Hacker News Reader

Hacker News

Comment by tanvach | Hacker News Reader

Parent Full thread

tanvach

We recently started using Waymo more often:

- Quality of Lyft and Uber rides have gone down significantly.

- Consistently spacious, clean and quiet cars. You know what you'll get.

- AC always works and not up to the whim of the driver.

- No chatty driver to disturb our sleeping baby.

Negatives:

- Rides have usually 10% mark up over Lyft and Uber.

- Pick up and drop off tend to be a small walk from requested locations.

Forgot one more positive - you can choose a soothing music play list in the car and it automatically resumes in the next ride. Small but really nice detail when traveling with a baby.

View on HN

> You can choose a soothing music play list in the car and it automatically resumes in the next ride

Oh wow! There is a non zero chance that was implemented because of some feedback I provided as a trusted tester many months ago. I napped my son in them a lot when they were free and just spent my time thinking up things they should do and reporting them in app.

Man. My daughter's head would sometimes detach from her body and fly around the room by itself.

Strapping her down in a car seat and taking her for a drive usually rectified the situation.

I look forward to a future where this process can become more automated.

(/s, sorta)

My father said the same for me. To this day, sleeping in the car is the easiest sleep i get. It's so soothing. Which is problematic considering i'm the sole driver for my household hah.

Has anyone ever attempted putting a bed on actuators to emulate a driving vehicle?

There are many cribs that do this, here is the top of the line: $1000+ : SNOO

Ford even did an ad campaign around this concept some years ago https://media.ford.com/content/fordmedia/fna/us/en/news/2017...

Many car seats are removable and have carry handles so it is possible to hang them and have them moving like a pendulum also spinning. For safety use a strong strap (like for a hammock) and hang just an inch or few cm or so above the ground and make sure what it is hanging from can not topple or come crashing down.

I swear I've seen a thing for a car seat... But not for an adult.

It is automated. Has been for many decades. Hit up ebay and search for "swyngomatic." There are ones that are spring-powered and more modern ones use a couple of d-cells (or a wall wart if you're enterprising.)

That and a white noise machine (set it low, their ear canals are very straight so they're sensitive to noise, although babies get used to noise around them given time and sleep through it, which is more practical / useful.

You can also just do what people have done for tens of thousands of years: wrap your kid against your body and go about your business.

Putting your kid in a car (self driving or not) is putting them in the most dangerous situation they could possibly be in - car crashes are the number one cause of death for kids (in the US.)

Something every new parent should learn (which some parts of car rides mimic), learn about the 4th trimester. Then learn the 5 S’s to mimic the womb:

Swaddle - wrap the baby

Side/stomache - position like the womb

Shush - white noise

Sway/swing - walking or bouncing

Suck - a finger, a sucker, etc

These were a life saver for my son. Most nights it was just swaddle + sway, on bad nights it took all 5. I think the reason cars work for so many baby’s is that car seats, the rhythm, the white noise from the car, etc… these all do the same thing.

Our first is just out of this period and it really was pretty rough. It look us a month or so to come across the 5 S's and it made a world of difference.

We also somehow got very very lucky in that he didn't need to be transitioned out of the swaddle, he just grew to dislike it so one night we just stopped cold turkey and it was fine. We've had friends that have had to do the one arm in / one arm out, swaddle the legs, etc. and transition over days or weeks to get any sleep at all.

My son would always wriggle around until his hands were up next to his face (this is like 11 years ago). I started swaddling him in that position. We had an ultrasound picture that showed his hand in a similar position, so I guess it carried over.

Same, almost all his US pictures show him with hands next to his face and he sleeps like that most of the time now too.

I disagree about giving them anything to suck on for the night. IIRC there's even research that could lead to late speech or trouble with speech later on.

> Putting your kid in a car (self driving or not) is putting them in the most dangerous situation they could possibly be in - car crashes are the number one cause of death for kids (in the US.)

It would follow that since no children die each year playing in active volcanos that's the safest thing they could possibly do. Kindergartens should be filled with magma.

Serious preventable accidents happen at 60+ MPH. Generally a slow 20-45 mph (35 mph average) drive around the neighborhood is much safer as there are almost no "serious" accidents.

Source: an organization with a fleet installed some driver tracking gps's and looked at the data over a year or two.

Those stats are probably very different for babies, who are more fragile and could get seriously hurt even at lower speeds.

Where are the stats on infant motor deaths where the baby is secured correctly and car speed is <= 40mph?

Most car seats are improperly installed. If you have one, get its installation checked by a professional.

In the SF Bay Area, you can have CHP check it for free. The full-time job of the officer that checked ours is investigating accidents where kids in car seats died. Almost all the child fatalities around here are due to improper car seat installation.

We had it dangerously wrong. Before the appointment, I read the manual twice and spent something like 60 minutes installing it. The seat used the latch system with a top anchor, it was a new car and we had a top-ranked car seat. All the straps were connected up correctly. During installation, I sat on it with all of my weight, pushed against the ceiling of the car with my back, pulled the tightening strap until it felt like it'd break, and then pulled harder. The straps were still too loose. I'm not small.

Our last car seat had a failure mode where one of the latch straps would just self-release every 1000 miles or so. I'm not sure if it was user error or not. It happened twice.

After that experience, I think part of the certification process for car seats and cars should involve proving that the majority of people that have never installed a car seat (i.e., first time parents) can get it right on the first try without reading the manual. The test group should include people with physical impairments, and should include at least a dozen popular car models going back a decade. (Car certification should be the same, but with multiple car seat models.)

I found some NHTSA NCSA stats and posted them above.

- "where the baby is secured correctly" => "Restrained", <1 age group and 1-to-3 age group

- car speed is <= 40mph : NHTSA does not separate out accident speed, or separate residential roads from freeways

Apparently the NHTSA stats aren't different for babies (likely because babies are usually heavily restrained, and probably brought out in a car less e.g. don't need school runs). Below are some US NHTSA publications with detailed answers, but here's a loose paraphrase a) it's dangerous being a nonoccupant in a driveway: nontraffic backover/backing crashes caused ~460 deaths in 2008 (~300 nontraffic + 160 traffic), see [3]. b) reported fatalities per year age of child seem to be roughly constant, i.e. babies do not exhibit a higher rate of traffic deaths (probably because they're typically restrained in carseats). See [1] p5: Table 1. Passenger Vehicle Occupants Involved in Fatal Traffic Crashes, by Survival Status and Age Group, and Restraint Use, 2021. In fact, the peak is for children passengers 13+ or 15+, not babies. c) (These do not report crash speed, or separate freeway accidents from residential roads (posters here asking about "a drive around the neighborhood"), and you'd expect people would drive babies slightly slower.) NHTSA only separately reports nontraffic accidents (e.g. parking lots). Or non-crash deaths such as hyperthermia.

[1] https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/... "Traffic Safety Facts, 2021 Data, Children, DOT HS 813-456"

Key Findings • Of the 42,939 traffic fatalities in 2021 in the United States, 1,184 (3%) were children 14 and younger. • An estimated 162,298 children were injured in traffic crashes in 2021, a 17-percent increase from 139,058 in 2020. • Of the 26,325 passenger vehicle occupants killed in 2021 in traffic crashes, 863 (3%) were children. Of these 863 child passenger vehicle occupants killed in traffic crashes, restraint use was known for 769, of whom 308 (40%) were unrestrained. • Of the 1,184 children killed in traffic crashes, an estimated 294 (25%) were killed in alcohol-impaired-driving crashes in 2021.

[2] From NHTSA: NCSA (National Center for Statistics and Analysis) Motor Vehicle Traffic Crash Data Resource Page https://crashstats.nhtsa.dot.gov/#/

[3] "DOT HS 811 144, November 2008, FATALITIES AND INJURIES IN MOTOR VEHICLE BACKING CRASHES" https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/...

A backover is a crash which occurs when a driver reverses into and injures or kills a nonoccupant such as a pedestrian or a bicyclist. Backovers can occur either on a public roadway or not on a public roadway, i.e., in a driveway or in a parking lot. The former are called traffic backovers and the latter nontraffic backovers. There are also “other backing crashes” that are not backovers, i.e., they do not involve a pedestrian or other nonoccupant, that occur when, for example, a driver backs into a tree or pole or when a driver backs out of a driveway or parking space and is struck by another vehicle. Together, backover crashes and other backing crashes are referred to as backing crashes."

Look at the table on page iv.: Fatalities and Injuries, by Type of Crash, Type of Vehicle

> Man. My daughter's head would sometimes detach from her body and fly around the room by itself.

Is this satire, or am I missing something (going by the serious replies)? Not sure I would ever use this phrasing to describe anything serious - much less about my own child.

Search "Why is my baby crying bloody murder" aka witching hour.

If you've never experienced this, you are very lucky.

Ah, thanks - So satire it is. Thank goodness my kid didn’t scream like that as a baby. Although we’ve had our own share of unusual challenges - she was/is a loud crier and could go on for 20-40 mins in the middle of the night. Not murder screams but still loud enough that my Apple Watch would warn me of prolonged exposure to loud noise :-/

I mean: That part was of course satire. It's not like her head literally detached and flew around the room by itself: She was a real, human little kid (and a pretty awesome one at that), not something from some horror film full of practical effects and CGI, with disembodied heads flying around and screaming tirelessly like their very essence is being destroyed.

