyo have an even more notoriously hardcore "city never sleeps" workaholic culture than NYC.
They also routinely upgrade their public transport, instead of NYC which incurs massive costs as it keeps 200 yr old infrastructure barely alive.
It is also unreasonably cheap. $127 a month is lower than the cost of insurance for many.
So just one car's insurance is worth two monthly transit passes.
I regardless heavily dislike driving in NYC and would rather take Subway or Express Bus to commute.
Is it cultural? Or is it a matter of political/economic policy? I have no idea, but I'd kill to be able to bring it to the US.
I'd happily use the public transportation if it was anything closer to Tokyo, Seoul, Singapore, Hong Kong, or many other cities outside the US.
Given this, I don't really know why he mentions the challenges involved in self-driving and his skepticism thereof -- it seems to me that his autonomous "rovers" and "software trains" face basically all of the same challenges as a self-driving sedan would.
Overall I applaud the imagination that the author applies, bringing together the threads of autonomous vehicles, micromobility, and lingering anxieties from COVID into a set of bold claims about the future of urban transit.
What I am curious about is whether the future he paints ("micro-sprawl"; "walking is passé") is one people find compelling and desirable. I think he's basically right that the underlying market forces will lead us there in the next couple of decades absent strong action from municipal governments, but is this the right way to model cities of the future, or are we sleepwalking into another Robert Moses style disastrous reimagining of the urban fabric? Personally I'm undecided.
How so? It seems pretty evident to me that having separate, dedicated lanes should make it a lot easier. What I'd say is by far the biggest concern, human-driven traffic, is just completely eliminated with that.
Self driving supervised trains can drive on non-ETS (not fully automated tracking and safety) rails too with additional precautions. Lines where trains are driven by humans, and where some safeties are manually operated, as long as the presence of those safeties and cars is programmed. These behave then more like driver assist and automated safety.
Missing from this conjecture is a synthesis with car and bike sharing.
There will also be massive land use, zoning changes when there's 90% fewer personal cars. Unused capacity (parking, roads) freed up and repurposed.
The role for autonomous driving for moving people (vs turking cargo) will look very different when our cities look more like medieval villas and college campuses.
I still cannot even guess if the future will be more concentrated or more sprawl. So I've provisionally decided "both", but very different from today.
Jokes aside, we're at the point where everyone who wants linux on the desktop can get linux on the desktop without having to struggle. Big, known enterprise-friendly companies like Dell and Lenovo will sell you their flagship and business laptops and desktops running Ubuntu, and it's not uncommon for engineers at the 8000+ not-a-software-company I work at to casually have a Linux machine instead of a Mac or PC. It's not 2010, where you had to be willing to navigate UEFI and "will my broadcom wireless work?" to weed out the casuals from the true-believers.
Messing around with chromium versions and netflix's plugin was not fun when you just wanted to watch a show.
It's not polished, any distro.
But Linux server is GOAT.
edit: I have to say that I don't use netflix or play video games.
I'm really sick of this trope. I have been using Linux on the desktop almost exclusively for 15 years. Yes, I still have to fix weird issues on occasion . . . about as frequently as I have to fix weird issues on my wife's Windows machine.
Statscounter shows Linux installs about on par with ChromeOS installs [https://gs.statcounter.com/os-market-share/desktop/worldwide]. Both are beaten out by the "Unknown" category; there are literally more people running we-can't-tell-what OS than a Linux-derivative OS.
I think these days, the roadblock isn't reliability; it's Linux being off-mainstream. Network effect, essentially. Most of the (non-Internet) stuff the average person hears about in tech media and by word-of-mouth is not guaranteed to be available on Linux and is basically always guaranteed to be available on Windows, MacOSX, or both. Linux archs have done a decent job of solving the "Can I see my screen with the latest graphics card" problem but are still behind the curve on the "Can I buy Photoshop off the shelf at Best Buy and run it on my computer" curve.
