The problem is kind of similar to people who insist on some overly complicated microservice architecture when a monolith would be a much better fit. I actually hope that self driving technology stagnates, at least until we can start designing cities for people and not just cars.
I live near where the accident occurred and there is definitely sufficient population to support alternatives to driving.
If, let’s say ubers and taxis paid a tax when they rode around without passengers and that tax specifically went to road upkeep and construction, that would close the gap. But this takes policy change, and engineers not only cannot independently make that fix in a democracy; they think they know better than democracy because they know how to program a car in a simulated road in abstraction.
So, not only is the focus often misplaced towards IT solutions, those IT solutions often accelerate the underlying problem. Whereas there are policy solutions but they require smart young politicians and just flat getting out the vote and encouraging optimism for progress through an imperfect but best available democratic system.
The amazingly fast capabilities of modern IT should not be unfairly compared to the more difficult democratic process; leading us to lose faith in the more equitable practice of democracy. Let’s be humble enough to know when we personally are not subject matter experts.
1) How expensive and disruptive to build would "good transit" actually be?
2) How much do people appreciate more direct point-to-point transportation?
This incident occured in a town of 51,000 people with a density of 1,700/sq mile, that looks like a suburb of Salt Lake City which itself has 200,000 people with a density of 1,800/sq mile.
What's the cost and timetable for turning that into a transit-friendly city even if everyone wanted to have smaller homes in a presumably-more-dense footprint?
On the other hand, self-driving cars would sit on top of existing infrastructure to enable even more personal privacy and land use. So even if people were 50/50 which way to go, the latter would likely be far cheaper.
If you go back 100 years and prevent cars from ever be mass-produced, yeah, American cities would've grown and suburbanized in a more British, rail-oriented way. But reversing that would be far, far harder.
Also, $100/hr to run a bus is probably a fine operating cost number, but you do have to buy the bus (capitalized cost) and take the depreciation hit just like cars.
https://www.npr.org/2022/07/19/1111765630/on-demand-shuttles...
15 minivans replaced 6 busses, at a cost of 1.6 million to 1.3 million per year - but with much better coverage than the busses.
Of course this equation will look very different in big cities where economies of scale from mass transit kick in， but hey, small cities need transit too.
We already have the solutions for the future of transport, but they’re not sexy like self-driving cars, and they don’t prop up the auto industry. Walking, bikes, buses, trams, metros, combined with more dense mixed-use zoning everywhere.
Public transit simply can't take you everywhere. Even when I e lived in areas with dense bus routes, how one would get to a stop was often a question. Buses are relatively slow and impede traffic, especially in larger cities.
More trains would be nice, but they share many issues with buses, including the fact that they become de facto homeless and mental health shelters on wheels. The only place where I have lived where this wasn't a problem was New Zealand, and that was 18 years ago, so I don't know if that's still the case. I always give up on public transit after seeing enough acts of human degradation and fights breaking out. Let's not forget the smell.
Cars and self driving tech will always have an edge over public transit. We might as well not waste too much time on public transit, but make sure it is adequate while making car-driving more sustainable.
In fact, I would rather prioritize making cities bikeable and walkable more expanding bus service. For some cities it's pretty much too late, but for ones that are kind of spread out there is no reason why there can't be dedicated bikeways.
Cars have an edge on transit because orders of magnitude more money is spent on the road system and on the vehicles than is spent on PT. Despite that, PT tends to be faster in cities, where it's simply not possible to jam more cars in (it turns out only carrying one person per car is a truly massive capacity constraint.
Cars will probably always be better in low density areas, as it's not reasonable to serve all areas with reasonable quality of service, but in any kind of reasonable high density environment properly separated PT (ie. not lumped in with car traffic) will have huge capacity and speed advantages.
In fact, looking at many of these public transit vs. car threads on HN, on the topic of safety, I feel transit advocates are not understanding the issue properly.
Most people that dislike using public transit in the US are not afraid that the train or bus will derail or get into an accident. The odds of that happening are definitely tiny.
What they're afraid of is some psycho on the train/bus committing anything from nuisance harassment to outright murder.
It doesn't help that many of the people that most advocate public transit are the ones most resistant to making transit free from being terrorized by such potentially dangerous individuals.
Even outside of that psycho scenario, you also encounter a lot of benign, but otherwise annoying or otherwise less-than-desirable-to-be-around people whether it's someone with horrific body odor or someone blasting loud music or someone coughing their lungs up.