But it sure did feel like that was happening every now and then, despite our very best efforts and resources. And sometimes, something like a trip to the store ("HEY, DO WE NEED MORE EGGS AND MILK FOR TOMORROW?" "YOU KNOW WHAT? I THINK WE DO NEED MORE EGGS AND MILK FOR TOMORROW!" "OK! LET'S TAKE THE KID AND GO TO THE STORE!") was the the only thing that allowed for a return of sanity for anyone involved.

It was not the first option. We always employed all positive options first and had good support for finding new methods; we really did do pretty-OK at that time as non-robotic human parents, I think.

Going for a short drive was the last option before "Let her scream. It's hard work to make that much noise and she'll tire out eventually, and fall asleep [0]" would come into play after everything else has already failed.

[0]: We once got this advice from her doctor as a last-resort option. And maybe that was good advice or maybe that was terrible advice, but it was absolutely untenable advice for us when we were living in a stick-framed apartment building: Letting a kid scream like that for a prolonged period with a multiplicity of neighbors on the other sides of the walls is an easy road towards getting the police, CPS, and/or building management involved through no particular fault or misdeed of our own. And as a practical matter, when third parties become involved like that, now there are at least two problems (at least one of which is now an ongoing major problem) instead of just one relatively minor problem of a clearly-tired little human who chooses to scream instead of sleep.

I'm very glad that you never had the opportunity to experience this kind of thing. It's awful, and if it were easy to resolve I wouldn't have any of this to write about.

It’s been that way for several years.

This also just makes sense when you conceive these taxis not as "Some other person's car" (as they would be with Uber, or with a conventional private hire service) but as interchangeable units. Of course they should all share state and so it makes sense that resuming the music carries on from where it left off.

Ya I think it does. But tell that to any C level after sometime a company grow… They will look at you wondering which weird idea you talking about.

Maybe it’s the Silicon Valley but it’s cool they still care about details like this.

Happy to be corrected. Zero chance it was implemented due to my feedback.

I believe in you.

Several years is many months

You should patent the idea, and then sue anyone who independently comes up with it.

> Rides have usually 10% mark up over Lyft and Uber.

Waymo says they are a premium service like Uber Black because they have nice cars (Jaguar I-Pace), plus the novelty and safety of being driverless. They’re not trying to be competitive with Uber X or shared rides for in their current form.

How nice is the backseat of a Waymo compared to an Accord?

A little nicer but not something I'd pay for if it wasn't driverless. It's significantly more spacious than a Civic though.

It has been explained to me that cars are an extension of ones personality, a social signal, and a hobby. So for some apparently what kind one rides in is significant.

I'm pretty sure I don't get any "social signal" from using fancy taxis. I think you've got to own the vehicle for that to happen.

Much like buying a museum ticket to see a famous painting isn't comparable to owning the same painting.

Some museums do try to frame memberships as something that is part of one's identity. A museum in my city sells annual memberships with a physical card with nice artwork on it, discounts on tickets (including a discount for a friend), and an invitation to a monthly evening event at the museum that is rather quite nice.

I haven't seen Uber or Lyft effectively go down the route of having their service be a part of a user's identity. But Waymo could make an effort, if they wanted to: they use a distinctive car model, and they could offer incentives to make carpooling with Waymo cheaper (to try and make a user 'known' in their social circle for using the service).

If that were true, I don't know how you could explain the persistence of town car and limousine services.

Ironically, the actual Lincoln Town Car has been out of production for years.

Really? Aren't town cars nicer inside than regular cabs? Limousines tend more in the direction of conspicuous consumption, though I imagine they are also generally more comfortable/clean than less-expensive options. Limos are also not that expensive for some use cases: we once had to get 8 people from NYC to MD, and it turned out that renting a limo was cheaper than us all taking the train or renting a vehicle.

Likely that is a factor, but people communicate social status and identity in so many different ways, which often appear meaningless or even absurd to people outside their perceived peer group, that I think it would be a mistake to discount the idea. (I recall one person I used to know who certainly did have opinions about the kinds of cars she would and would not prefer to be seen arriving in, regardless of who was driving or who owned the vehicle.)

> generally more comfortable/clean than less-expensive options

Some limos have poor or blocked views of the outside which can trigger motion sickness. I also wonder about their side impact protection and how much testing that gets.

A rarely discussed negative to Waymo is that they drive slower than human drivers. Anecdotally they can be 10-25% slower than the rest of traffic, and it's not uncommon to see human drivers do unsafe moves to pass a Waymo.

> Anecdotally they can be 10-25% slower than the rest of traffic

I don't know anything about USA, but my understanding of other humans tells me that "Waymo is 10-25% slower than the rest of traffic" actually means "Waymo drives at the legal speed limit for that location, without surpassing it".

> "Waymo is 10-25% slower than the rest of traffic" actually means "Waymo drives at the legal speed limit for that location...

Probably 100% safer too

Speed is the most dangerous aspect of motor vehicles

There is approximately 0% chance that the speed limit decreed by faceless bureaucrats is the optimal speed for the current circumstances on a given road.

It has always seemed strange to me that restrictive speed limits persist in a supposedly democratic nation despite the functionally unanimous opinion of the citizenry, expressed through actual driving behavior, that the legally prescribed safety margins are excessive. I doubt you could find any other issue on which Americans of varying political persuasions would demonstrate such a high level of practical agreement.

In comment threads like these, one invariably views vigorous venting of virtuous vitriol versus the vice of velocity, but simply getting out on the road and having a look around - pretty much anywhere in the country - will show that such opinions must either be hypocritical or held by a small minority.

People are bad at risk management. I see oodles of people making wild decisions in their cars (sweeping across 4 lanes at once, backing up on the highway shoulder to return to a missed exit). The norm for trailing the car in front of you is also way way too short based on real physical limitations of braking.

Speed limits can also exist for reasons beyond the safety of drivers. There are roads around me that are wide, straight, and empty enough to comfortably drive 40mph but are right next to a school. The school-zone limit exists to provide extra safety for children who might foolishly step into the road very suddenly.

One possibility is that speed limits are set so drivers are safe/efficient at the maximum observed traffic density.

If the maximum traffic density doesn't happen often, then people can understandably be annoyed when they can safely drive faster most of the time.

You're the one calling them faceless bureaucrats. Most speed limits are set in place by a mix of safety boards and urban planners, who do take these limits into account. Explain to me what makes them faceless and detached from the situation?

> Explain to me what makes them faceless and detached from the situation?

The fact that they’re not there.

The idea that the same speed is appropriate at 9am when kids are crossing the street going to school and 2am when there’s noone there and no cars parked on the side of the road, is insane beyond belief.

Speed limits are generally repressive and mostly just a tax collection scheme (though I strongly support most daytime city speed limits!)

What are you proposing? No speed limits outside cities in the day time? Or speed limits that change based on circumstances?

Don't we already have school zones which are 20mph when kids are around and higher otherwise?

That doesn’t make them “faceless bureaucrats”. It just means you don’t like speed limits because you know best

Because everyone drives 40mph on that main road that is marked 25mph, and there is no string of accidents or fatalities after decades of this.

> there is no string of accidents or fatalities after decades of this.

Ignoring reality

A lot of people killed on the road

Once you are doing 40 or 50 any accident you do have with a pedestrian is almost sure to be fatal. Lower speeds offer more of a chance.

All it takes is for a driverless car speeding to hit a pedestrian for the industry to be dead in its tracks.

Visit Texas where there are 85mph posted limits and people casually do 95.

I live in Texas and I would unironically feel safer driving the normal ~10-15mph above the speed limit that traffic is normally going than go exactly the speed limit, because I don't want everyone else flying around me recklessly because they all know that nobody goes exactly the legal limit here.

Yes! Speed difference is dangerous. So yes the difference in speed between vehicle and stationary objects it can hit (trees, poles, parker cars). On highways without a divider the danger is from incoming cars as that will double the impact speed. On highways with dividers having speed different from cars around you can be dangerous because it means either you are changing lanes to pass often else others are changing lanes often to get around you.

nothing compared to Italy, Slovenia or Germany haha

You're right! I've driving here as well, but one thing people don't realize is how big Texas is.

Texas vs EU: https://i.redd.it/jlqp9vv8flb31.jpg

Laws are necessarily very blunt things. The actual optimal speed limit will vary with car, driver, traffic, weather, time of day etc, would vary hugely along the road and wouldn't be in neat 5mph increments.

But making such dynamic limits would be an extremely expensive endeavour.

If you want to optimize for least deaths and injuries, thr optimal speed limit is 0 km/h. Everything else is a tradeoff between convenience and potential harm.

> Speed is the most dangerous aspect of motor vehicles

Speed difference is dangerous. So yes the difference in speed between vehicle and stationary objects it can hit (trees, poles, parker cars). On highways without a divider the danger is from incoming cars as that will double the impact speed. On highways with dividers having speed different from cars around you can be dangerous because it means you are changing lanes often else others are changing lanes to get around you.