Or I can use GIMP for free.
You are correct that Linux on the desktop is not mainstream and that is largely down to network effects, but I don't understand the next logical step in the argument. How does that make Linux on the desktop "not ready" or "incomplete?" In my opinion (and I recognise that this is just an opinion), Linux has been a superior experience to Windows for a long time. All common problems have several robust solutions. Uncommon problems often also have a good solution (which isn't guaranteed on Windows either, at least not without spending a lot of money). The idea that Linux isn't "Desktop Ready" is a tired trope. Just because people don't, doesn't mean they can't. Linux doesn't have a marketing department (and I wouldn't want it to).
But again, that's the problem to solve. You can use GIMP for free; most people can't. It's a space-alien UI relative to the Photoshop they know. And while I love GIMP and think it's a fine drop-in replacement for 90% of use-cases, it's still not good enough to fly (not fast enough, not robust enough, not the same size of plugin ecosystem) for the 10% that are using Photoshop in high-volume professional work settings. Businesses love cheap, and if GIMP could be substituted, they'd have already forced substitution on their art-houses.
You can even probably get Photoshop running on your box with the right cocktail of emulators and libraries. Most people can't.
The problem of picking up software from any brick-and-mortar store, taking it home, and running it on my Linux machine has no good solutions. Alternatives that you and I know work just aren't palatable for the average user. The average user still can't.
I'm not sure this is, practically, a solvable problem; I'm just identifying that it is the problem for adoption of Linux on desktop outside of the wonks like us that are willing to learn a lot of computer stuff off the beaten path.
Modern desktop Linux is great, but it's a whole commerce-adoption model away from being part of the beaten path.
And I still find the "Linux on the Desktop" trope to be funny.
Until some really big accident happens in a situation where even a drunk driver wouldn't have messed up that way.
After that self driving will be limited to situations like while you are in a traffic jam or limited to 30km/h and having it's own track or similar.
Outlier incidents shouldn't matter.
This systems expect the user to shim in fastly if something goes wrong, which just can't work reliable given how human brains work.
Furthermore the situations which can make a self driving car crash are not the same which make a human driver crash.
Such crashes can be pretty arbitrary. Random things like advertisement on a bus has the potential to trick of the self driving car. Similar at least some systems are not very good at handling situations where the lanes are unexpected changed e.g. around a construction site. I had been in more than one situation where the lanes markings what completely messed up and only common sense was left to prevent a potential dangerous situation. This systems don't have common sense.
There already was at least on deadly incident where a Tesla crashed for no apparent reason (blaming the dead driver for not taking over fast enough which again isn't realistic at all, through legal maybe valid).
Tesla is supposedly the leader wrt. self driving cars, but people where easily able to trick that car to drive into the lane for the opposite direction and it has since a while one of the worst emergency brake assistants.
The problem boils down to self driving cars not being good at "unusual" situations but situations which are unusual for the self driving car are in practice pretty common given how much people drive cars and they are not necessary recognizable as such by humans.
Another problem is that traffic's in different countries have different rules and dynamics a self driving system working well for idk. the US might not at all be safe to use in idk. insert arbitrary EU country here even if trained for that country.
Humans can normally adapt to a degree in the fly, which makes them not have a crash, or less bad crashes in some situations. But the self driving car has no such common sense and might not even realize it's in a situation it's not trained for.
I see a lot of potential to make traffic more safe.(non stuff driving but assistant systems, self driving in traffic jams, slow self driving buses in car reduced inner cities, on highway self driving truck convoys, etc.).
But at least on the close future I'm not convinced "general purpose"/"everywhere any speed" self driving cars are more safe.
I mean we also shouldn't forget that a lot of deadly accidents happen because of negligence, so while self driving cars still require a driver hand over, they still all happen because the driver is as negligent (or might intentionally not use the self driving feature to get faster to work etc.).
Edit: Swipy keyboard mess ups.