I never felt afraid, anxious, or even annoyed in any way getting on the subway in Seoul, Tokyo, or Hong Kong.
I always feel at least some degree anxiety and annoyance, and sometimes outright fear, getting on the subway in NYC.
You were supposed to describe public transit, not a rest stop on an Interstate Highway in the United States :-)
But seriously: the things you're describing are civic issues, not issues with public transportation. The only reason you see them less in your car is because the activation energy for them affecting you is much higher (specifically, instances of road rage, DWI, etc. culminating in someone colliding with you). But that doesn't actually make them less frequent; it just converts the drunk guy shouting on the subway bench into the drunk guy who's about to T-bone you.
I agree these are civic issues, not issues with public transportation. However I always get the impression that many ardent public transit advocates are the ones who are also resistant to cleaning up public transit so that it is more attractive.
Yes, let's definitely work towards providing mental health solutions, but let's also not let the psycho claim an entire subway car because "he has every right to be there" or "it's inhumane to forcibly remove him" and still expect the general public to delight in giving up their cars and take the subway instead.
They both are and they aren't. It is a civic issue that is inherently worse for public transportation, especially those that have nearly no supervision like subways. I totally agree that the issue is a civic one that should be solved, but if we're not going to solve it, then it's a problem that will prevent public transportation from reaching its potential. Would you rather encounter a crazy person out in the open or be trapped in a steel box with them? That's essentially the decision people are making when they board public transit in a major American metro. If a person is smelly out in the open, at least there's a breeze that can blow the odor away, but on a bus or train you are trapped in the recirculating smell cloud. Out in the open, one might have to pass by such things, but being on public transit can mean experiencing them for hours.
> But that doesn't actually make them less frequent; it just converts the drunk guy shouting on the subway bench into the drunk guy who's about to T-bone you.
That's a bit of a false equivalence. While there's indeed some mathematical soundness to what you're saying, I'd argue the vast majority of trouble makers (for lack of a better term) on public transit aren't there out of commuting necessity. They hang out on public transit because it's out of the elements and they are unlikely to be harassed themselves. It's somewhere for them to go. If all public transit closed, they'd just be out on the curb or in front of a gas station.
There is an incentive for city officials to ignore the problem as much as they can get away with. They know that the vast majority of their constituents aren't taking public transit, so what better place to allow the unwanted to be off the streets?
That's why it is a public transit issue. It's a much greater challenge to solve civic problems at a higher level, but cities could actually hire security to be on their trains and actually enforce tickets. No matter what, it's an inferior daily experience to being in one's own car. You can never get rid of smells and the low-level bad behavior. Not without society changing itself.
I believe that people with this kind of mindset do not understand nor can understand the value of time.
Example, for me going to work using public transit takes ~35 minutes, that is 8 minutes of train then the rest on a bus. (Not including the waiting time for the train/bus to arrive for departure)
Using a car, it takes me roughly 12 minutes.
So.. I save 2*20 minutes per day by not using public transit. How much is that per week? Month?
Some may argue I should move closer to work, but that would mean a way higher mortgage and a crampy apartment vs. my house out in greener areas. I think the choice is fairly simple...
Or that I should switch jobs, but there are no companies around my living area which pays the same, and I really enjoy job and definitely my living standard.
So unless the city put a direct line with no stops between my suburb and the area I work in, a car makes my life less stressful and more efficient.
Absolutely! An hour on a train where you can do productive work is far better than 30+ minutes in a car where that's not realistic. So sure, self driving cars that do make it realistic to work while commuting are the ideal, but in the mean time, ensuring public transport is sufficently uncrowded and well-connected that travelers can realistically make productive use of their travel time makes a lot of sense.
Of course, we don't want to waste 2 hours of our day commuting, but this is the case to advocate for better public transportation that could take us quicker and more comfortably rather than believing that self-driving cars are the solution.
Meanwhile, huge amounts of money is spent on improving mass transit. It's fairly conventional wisdom that it's a good thing. (The money doesn't seem to go very far though, compared to the need.)
I think that many people(including me) do not belive in self driving cars. I "know" there will be no level 5 self driving car anytime soon so why would I be excited? All the current and "near future" self driving cars seem to have a steering wheel so they are "fake self driving cars" and I'm not interested in that. Why would I use a self driving car if I still have to keep my hands on the steering wheel? I would be more worried that the car could get into an accident by itself and I would be blamed.