> Speed is the most dangerous aspect of motor vehicles

Not in the way you think. If you're going 55 because it's the speed limit, but everyone else is going 70, you're the most likely driver to cause a collision. https://en.wikipedia.org/wiki/Solomon_curve

The mental hoops drivers go through to justify their unsafe behaviour...

Oh please.

Let's assume people are going at a speed that the road is designed to handle, since that's usually the case. In that situation, the behavior you're calling out as unsafe is simply "going faster". Now, in a technical sense that's true, but it's always true. Going faster reduces safety when you're above the speed limit. Going faster reduces safety when you're at the speed limit. Going faster reduces safety when you're below the speed limit.

If you reduce safety to black and white, then you should always go slower, until you're barely moving. If the speed limit is 40, that doesn't magically make it safe to go 40.

Black and white thinking leads to a bad conclusion, so let's add nuance back in. Speed is dangerous but roads are designed to let you go a certain speed while minimizing the danger. You're not being dangerous until you go over that speed. And that speed is almost always higher than the speed limit, especially in good weather.

Going over the speed limit is not itself a problem. It's going over the design speed of the road that's a problem.

> Let's assume people are going at a speed that the road is designed to handle

Let's not, because most of the speeding I see is from people who very mistakenly think they know what speed the road's designed to handle, because they don't give a crap about pedestrians, bicyclists, visibility, other vehicles, or really, anything but getting to the next red light faster.

Perhaps you drive in a utopia that's solely populated by traffic architects, but here in the real world, it's populated by normal people, many of whom are bad drivers, and bad judges of road safety, but are really good at hitting the gas.

Our high death rates reflect this.

You can take that logic and apply it to any traffic law. Blowing red lights? Fine as long as you can tell there's no cross traffic, right?

Laws were put in place because people are not in fact qualified to make these assessments.

The only nuance here is that you are not required to go the speed limit at all, you can and should go only as fast as the conditions allow.

Kinetic energy is proportional to square of velocity. It doesn't take large speed reduction to halve energies involved and thus the danger. Even not taking into account how crappy human perception and reflexes are.

I'm quite confused why I keep seeing posts like this downvoted in this thread. This post is nuanced and probably novel to most readers. Even seems correct to me.

Rear-ending someone going slower than you is not their fault.

>Rear-ending someone going slower than you is not their fault.

who gives a shit who is at fault when they're in an accident?

the goal is to avoid the event all together, not to have a scapegoat.

The type of people who see a "lane ends ahead" and immediately merge, blocking up traffic and wasting a half mile of empty lane. "I'm in my assigned place and I did what I was supposed to!!"

The wiki you link finds the theory to be biased and without solid backing data.

Not sure why did you link it.

Regardless, Waymo is under the spotlight and they can't risk the regulatory hit of openly breaking laws, even if it's a perfectly normal thing for a human to do.

Indeed. Much as buying a smaller car makes you a hazard to everyone nearby because they cannot see you as easily. The most sensible thing for everyone is ever larger and more heavily armored vehicles. Won't someone think of the children. `/s`

All that said. If you cannot maneuver safely around drivers going the speed limit then they are not the problem.

No it is not. Distracted or impaired driving is the most dangerous aspect.

> Probably 100% safer too

Not really. Because speed works in relative. I.e. if everyone drives at 200mph it is much safer than if half drive 100mph and other half drives 200mph. Granted you can't have things appearing at 200mph.

Speed is not the problem deceleration is. Which happens more often when some percent of people drives slower than others.

Decreasing speed limits has the counterintuitive effect of increasing vehicle to vehicle accidents, and intuitive effect of increasing ticket fees. Pedestrian safety isn't that imprcted since a portion of drivers drives above limit anyway.

So it's a bad driver then, because matching flow of traffic is more important and safer than stubbornly following the exact speed limit

It's a bad driver for adhering to the speed limits?

I'm having a hard time understanding how that makes sense. I understand that matching the flow of traffic is important, but more important than the speed limit? They're there for a reason.

If the speed limit is 30km/h and everyone is driving in 50km/h, you're saying you'll opt for driving at 50km/h?

Besides, if there is multiple lanes, you take the right lane and follow the speed limit. The ones who don't want to adhere to the speed limit, have the left lane(s).

And in the case of it only being one lane, you can literally decide the flow of traffic by adhering to the speed limit, and everyone behind you need to follow it, and once it stabilizes, that's now the flow of traffic, problem solved?

(Just as a disclaimer, I do sometimes drive above the limits myself, but limited only to the highways, never on road with a limit below 120km/h)

I’ve heard a lot of debate about this point here in Australia. In the state of NSW, people on their learner permit are restricted to driving no more than 90km/h (55 miles per hour). Arguably, this is safer because learner drivers haven’t learned to control their cars well yet.

Here in Vic (the next state over), people think that law is stupid because it’s apparently less safe having everyone on a freeway driving at different speeds. It makes changing lanes much more difficult and dangerous. Here learner drivers are expected and encouraged to match the speed of traffic (obviously obeying speed signs too).

I suspect vic is right on this one. It probably is more dangerous having different drivers driving at different speeds. At least, more dangerous to human drivers. But what do you expect Google to do about it? Make their entire fleet break the road rules? Their licence to drive their cars on the roads at all is on a trial basis. They’re being closely watched by everyone. Even if it’s potentially unsafe for other drivers who are speeding, sticking to the letter of the law is really the only choice they have here. Maybe in time all traffic will flow at the posted speed. And officials will finally feel comfortable raising the speed limit to match the speed everyone actually comfortably drives at on the roads.

> I understand that matching the flow of traffic is important, but more important than the speed limit?

Yes absolutely. Speed limits are cooked up in a room somewhere and do not reflect the actual conditions of the road. The real world situation you find yourself in when on the road is too dynamic to suggest a number on a sign should be the ultimate authority

It's very understood that when it's raining or snowing or foggy you should slow down because you drive to the conditions of the road because that's safest.

If everyone around you is speeding then that's also a condition of the road, which you need to adapt to

> problem solved?

The consensus from the folks writing laws appears to be that the problem is not in fact solved in that case. There are laws which apply independent of your speed that require you to give way when more than X number of cars are behind you constrained by your speed (X varies by jurisdiction).

Of course the response to this is usually "but if I am going the speed limit that other law cannot apply." Very much like the laws saying you must keep right unless passing someone. Everyone seems to have their own opinion on which laws they will follow, and which they can ignore.

driving 50 in a 30 is a 66% increase, not a 10%-25% increase like gp was suggesting. When speeding is more extreme it probably crosses a threshold where it's safer to drive the speed limit than to drive with the traffic. But when the difference is 10%-25% (which is accurate IME in the US) then you only have to experience it a few times to know it's less safe to drive the speed limit.

Flow of Traffic vs Speed Limit Dangers

Many believe the myth that as long as we’re going with the flow of traffic, we’re not doing anything wrong. It makes sense– everyone else is going the same speed, so why shouldn’t you? There’s no way a cop could pull you over if you’re just going with the speed of traffic. Wrong.

If you argue that you were just driving with the flow of traffic, then you are essentially admitting to speeding. Other people breaking the law does not justify you breaking the law.

So in following the law vs being safe you say “follow the law and make the roads more dangerous” because… why?

> Many believe the myth

Do they? Almost without exception the argument for going with the flow of traffic is that it is safer, not that it is the most legally defensible.

Uh that's just not true and I would not trust any driver who says this.

If its not true then why do many cars with self-driving build it in as a feature to allow breaking the speed limit (up to a limit, in my car it still won't go above 15mph above the limit) in order to more closely match the flow of traffic.

I think theres a bit of nuance. If the flow of traffic was all being crazy lunatics then yes I'd say so. But I would feel safer without other cars flying past me/around me because I decide to go 70mph when everyone else around you is going 80-85.

I speed on the freeway too, I just don't believe it's "safer". The margin between the traffic speed and the speed limit is just not generally large enough to be a hazard, and driving at a higher speed is just more risky and potentially lethal in every way.

Here in Texas I'd love for more people to stubbornly obey the speed limit to slow down the flow of traffic.

Having just returned from Texas, I have to say your problems in that regard are entirely self-inflicted. And frankly, probably by design. You have wide open freeways with ample space in every lane, absolutely inviting people to cruise at high speed. Want to dial it back? Make the lanes narrower. The vast majority of drivers choose their speed based on comfort, not what the sign says. That won't change, so the strategy for how you slow down traffic has to.

You're definitely right. In most of the US there's a speed limit and there's an understood amount over the speed limit you can still drive and cops won't bother you.

Same in Australia - though that amount varies by state! In NSW, everyone goes 10-15% over the limit. Just like California. In Vic, they have speed cameras everywhere and they’re crazy strict about it. People joke about speeding fines being part of their taxes for the year.

Apparently they even have entry and exit cameras on some freeways - and they take photos of everyone to check your average speed. Even if they don’t catch you in the act of speeding, if your average speed is over the limit they’ll still send you a fine.