The right metric is simply a much lower accident and fatality rates for self-driving cars.
This flexibility reduces the challenge for self-driving cars since they are far from understanding situations at the level people potentially can, but still results in a safety win.
Using Autopilot is in general much safer even though it does from time to time make fatal mistakes that human would not do. (Like when it fails to recognise a truck and tries to drive through it[1].)
Of course AP is a very different thing to FSD, so this doesn't speak anything about current nor future FSD safety.
[0] Autopilot safety comparison https://www.reddit.com/r/teslamotors/comments/jim86m/i_graph...
[1] Accident https://electrek.co/2019/03/01/tesla-driver-crash-truck-trai...
That is why Tesla is not hyper focused on one city and is not building a team of exclusively AI experts, rather build a tool that other people can use to configure the specifics.
Tesla already are able to read signs in many languages and know about lots of different country specific rules.
> Tesla is supposedly the leader wrt. self driving cars, but people where easily able to trick that car to drive into the lane for the opposite direction and it has since a while one of the worst emergency brake assistants.
'Tricking' the car is kind of nonsense, you can also trick people into things pretty easily.
I have seen a number of reports where Tesla AHAD got the top marks and I have seen many videos of Teslas breaking when the driver was not paying attention. What are you basing this analysis on?
>“It’s almost getting to a point where I can go from my house to work with no interventions, despite going through construction and widely varying situations,” he said.
>“So this is why I’m very confident about full self-driving functionality being complete by the end of this year. It’s because I’m literally driving it.”
Though reading that it sounds more like the car can drive from A to B on it's own sometimes rather than with any reliability.
I think many users imagine being chauffered around while they read a tablet, potentially not even needing a drivers license.
Meanwhile I think realists and Elon Musk expect a car that is doing most of the menial parts of driving while a licensed human driver is still hands on but not doing much work.
We're close to that realist scenario in many ways for some areas - though I'm sure we'll see a spike in tragedies once it first goes out and people attempt to treat it like scenario one.
Scenario one is a very long ways away from being universal.
The term to use is driverless. Gives the correct idea of what users imagine. All thsi "self-driving" or FSD is just driver assistance given more capability with no regards to safety.
We've had self-driving train and airplane functionality for years. Meanwhile self-driving trains and planes are not closer.
The currency the federal reserve multiplies? That could collapse.
Since 2020, the dollar menu is gone, home and stock prices skyrocketed. In a year I've seen ~30% inflation. There's no reason for the inflation to stop there.
(Note, I have lots of investment, inflation may help me)
For stock info, look it up. Not sure where to get historical real estate prices, I'm just looking for a second property and everything is at least 10% higher than in February as you need to offer more than the selling price in this economy.
Not sure if you are looking for Data or a psychic. I do the former.
Will it happens tomorrow? In 20 years? In 5 months? It's just a feeling you have that isn't backed by data. I can't really ever disprove this because you're not really making any claim.
This is every FED conversation on HN. People totally believe hyperinflation is "just around the corner" but it never happens only standard normal levels of inflation always within FED projections.
What's your website BTW?
Also, why are you discounting the last 100 years of federal reserve policy? We obviously had greater than 10x inflation. Is 30% inflation in 8 months not considered hyperinflation?
My kids bank account is 30% less valuable now. I'm not going to fight over definitions that 30% inflation isn't enough to be considered hyperinflation.
It will be possible to buy large amounts of crude oil without owning USDs by 2026.
Still don't think it will happen.
CPI is not the best measure and isn’t always 100% accurate but it’s not going to be off by double digits.
> The index for food away from home continued to rise, increasing 0.6 percent in September. The index for limited service meals rose 0.9 percent in September, the largest increase in the history of the index, which dates to 1997.