Then the more you learn about the tech the more I get worried. The car is supposed to drive itself based on solved captcha or an army of people trying to label everything?? I don't want to be anywhere near a "fake self driving car".
Self driving cars are like those miracle cures for aging or cancer that seem to work only on rats or in lab specific conditions. One day they will work on humans but you never know the day.
So if a miracle happens and "real" self driving cars are developed I think all the people will go crazy after them just like for a miracle drug to reverse aging.
https://amp.cnn.com/cnn/2022/06/02/health/reverse-aging-life...
You're learning is a little bit off. Self driving cars are not a big truck. It's a series of tubes.
I think you're right: companies like Uber and Tesla are pushing self-driving cars, but there isn't really popular support for them.
Huge amounts are instead siphoned from public transit to shore up car makers' and suburb real estate profits.
In California, $10 billion so far has been spent on high speed rail. The new bus terminal in SF was $2 billion, though admittedly that includes a skyscraper and a park. The Bart extension to Milpitas cost another $2 billion.
In Manhattan the first phase of the second avenue line cost $4.5 billion and the second phase will cost $6 billion.
As I type this I'm in the New Orleans city center, where the noise eight floors up is constant and overwhelming. Like, had to pause the movie last night multiple times because street noise was drowning out dialog. In my suburb back home, things are very, very, very quiet. It is much easier to imagine public transit where I am now than where I normally live, but that wouldn't help me in the suburbs.
In theory, I love urban density. In theory, I hate parking minimums and zoning restrictions. In practice, like most Americans, I'm choosing to live in the land of parking minimums, zoning restrictions, and little to no public transit.
Viable public transit in metro areas is economically a tough sell, but culturally it's even tougher.
I think self-driving vehicles are mostly fraud, and that you'll get your wish about stagnation, but that doesn't mean public transit will step up in the gap. I think it will just mean we continue to sacrifice 35k-45k people every year and keep our cars.
To me self driving EV's appear a much more viable way to "solve" public transit everywhere beyond major metros. They'll drastically drop labor costs and are extremely flexible. No major infrastructure changes are needed compared to adding things like rail. No cities need to be redesigned in one fell swoop.
Self driving buses or shuttles may actually come first around planned retirement communities. Simple routes, good weather, with flexible demand.
But it doesn't solve the problem of using something sized for four people carrying one person.
Reason 1: people in the US aren’t judging transit vs car ownership on the axes you mentioned. They largely don’t care about climate change and efficient land use (at least not enough to change their own habits.) They care about comfort, control over their own surroundings and mobility, and yes, signaling status.
Reason 2: The way of life that previously made transit make sense… daily point-to-point travel at peak hours when driving would have been prohibitively expensive… is quickly going by the wayside. Even if I go into my office 1-2 days a week, it’s now more cost effective for me to drive, and climbing aboard a bus or train feels like a relic of the past that I’m no longer accustomed to because I don’t have to.
It's not trivial to just redesign our suburbs. And the younger generation flock to big cities, not many seem interested to live in a small town that has potential for pedestrian centered design.
What exactly are you proposing?
I'll be steering well clear of these things in the future.
Humans make mistakes.
What will it take before we indict Musk for reckless endangerment, and for fraud? What will it take for the NTSB and FTC to issue a recall, and require Tesla to openly refund all the money collected for it?
You can go on YT and step through self-crashing videos frame by frame.
The burden is on them to show it is safe before they turn it on.
A tangent; Tesla did show their self driving is safe. Doesn't mean it has no bugs.
Actually, scratch that: what matter is that the autopilot sounds cool and high tech. Given that, who cares about the casualties?
So not really apples to apples comparison. Unless FSD is as safe in middle of blizzard when positioning system is being jammed it fails for me...
However, that's not a fair comparison.
Tesla's "autopilot" only works on clear motorways with nothing "exciting" going on. It can't cope with sudden changes in conditions very well, and hands over to the human driver. It can't cope with two-lane roads with cars coming at it at all, never mind single-track country roads.
So, if you compare Tesla "autopilot" with cars driven in the same conditions as when "autopilot" is in use, you see that they're no safer or just a little bit worse.
TL;DR Tesla "autopilot" only works in conditions where cars don't crash anyway, but they are compared against cars driven in all possible conditions.