The moral of the story: Don’t speed in Melbourne. It’s a meme that everyone who visits the state gets done at least once. Everyone in Sydney has a friend who’ll tell you the harrowing tale of their vacation speeding fines.

It varies by the state and town in the US. There are some that get a lot of funding from speeding tickets and the police stake it out. Then there are places like southern california where I’ve never seen a cop taking radar ever.

This is exactly what they do. They don’t drive slower, they drive the speed limit. I’ve taken a lot of Waymos and their priority is safety, for everyone.

The legal speed limit has nothing to do with how fast traffic moves on a road.

Roads can be designed to slow down traffic, but slapping a 25 mph speed limit sign up on an 80-100’ wide 4-lane road does not slow down traffic.

A road near me was reconfigured with a center turn lane and two lanes from four lanes and traffic slowed from 40 mph to 30 mph, the speed limit for this particular road.

https://www.strongtowns.org/journal/2021/8/6/the-key-to-slow...

https://www.smatstraffic.com/2021/08/30/traffic-calming/

10% slower on a 30-minute journey is 3 minutes. if i'm in the back of a taxi, i'm on my phone or reading a book or something. getting there 10% slower really just doesn't matter.

humans drive fast becuase when they're late it makes them feel like they're doing something to solve their problem, but the time savings are almost always inconsequential.

It's probably not even 3 minutes -- it depends at least as much on how many red lights you hit.

Hard disagree. Time savings from speeding comes from discontinuities like beating traffic lights.

Also 3 minutes is huge if you’re almost late to a 30 minute private lesson.

If you arrive on time you're already late. I think a lot of problems arise from people trying to optimise their journey time so they can leave at the last minute. Any small inconvenience then leads to road rage.

> discontinuities like beating traffic lights

Sometimes you beat them, sometimes you don't. It's not always in your favor, it averages out.

This is only true if they are random, which they aren't.

You can also pay attention to them, and learn the cycle lengths, timing patterns, and pedestrian patterns.

The average of gain and neutral should still be a gain.

If you are almost late and driving then you are late. Do better next time.

Is that a negative of Waymo or a negative of the aggressive/illegal way many humans drive? My assumption is the anecdotal speed difference you notice is Waymos actually following the speed limit, and I imagine Waymo isn't really looking to program their cars to break the law while they're trying to expand their ability to operate.

Not just humans, but Uber/Lyft drivers. It is very common for an Uber driver to perform unsafe and illegal maneuvers while driving me. I don't report them because I don't want to get them fired. Usually I don't even remark on it, unless I think they did it in a way that was especially unsafe.

If you don't hold people accountable for dangerous driving, 1)more people will drive dangerously while doing lyft/uber and 2)you are indirectly endangering vulnerable road users (people on foot and on bike, motorcycles, etc.)

Trust me, you'd feel very differently if you were on a bike regularly in the city. The lyft/uber drivers are the most dangerous drivers in the city. Have been since almost the very beginning of this ride "share" app crap.

Any time a driver did something dangerous with me in the car, they got reported and a one star rating with a comment explaining exactly why. If they did something particularly dangerous around a pedestrian or cyclist, I'd tell them the ride was over and to let me off immediately, and then call uber/lyft to report it to an agent instead of just clicking the "unsafe" button in the app.

You don't have a right to drive for a living.

If you do, you damn well should act like it is your living and be safe about it.

The whole fucking reason we have a massive problem with traffic safety in the US is because police and courts and legislators act like it's so necessary that we must endlessly tolerate people endangering others with their cars, and people get "hardship" exceptions where they're allowed to keep driving even after proving themselves to be a complete fucking menace, because "they need to get to their job" instead of "you knew you needed to be able to drive to get to your job and you still drove the wrong way down that street and hit someone? Sucks to be you."

This is the aspect that makes me most excited about Waymo and the like –the prospect of no longer having to tolerate terrible driving because removing someone's license is akin to an economic death sentence. Fully commercialized, self-driving should provide an economic alternative to humans driving themselves. And we should be able to leverage that to stop the worst drivers ever getting behind the wheel again.

So instead of penalizing unsafe drivers, we'll penalize all drivers by replacing them with robots?

I'm actually a Waymo fan but this particular argument doesn't really hold water.

Wait, how is offering driver less cabs penalizing all drivers?

The only motivation I see that drives self-driving taxi development is reduced labor costs. But that isn't different than what has happened since the industrial revolution.

This is a far-too-infrequently discussed benefit of self-driving cars. Right or wrong, I (and many others) have qualms about narc’ing on a human driver. I have no such qualms about doing so with a robot.

Uber/Lyft drivers won't improve their behavior on their own without feedback from the system, so please for everyone's safety please do report it.

You're not wrong but I do feel a bit guilty every time I give less than 5 stars. The rating system is so bad that you can't give any small feedback without serious negative financial consequences for the drivers.

Basically anything negative is rolled up into one signal no matter what the safety aspect is.

Uber/Lyft/Taxi drivers as sin-eater is one of their great values. Rider gets all the benefits of drugging like an asshole with none of the social/moral shame

Yes, most human drivers go above the speed limit. I agree that it can be unsafe, but wouldn't a self-driving car be safer at going above the speed limit than humans? I feel like it should be OK for it to go 5-10 miles over, especially if that's what the flow of traffic is.

I don't know if it's actually safer, but I feel like programming your self driving car to regularly break the speed limit would be a hard sell from a regulatory perspective. It's different in "self-driving" cars with drivers who can choose the speed of the vehicle. In this case, Waymo is programming the car to independently pick a speed, and making that programming decide to break the law seems like it could be a problem.

Probably a better solution is more reasonable speed limits and more consistent enforcement of those limits, but now I'm just engaging in wishful thinking.

If they're really safer than human drivers, like the Google-funded studies claim they are, then this seems like a positive rather than a negative. Perhaps we humans should be slowing down and driving more carefully?

But I'm not sure if we can trust these studies. I'd really like to see a completely independent evaluation, of how the safety of Waymo cars compares to human drivers, and to the safety other companies like Zoox.

I agree that an independent study would be very useful to get a better idea of what's actually happening.

That said, human drivers in the US are bad. We have the highest traffic crash rate among all developed countries by far. Pedestrian and cyclist crash rates have been increasing in the US, one of the only developed countries in the world that this is happening in. Much of the US remains opposed to automated traffic enforcement or speed governors of any kind. Privacy advocates use privacy as an excuse but given how much Americans care about their privacy in other aspects of their life, I'm doubtful. And anecdotally it's common for friends to talk about driving 10-15 mph over the speed limit, and we know speed is the leading predictor for severity of a crash. Most US drivers, especially of higher income classes, are probably well aware that they speed and break plenty of other traffic laws on a regular basis and don't want to reckon with that fact.

> And anecdotally it's common for friends to talk about driving 10-15 mph over the speed limit, and we know speed is the leading predictor for severity of a crash. Most US drivers, especially of higher income classes, are probably well aware that they speed and break plenty of other traffic laws on a regular basis and don't want to reckon with that fact.

Because 99 44/100 percent of the time, this is not a factor. The true threat is not someone going 10 over on the highway, it's the loons you see zooming ahead of everyone else going 20+ over, weaving like they're in a video game. Equally unsafe are the self-appointed speed police camping in the left lane doing 2 under (looking at you, Greater Seattle). This forces people to pass on the right to keep up the flow of traffic, which is much less predictable.

The correct answer is to keep right except to pass, keep your speed within reason (<10 over max in good weather), keep a decent following distance, and let the cops take care of anyone driving recklessly. If speed was so dangerous by itself, Germany would not have Autobahns.

As a German, I'd like to point out that about half of the German Autobahn network do have speed limits in place, or had in 2019 (this oscillates a lot between about 30-60%, I think).

According to a government study into Autobahn fatalities that occurred in 2018, "non-adapted speed" (defined as the appropriate speed for the traffic conditions and allowing full control of the vehicle) was the cause of death in 45% of cases overall. For stretches without a tempo limit it was 46%, for stretches with one it was 50% -- this may be a good statistical for the tempo limits, as presumably there would have been more deaths on those stretches without the limits having been set.

Overall, 70% of Autobahn deaths occur on the segments without a tempo limit for a variety of reasons.

Speed limits on the Autobahn are a frequently-discussed and controversial topic in Germany, but for what it's worth, they're quite actively managed for safety reasons.

Gov source (in German): https://www.destatis.de/DE/Presse/Pressemitteilungen/Zahl-de...

I'm not denying any of this. I'm just observing that there's a faction in the US which seems to say "speed is a leading factor in car deaths" (which it is) and then insists on everyone never going 1mph over our often-arbitrary-and-politically-set speed limits. When what you're describing as "non-adapted speed" is exactly the actual problem . . . driving beyond one's capabilities or the capabilities of one's vehicle.

Our freeways stateside were largely built in the late 20th century to a uniform 70mph (~110km/h) speed limit engineering-wise. But after the 1970s oil crisis, speed limits became a political football that often have as much to do with driving traffic ticket revenue for local governments as actually promoting driving at a safe speed.