> Despite the September decline, the food at home index increased 4.1 percent over the last 12 months. All six major grocery store food group indexes rose over that span, with increases ranging from 2.6 percent (cereals and bakery products) to 6.3 percent (meats, poultry, fish, and eggs). The index for food away from home rose 3.8 percent over the last year. The index for limited service meals increased 5.5 percent and the index for full service meals rose 2.8 percent over the last 12 months.
Of note:
> The index for used cars and trucks rose 6.7 percent in September, its largest monthly increase since February 1969.
But mechanically, yes. The Fed can't collapse while it's the controlling institution of the dollar. The dollar is the world's reserve currency. It's the denomination unit for paying IMF loans. It's the unit of money other countries shove under a big ol' mattress (with scary-looking guards around it) to backstop a collapse of their national currency. The dollar is so deeply baked into international economics that only global shocks can fundamentally shake its value as a currency.
If that changes, the Fed's power base becomes heavily disrupted, but most of us in the US won't notice because we'll be too busy riding in gangs with our fellow War Boyz hunting down the last reserves of Go Juice for our Death Rigz under a burning sky. ;)
I don't buy this argument. Yes, retro-fitting costs money, but even if one were to lay down guide wires next to every major road, the costs wouldn't be as insurmountable as it is being painted here. The cost of digging up a ditch next to the road just isn't that bad.
Furthermore, it's not just costs to bear, but also benefits to weigh. If an AV capable road might get you from city to suburb, the suburb might just pay for the retrofit themselves.
I’m also curious what the tolerances of these magical “wires alongside the road” have.
For example, Google Maps can already hyper-accurately predict traffic down to individual road sections simply by observing the position and speed of Android phones on the road.
Guide wires don't tell you whether the surface is flat. The surface could be damaged/potholed, and the guide wires would still be there.
The guide wires don't tell you whether the surface is clear. For that, you need to have vision, to tell you whether cars, pedestrians, obstructions, and toddlers about to run out into the road are there.
The guide wires don't tell you what the turn radius in the road is ahead - it only tells you where the road is right here. Although technically this could be achieved, it would make it much more complicated.
Ultimately, the car needs to build its own picture of where it is safe to drive that is reasonably resilient to adverse conditions and even malicious attack. Trusting road infrastructure for that I think is not resilient enough.
Ten, twenty thousand at most, I'm sure. ;)
Caltrain tried to roll their own implementation, got caught with their pants down, and iirc are now trying to save face by switching to one of the standard vendor developed systems
No it won't? Why would it? Plenty of companies are testing in other areas, and I haven't heard any suggest that they need special hardware on the roads themselves.
Yes, the Phoenix suburbs are the easiest place to start, but that doesn't mean you need special infrastructure in other places. The tech will keep improving, and I predict that this author will turn out to be incredibly wrong.
Sure this stuff works fine in Arizona where there is no inclement weather most of the time, and the roads are in perfect condition because freeze/thaw cycles don't cause excessive wear, but what about the rest of the country and the world?
What do adaptive cruise control systems have to do with full self-driving systems? You realize these aren't the same thing, right?
> Sure this stuff works fine in Arizona where there is no inclement weather most of the time, and the roads are in perfect condition because freeze/thaw cycles don't cause excessive wear, but what about the rest of the country and the world?
Again, what's the problem you're pointing to here? Arizona is easy mode, no doubt...and? We already know companies are also testing in areas with fouler weather, after all. Waymo even demonstrated how their sensors handle snowfall two years ago, and presumably they've improved further since then as they've continued development: https://www.engadget.com/2018-05-08-waymo-snow-navigation.ht...
It's absolutely bizarre how many HN commenters assume that the engineering departments in every self-driving car company are chock full of the biggest morons on the planet. "Oh ho ho, did they ever consider that other places have different weather than Arizona? Checkmate, tech companies."
> Plenty of companies are testing in other areas
For example, many have tested in the Detroit area that does indeed get snow.
E.g. suppose the car can only drive itself on highways in daytime in good weather. It still a great convenience feature to have if you don't have to drive for 80-90% of your commute (with the above conditions) and you only have to take over in the city.