Autopilot (as I understand it) is a lane-following dynamic cruise control so should have detected the bike and slowed down on approach. Human beings have a weakness in their stereoscopic vision where because their eyes are not far apart they can’t accurately tell the speed of a narrow object (like a bike) if it’s travelling towards or away from them. It seems like this should not be a problem with a car (because you can put the cameras on either side of the vehicle for example), but I wonder whether Teslas detectors or software have weaknesses with narrow objects.
[1] And even though with good gear you can skid along the road relatively safely, other cars are bound to hit you.
I'm by no means a fan of self-driving cars, but despite tragedies like this one, I wouldn't be surprised if overall they're much safer than human pilots even for cyclists.
It was on a highway not a regular road to be shared with cyclists.
edit to add as I dont think it was clear- the tesla's system should have been able to detect the motorcycle as it was on the highway operating normally, so imo this makes it all the more egregious.
Both Autopilot and FSD (Fools Self Driving) needs to be investigated urgently.
That was a literal quote from one owner, to me, in a previous thread here on HN.
Perhaps Tesla should have a proper driver monitoring system to determine that the driver has their eyes on the road at all times whilst having autopilot or FSD when driving?
In the article: [0]
> "Just because your vehicle may be equipped with driver-assisted settings or auto-pilot features, all the vehicles still require the operator of that vehicle to still be attentive and still be watching the road,"
So it's really no surprise why they are getting lots of investigations by regulators over their contraption.
There's only one investigation, not lots. And they're only investigating whether FSD is "exacerbating human factors or behavioral safety risks by undermining the effectiveness of the driver’s supervision". They aren't even saying FSD is bad, or should be recalled. They're only saying that it needs to not alleviate the driver from their responsibility to watch the road, and they haven't even determined if that's the case.
The car will constantly alert you if you do not move the wheel and then stop the vehicle if you don't respond. Your comments are more concerned trolling then anything meaning that'll help people
I wonder whether the UHP pulled data from the car to verify this, or just took the driver's word for it.
Ironically in these situations a more understated name would probably work better for PR. Headline of "Tesla on cruise control rear ends motorcycle on freeway, killing rider" gives a much stronger impression of it being the driver's fault.
From the origin of autopilot from its aviation history, its pretty clear that autopilot was not designed to prevent you from hitting things.
I can see some possibilities: - reduced visibility because of fog, or some other reasons - the motorcycle rider was driving under the influence, and made some sudden move that the Tesla autopilot was not trained to predict (e.g. aggressive cutting in front of the vehicle) - somewhat similar: the motorcycle rider lost control of his own vehicle for whatever reasons, and the Tesla rear-ended them while they were skidding on the pavement - the Tesla owner had overridden the autopilot speed limit factory setting by 20 mph more - Tesla was not on autopilot at all, and the owner is simply lying; maybe the owner was DUI - a case of road rage: the biker did something, the Tesla guy honked, the thing escalated, maybe both were a bit inebriated, and the Tesla guy rear-ends the biker not with the intention to kill, but just to "teach the guy a lesson"
I'm not trying to say Tesla is not at fault. In the first 4 cases I listed, Tesla is clearly culpable.
I just simply doubt the story is "biker riding normally, and suddenly a speeding Tesla rear-ends and kills him out of the blue". At 1:10 am on a nearly empty highway.
As to how the driver didn't see the motorcycle, if the car is driving for you why even pay attention? Especially at night when there are so few other drivers. The whole reason to use autopilot is to interact less with the vehicle. I know telsa says you have to pay attention, but the point of the feature is not to.
Tesla's Autopilot feature compels you to pay attention. If it doesn't detect your hands on the wheels applying slight pressure, the car will nag at you, first by flashing blue on the screen, then by sounding alarms, until it does feel slight pressure on the wheel, and if you ignore that, it'll slow the car to a stop.
> The whole reason to use autopilot is to interact less with the vehicle.
According to who? Tesla? I've yet to see any documentation saying what the point of Autopilot says. The official user manual for the feature doesn't even try to explain why the feature exists.
I've used the feature in my Tesla, and frankly, it's lane aware cruise control. That's it. It in no way allows you to do other stuff. If anything, I have to interact more with my car when I'm using Autopilot.
I also think there's more to this story than just an inattentive Tesla driver ignoring their Autopilot's warnings. Especially when it happened at 1:10AM on a stretch of highway that's straight and smooth, and likely had very few other vehicles.
I have to wonder if the motorcyclist had a non-functioning tail light. Then it would be hard for a human driver to see, much less computer vision.