So the problem is that a portion of Americans commenting on our speed limits don't understand there's a difference between "what the government puts up on the sign" and "what is actually a safe speed to drive without endangering other people" as if what's on the sign is some sacred totem that shall not be questioned because We Must All Follow The Rules Like Good Little Boys and Girls.

What the government puts on the sign has significance you know. Maybe the highway was made in the 1950s vs 1970s and interchanges are too tight. Maybe its a surface road with smaller measured sightlines than you might realize. All the yellow lights are going to timed with respect to the posted speed limit and braking distance. All green waves timed with the anticipation you go the posted speed limit.

So let everyone decide to decide for themselves what a safe speed should be on any given stretch of road and see what happens? In other words, let evolution find the 'safe' speed.

I imagine people will be more tolerant of a slower ride if--as a passenger--they can be engaged reading a book or watching a movie or whatever.

Not just because it's easier to pass the time, but also because there's no push to make the experience of driving "more interesting." (Too much, and someone drives unsafely, too little, and they aren't really paying attention.)

> they can be engaged reading a book or watching a movie or whatever

Given how commonly understood this situation is, I wonder what fraction of the population can read a book or watch a movie while riding in a car without getting fairly sick.

A matched case-control analysis of autonomous vs human-driven vehicle accidents

Published: 18 June 2024

https://www.nature.com/articles/s41467-024-48526-4

"The analysis suggests that accidents of vehicles equipped with Advanced Driving Systems generally have a lower chance of occurring than Human-Driven Vehicles in most of the similar accident scenarios. However, accidents involving Advanced Driving Systems occur more frequently than Human-Driven Vehicle accidents under dawn/dusk or turning conditions, which is 5.25 and 1.98 times higher, respectively."

Unfortunately, that seems to include data regarding Tesla's FSD function, which renders it useless for actual comparisons between human driving and responsible implementations of self driving technology.

Nitpick, but Tesla does not yet have full self driving technology. They have a beta which is "supervised" by human drivers. Repeating the myth that they have self driving perpetuates the misconception that Tesla's marketing has pushed in the past.

> Perhaps we humans should be slowing down and driving more carefully?

Not necessarily. Safety is not the only thing we care about, we also want to spend less time in cars.

20mph motorways would be much safer than current limits or even 50mph, but the loss of QALYs from people sitting in cars would not be worth it.

That's quite true, but still surprisingly human like to me when compared to, say, Tesla FSD. It will also take longer to call one due to less supply.

If we want nice, relaxing ride with no time pressure then we call a Waymo. Uber is still our go to if we are in a rush.

Human ride shares drive so wildly they make me sick like half the time. Never happened in a waymo. Yeah it's a bit slower, but everyone always overestimates the gains from driving so fast. We all have to stop at the same red lights, or get in the same lines for the stop sign, even if you get to the line way faster. I'm happy to get there 30s later for a smooth ride.

This is evidence that most speed limits are too low. The proper speed limit on any given road is the 85th percentile of what drivers actually do.

No, the proper speed limit is one that is decided based on traffic engineering safety factors such as sight distances, and who else uses that road.

Cities aren't dropping their speed limits to 25 mph for shits and giggles. They're doing it because the odds of a person on foot or bike surviving being hit by a car goes up dramatically when speed drops from 30 to 25. Fatalities drop to nearly zero at 20mph, which is why many dense residential side-streets are 20mph. It also has the nice side effect of discouraging people Waze-slaloming their way through neighborhoods instead of using major routes.

https://nacto.org/publication/city-limits/the-need/designed-...

The whole "make the speed limit what most people are doing" was just auto industry bullshit that helped make our roads even more dangerous to people not in a car or truck.

> It also has the nice side effect of discouraging people Waze-slaloming their way through neighborhoods instead of using major routes.

I wish cities would solve this problem by making the main roads faster instead of by making the side streets slower.

You aren't solve the problem of "Too many people are driving on the arterials, so traffic is going too slow" by making driving faster and more attractive.

> The whole "make the speed limit what most people are doing" was just auto industry bullshit that helped make our roads even more dangerous to people not in a car or truck.

Hard disagree. It is simply an observation of reality, and a very useful reminder to anyone who wants to waste time and resources putting different numbers on signs. You have to actually engineer the roads for the speeds you want, and I get that it's more expensive than just painting new signs and telling your constituents how much you're trying to help.

We have neighborhoods where cars cannot pass without one giving way and pulling off into the parking area. Is it inconvenient? Sure, sometimes, very much so. But you know what? People drive damn slow. Nobody cuts through the neighborhood to save time when traffic on the arterial road is congested. It matters not one bit what the speed limit signs say (there aren't any, actually), because road design solved the problem neatly.

If the speed limit that was deemed safe for an area is lower than the 85th percentile driver speed, then that road is poorly designed. It should instead be redesigned such that people slow down themselves, regardless of what the sign says.

Speed limits in California are capped at 65 mph by state law (or 55 mph for undivided roads). Any agency wanting to set a lower speed limit has to conduct a speed study and cannot set the limit lower than the 85th percentile of measured traffic speeds.

Of course they can skip the speed study, or not repeat it every 5 years as required, but then tickets are effectively unenforceable on that road.

They can also increase the speed limit:

> Department of Transportation, with the approval of the Department of the California Highway Patrol, may declare a higher maximum speed of 70 miles per hour

https://leginfo.legislature.ca.gov/faces/codes_displaySectio....

Highways like the 101 are 55mph in some parts. Whats most dangerous about californian freeways are the people with ten feet of metal scrap in the truck bed going 40mph in the middle lane.

While most roads are indeed capped at 65, stretches of I-5 are set to 70 , there is a process for this I believe.

Honestly as a cyclist who takes Waymo, all the Waymos I've seen are much nicer to cyclists than any human driver. Most human drivers in SF either buzz past the 3 ft legal limit going 15 mph over (good luck when their mirror taps you at 40 mph) to try and overtake the cyclist or they'll just edge the cyclist out any time there's space to merge into the lane. Waymos usually give cyclists the whole lane comfortably and take time to merge out properly.

(I've read comments that if you're on a skateboard this isn't true but I've never used a skateboard in SF so I don't know much about the experience.)

Sometimes I wish I could load my bike into a Waymo. I hit a flat yesterday and left my replacement tube at home and it would have been so much nicer if I could have loaded my bike into the car and gotten a ride home.

What I found is a hack for cycling on LA streets is just taking the entire lane. It makes it feel like there are actually bike lanes everywhere. It forces the drivers to pass by merging vs squeezing you out. They honk and cuss no matter what you do so you might as well do what makes you most safe and visible to other traffic.

I used to work with a guy who liked cycling, and liked doing it light and in open, fairly rural areas near a city.

He didn't carry tools or spare parts. He'd just summon an appropriate Uber to his location and get a ride home home with his bike if he had an issue.

You probably can (although I haven’t tried). The trunk is accessible and you can flip down the seats and ride in the front.

Marked NOTABUG. One reason to like AVs is that they follow the rules, and a critical mass of AVs, even as a minority of vehicles, will impose a "phase change" on traffic.

Waymo has mounting evidence from millions of driven miles that the presence of their vehicles makes the roads overall safer.

it boggles my mind. why this isnt talked more.

> - Consistently [...] clean [...]

I don't expect this to hold true once Waymo become generally available, alas.

I do. It'll be easier to keep the fleet on a cleaning schedule, and won't be left to the whims of individuals.

I have a good friend who's a Lyft driver. According to him, all drivers are rated on cleanliness by passengers; if you're dinged for a weird smell, there are lasting financial consequences (even if it was for reasons outside of your control, e.g. using a Lyft provided rental while repairing from a traffic accident).

We'll see how Waymo handles it! It will definitely be Waymo's problem to solve, though.

Their policy from an email sent out Oct 2023:

> Many of our riders choose Waymo for the clean and consistent vehicle we offer. To ensure every rider gets this experience, we’ll be applying a vehicle cleaning fee for riders who leave a mess behind in the vehicle, such as vomit, excessive trash, and smoking odors.

> For those that self-report their mess during their ride (not including smoking), the fee will be $50. For issues that go unreported, we’ll charge riders $100 for the first violation and increase the fee for subsequent violations. Repeat trash and smoking related violations may also impact your account standing.

That’s an awesome policy. Compare to car share services (in SF.. apples to apples) such as Gig Car and Getaround which allow unsupervised general access (i.e. no driver there to witness car treatment)… those are generally pigsties. Blunt ash all over the dashboard, used kleenexes in the door handles and cupholders, trash on the floor. It always blew my mind that the perpetrators weren’t fined into the dirt. Good for Waymo.

Considering that I’m in a social bubble of considerate people all of whom wouldn’t leave a single bottle cap in a car, this makes me despair at how people must be outside of my bubble.

I'd say the vast majority of people don't litter or trash things, but the few that do ruin things for everyone. It only takes a few bad apples (on in this case trashy people) to ruin something for everyone else.

Its like living with 4 roommates, but one of them leaves their shit everywhere. It could make the whole place look unkept for everyone else.