They have to operate in weather. Rain, snow, dark. They have to adapt to novel conditions.
I put the self-driving car in the same category as the flying car and jet pack. This one just happens to have VC funding.
I don't think we'll see these things for another decade, and when that next decade arrives we'll be saying the same. Another decade.
But no safe full self-driving for a while.
Although I do feel like the money spent in the US so far would've been better spent setting up better public transit/train lines.
There's very little competition because there is not much to compete for, except in longer distance rail and bus.
You also need permission from the city to use public stops, consult routes and times, which is a barrier of entry. Not an insurmountable one and nobody prohibits you from running private stops or garages. You can even negotiate the use of bus lanes or tram lines.
The other barrier is much higher capital investment. Both bus drivers, busses and stops are not cheap, while the margin on a ticket is thin. (Compared to taxi fleet.) But it can be done even encouraging competition - see South Korea for an example of where it worked.
How much are talking? Waymo is 10 billion in the hole. Tesla spent maybe around that much developing autopilot. This is also not public money being spent.
A typical public transit system costs orders of magnitude more than that. Example: California's High Speed Rail project started at 40 billion and projections have already ballooned to 98 billion.
How could you even calculate this? By running massive experiments and seeing if fewer people die over a given period of time driving versus self-driving? Do you really think we can run such an experiment?
It is easier to do it if you consider 1% worst drivers, either older people with long reaction times, distracted drivers, drunk drivers etc. If you have a car that is obeying all traffic rules, it never goes above the speed limit and it has sensors that makes it brake before rear-ending other cars you are already better than the 1% limit. It is a simplistic approach, just giving an example of how 99% is not that hard to reach.
No they don't - a self driving car that knows when it can't drive is perfectly acceptable - not everyone can drive in extreme conditions reliably - doesn't mean they can't drive.
I don't agree, because of the grey area between the car knowing it can't drive and believing it can, but in practice doing so poorly.
That depends on your perspective. I think most of us just wants it to be more reliable than a human driver that's not drunk or sleep deprived. But the liability challenges could mean you'll want more "perfect" reliability.
> They have to operate in weather. Rain, snow, dark. They have to adapt to novel conditions.
Seems like the solutions out there are already getting pretty good at handling the kind of rain, snow and darkness you'd see in most urban/suburban areas. Maybe you'll "never" be able to drive autonomously to your cabin in the woods, but so what?
> I don't think we'll see these things for another decade, and when that next decade arrives we'll be saying the same. Another decade.
What exactly do you mean by "seeing these things"? By some measures they're already here. The ones that are truly self driving are slow, and mostly on static routes. The ones that drive on highways need to be monitored by a human, but they're doing pretty well. I think you'll see these different solutions just slowly become better and better. I'd be very surprised if you couldn't hail a fully autonomous robotaxi in many urban areas in a decade. Maybe it can't drive it anywhere, maybe you can't use it in a snowstorm, but it'll be a case of "good enough".
That is a seriously low bar, and in no way do I think the majority of people feel that driving "better than a drunk" is good enough. In fact, I think it will be the opposite: people will want it to be almost flawless before it's accepted.
>and in no way do I think the majority of people feel that driving "better than a drunk" is good enough
If that the case something can be done such as smart marketing/social media campaign or something like that to change people perception.
> more reliable than a human driver that's not drunk or sleep deprived
emphasis mine.
No thank you. Almost all the humans I share the road with are neither drunk nor sleep deprived. Putting robotic drivers on the road that make daylight driving as dangerous as 1am driving is not my idea of progress. It'd be a bloodbath.
Most of the roads may be clear, but “all” of them never are. Shaded areas are the worst, and can remain iced over for months at a time. Ice, especially intermittent ice, is bad. Slush, deep snow, and packed snow can be miles worse.
Winter driving, as any Californian who has migrated north will tell you, is a dramatically different skill than driving on the roads of California or Arizona. It’s non-deterministic - turning the wheel doesn’t always turn the car. The car will move sideways on the road with no steering input. The speed of the wheels is completely divorced from the speed of the car.
Flying cars and jet packs are limited by physics (specifically energy density) and progress in this area is almost non-existent. We barely got batteries to be vaguely competitive with internal combustion.
Self-driving cars are limited by software and hardware accelerators and this field is making real progress.
I remember a friend saying my kids would never need to learn how to drive. My eldest started her driving lessons last month.
People also posted this about 3D TVs and Cuecats.
Self driving vehicles will save $250+B/year in labor costs in the US alone - this is more impactful than the cotton gin and rivals the impact of refrigeration. 3D TVs and cuecat are 3-6 orders of magnitude smaller impact
Finally, it's worth mentioning that cuecat became QR codes which took off after getting built into cellphone cameras. 3D television didn't take off but 4K did i.e. consumer demand existed for richer TV images just not specifically 3D. Projectors remain niche as well even though they promise to take TV from 80" to 130+" - if ultra short throw (UST) projectors come down in price, I could see them going mainstream, especially in tandem with 4K/8K.
So maybe all new vehicles will have and old ones will be retrofitted with corner reflectors and infrared beacons or some such, and roads will have markers for AI cars, regularly checked and maintained too (by robots). Digital maps will be public domain. Clothes and bikes etc will get AI visibility signs integrated too.
In the US, if one pedestrian is killed by a self-driving car, the losses are unacceptable. This is true even if thousands of lives are saved.
In China, the opposite is true. If thousands of lives are saved, and one pedestrian dies, the self-driving car is a success.
China will be the first country to see widespread deployment of self-driving cars for this reason.
In the US, getting approval to construct a 10'x10' dog park requires moving heaven and earth to get permits signed off to even start planning, and then come the lawsuits.
Mmrright. Imagine a future where anytime public infrastructure stops working you have to contact Google's tech support. Just think of how that turns out today, when Google's algos decide someone's app is violating TOS or something similar, like the news that comes up every once in a while on HN, where the victims are met with a wall of automated messages that explain nothing.
"If you want an image of the future, picture a human trying to get through a bot to a human operator for ever".
Cars travel at a high rate of speed and a 1% error can easily be fatal. Either by hitting a pedestrian or someone on a bike, or by the car failing to acknowledge something and crashing itself head on into something.
Because cars travel in a highly dynamic environment this 1% error and the fatality risk associated with it is very high, so you need to have something that is 99.99999% safe which is extremely challenging because of the dynamic environment.
Meanwhile, if you look at automated train and rail systems. Because the environment is 99% controlled and less complex a 1% error is actually much smaller, because the environment is so controlled.
The high rate of speed, coupled with fatality risk, coupled with dynamic environments, is why this is such a challenging problem.
The car has created more problems than it has solved, what with CO2 emissions, brake and tire dust, and the incredible amount of car crash and pedestrian deaths.
Electric and autonomous cars would be positive steps forward to solving these problems, but are fundamentally still doing the wrong thing better.
The technology to solve all of our transportation problems already exists, and often has existed since the 19th century. Autonomous rapid transit has existed since the 1980s.
Compact towns with walkable streets, safe cycling lanes and high quality public transportation are all more effective and more affordable approaches to how we primarily move about our world.
Surely there's lessons there worth exploring?
Just say "parents".
https://news.un.org/en/story/2020/09/1071502
https://www.npr.org/2020/05/09/853073274/women-bear-the-brun...
In the US at least (and anecdotally), the driving factor appears to be that in a mixed-gender two-adult household, it is basically assumed that if someone has to step back from their job to watch kids, it's the woman---either because she makes less money than her male partner or because businesses are less flexible with granting men off-time for parenting than they are granting women such time. These factors aren't universally true, but I suspect they're on-average true enough to be driving the visible stats.