And because we generally don't want to pick up after other peoples shit/mess things may be left trashy for a long time before anything gets cleaned.

A very small percentage of people will trash a clean environment.

A much larger percentage of people will add trash to an already trashed environment.

Plus there's a camera in the car. Pretty easy to keep riders accountable, and solve any dispute.

Compare that to other car sharing options (e.g. Zipcar, where cars are usually a mess, but there's no way to keep users accountable).

Hmm zipcars in my area (London) are usually clean.

I wish it was the same here. In NYC (Brooklyn specifically) it's 50/50.

I haven't used Lyft in awhile but as an Uber rider what I can tell you is, if that company has a similar policy, it hasn't worked for me. The quality and cleanliness of the vehicle, and the quality of the driving, has been vastly inferior for me in San Francisco than Waymo has been.

I travel a lot for work so use Lyft a lot and I can say the same. Its not every time, but maybe 1 in 10 times I'll get in a car that smells like absolute shit, smoke odor, unclean, etc.

I think theres a stigma against reporting drivers so most people don't, but then that leaves the people who don't care about these policies to continue giving people bad experiences.

I feel that. With about that same frequency, I encounter a vehicle or a driver or driving that's just dreadful. But, I don't report it or even ding the driver because I feel like that's a hard job that I don't wanna do, and I feel bad. I dunno, maybe I should.

I think the difference is that as an individual driver you have an incentive to keep your car clean, so that Lyft continues to dispatch riders to you. For Waymo the selective pressure is less direct and also spread across their entire fleet. They can accept a level of dirtiness, given some probability that the rider would reject the car x cost of rider requesting a (partial) refund x etc. etc.

More cynically, there are simply too many people that won't take care of "public" property. If every 3rd rider (exaggerated for rhetorical purposes) trashes the car, it's gonna be dirty no matter what.

Assuming the cars have internal cameras they should be able to surcharge riders who trash the cars.

That doesn't cover all the normal cleaning needed just from people being in the cars, on top of environmental stuff (sand, dirt, mud, leaf debris, pollen...)

Like others said: the second they have run drivers out, the cars will stop getting cleaned obsessively because you won't have a choice.

Same thing that happened when Lyft and Uber when they ran the taxi industry out of business.

Same thing that happened when Zipcar was established. Cars went from being spotless and well maintained to damaged, dirty, and half the dash being lit up.

> you won't have a choice

I will have a choice. I will choose one of the competitors.

> Same thing that happened when Lyft and Uber when they ran the taxi industry out of business

There's still a taxicab industry in San Francisco. It became better after competing with Lyft and Uber.

I don't know how many people there are who won't take care of public property, but whatever it is I doubt it's 1 in 3.

The metric will be "cost of time spent cleaning" / "cost of sending out a dirty car." If this ratio is > 1 the "recommended cleaning schedule" will be the lowest priority item in the entire fleet.

The cost of cleaning goes way down when you do it in bulk at a service center rather than individual lyft/uber drivers trying to do it. A standardized car also helps.

If you can incorporate the cost of large horizontal demand spikes into the off hours and you can find cheap enough labor to fill it, perhaps.

The time spent travelling out of service, in cleaning, and back into service are all lost opportunities. Hopefully you can clean a very large number of cars in a very short period of time.

The cars may be standard. The messes, obviously, will not be.

The cars already need to go out of services to recharge, they can just clean the cars while charging

The have multiple charging centers strategically placed throughout the city. These seem to currently only have security guards there. The logistics of having cleaning staff there and trying to match their schedule to expected charging times is probably not very difficult but also not very reliable either.

The win they do have, that I did not consider is, they have cameras _in_ the car. So visible cleanliness is something they can manually check before and after the rid and schedule for service if required; however, it currently seems that this requires the vehicle to go to the larger centralized maintenance facility, which I guessing takes quit a bit more time than the auxiliary charge only lots.

Not trying to be super pessimistic, but mixing distributed autonomous operations with centralized manual service, especially in an urban environment, seems fraught with novel challenges.

The cars will all charge overnight. Clean them there. Getting to and from the charging stations / cleaning depots is roughly free. It’s just the electricity which is like 5 cents a mile roughly. Utilization at night is really low anyways so not much opportunity cost.

The overall point being way more efficient to clean than an Uber.

Once you have a large enough fleet of roughly the same shape, it starts to make sense to build some automated cleaning thing. I'm picturing like, a long tube with vacuum ports in it that can just be shoved in by a robot arm to vacuum each seat. Unless there's a wet mess, that'll basically get it 90% of the way there.

I think we'll see such self-driving taxi interiors optimized for staying clean and then ease of cleaning. At the limit, think like a stainless steel kitchen that can just be sprayed down. Or these sort of self-cleaning public bathrooms: https://www.youtube.com/shorts/z81KtV9w5fo?feature=share

That assumes a specific low budget market and does not apply to all segments.

So you're saying there will be a higher price point for the users that want a clean ride?

By this logic public transportation should be the cleanest of them all. But, it is unfortunately far from that.

Public transit is actually pretty clean. The main problem is that they can't refuse to board disruptive passengers, who just trash everything around them. Waymo has cameras inside, so they can fine/ban you if you litter.

The driverless cars can also, well, drive themselves to a cleaning center during the night-time when it won't affect the income generation.

Another interesting possible feature is to redesign cars to be more easily cleanable. Driverless cars can dispense with the central console, steering wheel and pedals, with simple hard plastic covered in detachable covering. Even the seats can be made detachable.

> Public transit is actually pretty clean. The main problem is that they can't refuse to board disruptive passengers, who just trash everything around them.

They can. My local state operated ferry plays an announcement on every ride that if you harass the staff, they will issue a no trespass order banning you from the service for 60 days. Violating the order is criminal tresspassing.

Of course, enforcement isn't easy, and confronting passengers to issue the orders isn't easy, and both can contribute to confusion and delay. It's a little easier to do on a ferry vs a bus because there's more staff.

If you're looking for easy clean surfaces, you could go for police car backseats. Not very comfortable from the looks of them, but very amenable to cleaning.

> They can. My local state operated ferry plays an announcement on every ride that if you harass the staff, they will issue a no trespass order banning you from the service for 60 days. Violating the order is criminal tresspassing.

I might believe that for ferries, they have personnel for that.

Our local bus drivers were told that fentanyl fumes are not harmful, and that asking the state to do anything about junkies smoking _on_ _the_ _bus_ is systemic racism or something.

Its not the refusal of certain groups of people that causes the cleanliness issue. Its the fact the bus doesn’t see a custodian until its out of service. The difference in cleanliness between poor and dirty areas and rich and clean areas of a city always comes down to investment in routine custodial service vs any behavioral difference. Rich people litter just as much, more sometimes knowing theres someone to clean it up.

Unsure where you’ve lived but in Philadelphia/New York/LA/SF/Seattle/Portland public transit is pretty disgusting, troublesome passengers or not.

> in Philadelphia/New York/LA/SF/Seattle/Portland public transit is pretty disgusting

In New York right now. Subway stations are dinghy. But the cars themselves are fine. Hop on the Metro-North or LIRR and they're on par with European cleanliness standards. (If not behavioural.)

Dingy/dinghy is a great one for the Eggcorn Database!

https://news.ycombinator.com/item?id=40720548

If you think the cars in NYC are clean, look under the seats. LIRR and Metro-North are OK, but also not exactly clean.

European trains are far cleaner.

In LA county the newer lrt is fine. The old purple and red car trains though are disgusting even when they are clean. Theres usually a constant unpleasant background smell, halfway between stale cigarette and old puke, the cars are usually hot as hell causing everyone to sweat, and sometimes the already quite poor and noisy ventilation is off entirely in the traincar. It also doesn’t help that the aesthetic is beige and brown 90s style, which makes it even more dirty looking than the newer grey and stainless interiors. Most of the stations are overdue for a power washing. Especially down in the tracking thats full of trash and pee. Its not bad enough to be a deal breaker to me even with the occasional tweaker actively smoking some rock, but still its not pleasant.

> public transportation should be the cleanest of them all

Privately operated public transport (e.g. jitneys and ferries) frequently are when positioned at a premium, as Waymo is.

This is counter to my personal experiences - but I'm certain it does depend on location/region.

Yes because a $3 ride on a public bus is exactly like a $23 ride in a Waymo car.

Just wait until they 1) have a near monopoly and/or 2) decide it’s time to actually make a profit. Because it won’t last either way.

That might be a long wait. There's no particular reason a monopoly need appear in this market. If it does, that will be a matter of public policy at it always is. That would be unfortunate, as it always is, but it would also be larger and different problem that goes well beyond this particular market.

Waymo has a partnership with Avis Budget Group to use their fleet cleaning and refueling facilities. I imagine if a user reports a dirty car someone can pop in on a camera, check it out, and divert the car to a depot.

They have some rider rules around keeping the car clean (screenshot link below). I'm not sure if they are enforced to this level but they do have at least one camera in the car (above the middle rear seat).

https://i.imgur.com/vCRvWOc.jpeg

I suspect they visually inspect the cars after each ride and send them to the depot if there's something that needs to be cleaned up. There are several cameras inside the car giving a good view of the interior.

If they keep the 10% mark up over competitors maybe there is still hope.

I'm curious why they can't pick up and drop off at exact locations, as long as the location is in their operating boundaries.

Because they have courtesy to not block traffic. Anecdotally, drivers are also MUCH more aggressive near them, so I’d be pretty nervous as a pedestrian getting in/out of one near heavy traffic. I’m not handicap in any way, and I’m totally fine walking to the corner of a city block to get in it, in exchange for a much safer ride than an Uber.

I’ve never had it be more than a few hundred feet - usually it’s 2-3 cars away where it can parallel park on the other side of an intersection in the WORST case. Oh I guess they do avoid some of the intense AF hills but I’ve had Uber drivers do the same.

In my experience they are very strict about finding a “safe” place to pick up and drop off, which often means turning onto a small side street or looking for a gap in parked cars. They won’t block traffic like a Lyft/Uber driver might.

> They won’t block traffic like a Lyft/Uber driver might.

Wow, then I don't know how they'd ever expand to NYC.

In most areas there's no such thing as a "safe place" that doesn't block traffic.

If there are sides of the road that aren't traffic (or bicycle) lanes, then they're taken up by parking.

Taxis, Ubers, delivery trucks -- literally everything just stops in the street (or bicycle lane) and traffic temporarily goes around it.

They'll just have to change the behavior, I assume the current behavior is just out of caution not some technical limit.

It's because it's illegal to double-park.

Everyone in NYC does it because they have to, but Waymo is obeying the law.

The app forces you to pick locations that the car can park safely (no double parking) that's closest to the requested locations. They can be 1-2 minutes walk.

Not a big deal for us to be honest, except when going to the theaters in SF, where the car can stop a block away in sketchy Tenderloin.

Human drivers are willing to break the law for pick ups and drop offs, and as a society we largely tolerate that as long as it's not egregious in terms of safety or blocking others.

But programming a robot to deliberately break the law is uncomfortable for people to think about.

This might lead to a beneficial feedback loop for some traffic laws.

When an Uber is near people wave to it, and the driver can stop where they are.

The Waymos probably don't have that kind of social skills.

They should have flashing lights and stop in the middle of the road like school busses do.

Easier to scale and make tech cheaper than humans cheaper.

Waymo just doesn’t want the backlash of them replacing human jobs right now.

Slightly expensive than humans for a premium consistent service is a good play.

Kudos to consistent execution for Waymo.

Wish we could say the same for Tesla.

They’re amazing, I only take Waymo in SF now and will do the same when possible on the peninsula.

do you have to tip the 'driver'? jokes aside, i would be happy to take them once the expand to SFO region so i can have a consistent experience.

I’d love to take it to the airport, but I imagine it’ll be a mess trying to pull up and park at departures. It’s already incredibly chaotic with the way humans are, and I don’t expect they’ll even pretend to extend any curtesy to a robot.

Maybe I’m wrong though? I think they go to the airport in phoenix, I just haven’t heard any reports on the experience.

Allowing Waymo at SFO is a political mess because the airport owned by the city and the city hates robotaxis. Waymo did apply for a permit to operate at SFO, but were denied.

That’s annoying - I hope it gets worked out, that’s the last bit I need to take Waymos exclusively.

> it’ll be a mess trying to pull up and park at departures

I’ve taken it to and from the airport at Phoenix. Pick-up was an in-airport light-rail station to the first stop. Drop-off was right to the curb late night.

Worth trying to sign up now. The waitlist takes a while.

It's not on a waitlist in SF anymore, Waymo is just open.

Interesting, my wife was still on waitlist in SF last week.

10%? you’re lucky. Every time I check the app it’s usually twice the price of an Uber or Lyft.

Does that 10% markup account for the Uber/lyft driver tip?

You guys are tipping your Uber drivers? IIRC you couldn’t even tip until a few years ago.

Well, you can now, and I do. It's hard enough living in California and especially San Francisco, and I believe being in the service industry here is no picnic. Tipping is the least I can do.

I know this feels right, but introducing tip culture into a market currently free of it is bad. Where there's a norm of tipping, it doesn't make the servers more money than where there is not; what you should expect it to do is to decrease the quoted price and increase the variance of income to the servers. And incidentally increase the mental overhead for everyone, increase the artificiality of interactions, etc.

> I know this feels right, but introducing tip culture into a market currently free of it is bad

I couldn't possibly agree with this less. My view is that "it depends." Level salaries in San Francisco and I'll consider not tipping. Until then, I'm comfortable with my choices.

Do what you wish, of course. But you seem to think this helps to level income, given the "level salaries" bit. What I'm saying is it doesn't. Markets equilibrate; they adjust when the form of payment changes.

I'd agree that after tipping culture is definitely established in a field, then stiffing someone is unfortunately bad, unless you can coordinate on reversing this harmful cultural norm. This is just very different from not helping to establish the harmful cultural norm, which is a positive, prosocial good; I believe you are doing the opposite of a prosocial good here, without realizing it. Of course I'm a fallible human.

Well. I don't want to speak for other people. If I were an Uber driver I personally would like to make more money rather than less, and if that happened with tips I wouldn't grumble too much about it. Maybe that's just me.

I think you both are speaking over eachother. You say on individual terms tipping makes sense and it does make sense in a vacuum like that. The other commenter is speaking in macro terms though. Having tips means theres less pressure to actually raise wages. It also fractures labor. You will have people getting no tips who really want higher wages and then people who are tipped well and comfortable with the status quo. This is a big issue with the restaurant industry; servers and bartenders in big cities might make most of their take from tips yet back of house struggles on the same base hourly wage since they aren’t tipped. Incentives are such that a high tipped server doesn’t want tipping to be replaced by wage increases sustained by menu price increases, since they will be splitting that now with back of house or less well tipped workers and will be making net less.

Uber/Lyft drivers only get tips on 15-30% of rides, so tips are like ~3% of revenue. If you choose to tip, then prices are going to be very similar between Waymo and Uber/Lyft.

In my experience, after tip Lyft/Uber are more expensive. The exception is during surges when Waymo's price shoots up like crazy.

From my experience prices for Waymo are at least double that of an equivalent Lyft/Uber ride and wait times are usually 20+ minutes. It is a great novelty but nowhere near where it needs to be to handle real scale.

Doesn’t match my experience in SF. I see 10% premium over Uber/Lyft, with Lyft and Waymo displaying wait times that are ~3 slower, but actual wait times are the same. (Uber consistently underestimates in my experience.)

I also haven't had Waymo assign a car and then decide to assign it elsewhere sometime before arrival.

I have had Uber/Lyft cars that were going to come to me and then some time before picking me up, they change their mind and I get to start a new wait for a new car/driver... worse, I've had that happen a few times in a row on some occasions.

I've seen more in the 10% premium ballpark for most Waymo rides (for just the base fare, excluding tips). There was a very stormy day once where the price shot up pretty high very quickly, but that's not the norm I see.

As an aside, Waymo does have a feature which allows you to add stops that, if within certain parameters like the trip being within 30 minutes (I believe) that it kinda treats it as one trip which is rather cheaper than multiple independent trips.

So, for example, I had to pick up my kid from school the other day and found out last minute I wouldn't have a car. So I booked a Waymo, selected my destination, and then also selected that it would be "round trip" in the app. The car got me to the school and an option in the Waymo app for "ready for pickup" was available; the car left after it dropped me off. After my kid came out, I clicked that ready button and another car came to the school and took us home.

The price for the round trip was just over $20. The price for two independent trips would have been about $10 more (if I'm calculating & remembering correctly). So another interesting feature on the pricing discussion which, to my knowledge, isn't an option with the other ridesharing/taxi businesses.

This feature exists on Uber and Lyft too, although iirc it’s buried underneath some ads

This won’t last long. Waymo has a treasure chest waiting to be deployed as soon as they start expanding more.

I genuinely expect that there will be a point at which manufacturing becomes the bottleneck.

That's going to be a lot of cars to build.

I really wonder if and when they expand to a lot more car manufacturers and models, just to avoid supply chain issues.

I live 15 minutes from downtown Phoenix, and it took them 2 years to expand to my location. Over 4 years after Waymo launched beta testing here..

Waymo cannot take me on the freeway (backstreets only)

Waymo cannot go to the entire city of Peoria, Surprise, Goodyear..

Waymo cannot go to the entire western half of the state.

If you live 30 minutes from the downtown city centre, there is a 30% chance Waymo can go there. Waymo serves higher-end areas only today (Scottsdale, Tempe, Chandler) - if you live in a poorer area, no Waymo for you.

I have family who lives in Scottsdale and Surprise, Waymo can't take me there.

My dentist is less than a mile from my house, Waymo can't take me there - but can take me to a fashion mall in chandler.

Is it better than Uber/Lyft? Experience wise, sure, but their snail's pace expansion is disheartening. Not to mention the freeway issue.

It will be more profitable for them to deploy cars to new cities at a premium than to deploy them to existing cities to capture more of those cities' markets. We won't see affordable driverless taxi service for many years unless competition appears. After Uber and Cruise imploded, I don't have much hope for that.

Is that true? There's a ton of overhead required to have any presence in a city (all of the fueling/cleaning/repairing/storing/etc depots) that you need to amortize over lots of cars. If they can keep the cars busy, I think they'd much rather have 5,000 cars in one city than 500 cars in 10 cities (with 10x the overhead).

They have a big upfront cost to add a new city, but once they add it, their marginal revenue per car will be higher there (partly because they can be sure to keep those cars busy and partly because they can charge a higher premium).

They have to prepare the infrastructure anyway in all the cities to get to the scale they want to get to, but each new car built is going to be sent to the least scaled city so far.

Does that make sense from a consumer brand perspective though? If everytime I check Waymo there's no capacity or it's super expensive, pretty soon I won't even bother checking it.

It's worked so far. People are willing to pay the premium.

Cruise is still here, they just stopped their unattened car testing.

If they expand their supply, I would expect that to lower prices, not raise them.

> nowhere near where it needs to be to handle real scale.

This is a phenomenally solvable problem. And it’ll scale without requiring more humans to train and monitor.

in my experience the prices are competitive to 10% more and wait times are 5-12 minutes longer than Uber/lyft

the comparisons are worse if i am coming from the outer richmond

> - Consistently spacious, clean and quiet cars. You know what you'll get. - AC always works and not up to the whim of the driver.

How much of the cars being clean, quiet, and having HVAC working is just because it's new?

A few years down the line the cars will be old (until they're replaced) and won't be as clean or quiet or have reliable AC. It will probably become like aircraft where they prioritize maintenance of safety-critical features but not so much the seat comfort.

There are lots of upsides to not having a driver in the car, but I wouldn't count on the cars always being nice.

The cars are operated by Waymo, so I have hope it’ll be more well maintained across the fleet.

My guess is that AC is off because keeping it on increases fuel consumption and that cuts into the driver's tiny profit margins.

> A few years down the line the cars will be old

Maybe. Then again, maybe not.

A car that's been around a few years is old by definition.

Hard disagree. A car that has been around for three years is not old.

How are the starts / stops? Jerky? I have c-spine issues. That's my main concern.

Why are they more expensive? I hoped they would be cheaper without a driver to compensate.

Extremely smooth! I can't comment on the c-spine issues but I'd argue that it's the smoothest driver I've ever experienced. Surely they have some math in their AI driving code that clamps acceleration and jerkiness. It's NOTHING like Tesla Full Self Driving, which I find to be incredibly jerky for both steering and acceleration.

Why are they more expensive?

Likely because they're not pricing based on their costs, but rather what the market will bear.

Although they have a lot of R&D to recoup, so that's a cost.

Also, perhaps it's a consequence of limited supply. If the SF fleet is still just 250 cars that seems quite small to me for a city with a lot of demand for ride services.

much smoother than the sf uber teslas that all have jerky regenerative braking on

You take ride share with a baby? How does that work, with regards to car seats?

We take an infant car seat and use the European belt routing (or if that fails due to short belts in American cars, American belt routing) to secure to the back seat.

WTF belt routing? Don't you guys have Isofix over there?

We do, and I have the base with isofix in my own car.

Wait how do they justify having an even more expensive ride with fewer people to pay? The whole point of cutting the driver out is to save on cost, without that the entire project is no better than uber or lyft (which already overcharges by an arm and a leg)

If they priced below human labour they might have their testing licenses revoked due to political pressure. Right now they're still in the testing phase, there's no need to rock the boat by pricing below uber and lyft yet.

You raise the price to reduce the demand to what you can comfortably supply. Price is not a tool of justice. It's a tool for incentivising or disincentivising the customer.

Initially, automation is always more expensive than cheap labor. That’s why we don’t automate things until human labor becomes more expensive than the ease of automation.

To some extent this is true, but most of the automation I've witnessed in my lifetime led with direct cost savings. My main conclusion is that waymo's customers are more likely to be scared by other humans than they are to be price sensitive.

> most of the automation I've witnessed in my lifetime led with direct cost savings

Direct cost savings to the business or to the customer?

Waymo has been operating at an incredible loss for a very long time. They now have a product that people are willing to pay for and, in fact, find to be more comfortable—from OP's description it's basically a luxury Uber. Further, they expect to not be the only autonomous taxi system forever, competitors will eventually catch up.

Given that, I think it's only rational for them to operate at very high margins for as long as they can before competition drives the prices down. They probably won't come anywhere close to recouping the loss in that interval, but they'll get a head start on it.

Yep. They also probably don’t have that many waymo cars yet. I’m sure eventually they hope to blanket the city but for now, I assume they price rides such that supply = demand. If it was so cheap that everyone ordered them, you’d never get a car. That would be a really bad experience for customers and they would bounce. It’s incredibly hard to get someone to try your service a second time if their first experience was negative.

without that the entire project is no better than uber or lyft

Obviously not, because consumers appear willing to pay a premium to not have a driver.

You are using a product before enshittification begins, of course it's much nicer.

When it's been 28 quarters of general availability of Wayno and profits need to keep growing faster than before, they'll get dirty, the AC will stop being fixed, they'll start getting flats / engine trouble, etc. in the middle of the drive.

Enjoy it while it lasts. Be sure you move quickly to the next new thing before this one goes downhill too :/

Doesn’t the price end up being the same if you factor in tips with Lyft?

> Rides have usually 10% mark up over Lyft and Uber.

I found that to be true as well, but when you factor in the tip to the driver they come out to more or less the same price.

> - Rides have usually 10% mark up over Lyft and Uber.

Isn't the whole point of Waymo is that it'll be cheaper than a human driver?

The point is that it's more profitable for who runs the taxi company.

Eventually price will always be what people are willing to pay.

I bet what that in a decade the price will be identical to (disappearing) human drivers and then gouged higher.

It's depressing how focused we are on the wrong things for cities and planet, self driving taxis rather than quality public transport. We are destined to extinction and we deserve it.

I had trouble with it not wanting to drop me where I asked, or wanting to pull over or wanting to let me out. Never again.

Sorry, but any argument that basically boils down to "it's better because I don't have to talk to a driver" is not a real argument. Drivers are people too. Nevermind the fact that you are cheering for people to lose their jobs.

Would you accept any historical development that reduced demand for some kind of labor (or human interaction) as potentially positive? There are a lot of those.

(I'm happy to hear if the answer is "yes".)

For example, most kinds of shops were (like jewelry stores today) apparently historically not self-service, so you had to ask the shopkeeper for every item individually, and the shopkeeper would have to retrieve it for you.

Maybe a more direct analogy is that you once had to ask human operators to complete telephone calls for you, and gradually this was replaced with automated switches and direct-dial systems. That has benefits and drawbacks, but I think I appreciate it quite a bit. Do you think it would be better if that change hadn't happened?

This is not a reasonable comparison.

In Tesla's model, you buy a Tesla, and allow it to operate as a robotaxi when you're not using it. It generates income passively for you. The technology has introduced automation that might have displaced your job responsibility but not your means of generating income.

In Waymo's model, you have completely displaced both the job itself and the means of generating income. This is a strict lose-lose for everyone except Google, who reaps the profit.

Why is it not a reasonable comparison?

Automated telephone switching fits the latter pattern. The job and the income went away. So the question is valid, would you prefer automated telephone switching was never deployed?

Putting aside the fact that Tesla's robotaxi model will likely never actually ship, I don't see how you're not displacing a driver's job by operating your car as a robotaxi instead of letting it sit idle.

If the vast majority of the robotaxies are owned by normal households that have 1 or 2 of them, then that's the people owning the means of automated production. Ideally it would be the people that already own taxis getting them upgraded, but it's a lot better than corporate fleets for putting the revenue into the pockets of people.

It's real for us. We had a ride when the driver wouldn't stop talking and the baby was trying to sleep. Not all people are nice.

super bad take. by this logic nobody should be excited about any technology whatsoever

Many technologies enable some new cool things and in the process automate some stuff.

Here we are talking about paying extra pretty much solely to not have to deal with a human.

Waymos enable fewer families torn apart by car crashes.

> No chatty driver to disturb our sleeping baby

San Fransisco slave-owning class misanthropy on display.

I'm sorry that the poor tried to talk to you instead of shutting up, and performing his digi-task without disturbing your innocent baby.

Is it any wonder that the same upper-middle class NPC's that treat human beings like machines also (in error) dream of a future where machines will replace human beings?

Yikes, you can't attack another user like that here. We have to ban accounts that do, so please don't do it again.

While I have you: could you please stop posting unsubstantive comments and flamebait generally? You've unfortunately been doing it repeatedly. It's not what this site is for, and destroys what it is for. If you wouldn't mind reviewing https://news.ycombinator.com/newsguidelines.html and taking the intended spirit of the site more to heart, we'd be grateful.

Lmao I can't tell if this is satire or not but calling people NPCs always reminds me of this XKCD I once had up in my cubicle. https://xkcd.com/610/

Reply on news.ycombinator.com