Looks amazing! (Also, I've known Dexter since before Human Layer and he's a force of nature. If you think this is interesting now, you're going to be amazed at where it goes)
Great idea. I used to work at Instabase, which you probably compete with. The better you are at dealing with dodgy PDFs and document scans, the more valuable this will be to big banks, shipping companies, etc.
They've been in Phoenix for the past two years (generally available for the past year). They're so commonplace now, I don't think anyone really notices them.
Sort of I guess, but there are things like how they look very different and that F# is very very fast, has static types and OCaml is not exactly slow but not fast exactly fast, either.
Here's one concrete example -- these cars almost never take left hand turns. They'll route multiple right hand turns to get somewhere, which makes it feel much less risky when pulling onto a busy street from a neighborhood, for example.
Same here (downtown Phoenix area) -- the driverless Waymos have been awesome. Occasionally they get confused but it seems pretty rare. They're better than a lot of the other drivers on the road out here.
I'm mostly excited that they upped the allowed passenger count from 3 to 4. Usually we take a car if we're going out to dinner with friends but the 3 passenger limit made taking one of these a pain in the ass or impossible.
I haven't taken one since they announced the change (I'm in a part of Phoenix that's pretty walkable), but I think it will be 3 in the rear bench and one in the passenger seat. They've been pretty clear that they don't want people sitting in the driver's seat, and I'm guessing that will continue to be the case with this change.
I wonder if there's some regulation behind why they haven't got rid of the traditional front-facing layout entirely and replaced it with train carriage style seats around a table?
These cars still have to meet all federal crash safety standards. Modifying the seating position would also require modifying the airbags, seat belts, seat belt pre-tensioners, etc. Then the vehicle would have to be crash tested again to ensure it is safe for the passengers.
Part of this announcement is the completion of retiring our Pacifica fleet. So while we used to have the vans on the road in Chandler, you'll now see the same vehicles that were in Downtown Phoenix and San Francisco.
It blows my mind that someone living downtown in a city the size of Phoenix (relatively large) needs to use a car at all. It’s really a shameful failure of urban planning.
Is a 20th-century mass transit system actually preferable to robocars once the technology is mature? My guess is it's the cities like Austin, planning new transit systems without designing for synergies with robocars, who are failing at urban planning.
Good public mass transit is absolutely superior. Waymo cars don't solve traffic, it only partly solves parking and paying a human.
Having (on average) a single person take up 40 square feet of space on the road for travel is horribly inefficient. You do a few orders of magnitude better with a full train car or bus.
Not only don’t autonomous cars solve traffic, they will actively make it worse. The marginal cost of operating a car drops dramatically when you remove the human drivers from them (who also suffer from pesky conditions like boredom and drowsiness), and so we can expect these vehicles to drive signficantly more frequently than their human-operated cousins.
Think about uses like DoorDash. Instead of (in a city) walking to nearby restaurants, we now replace a larger share of those dinners with ones where cars are driving back and forth across the city to pick up the food and deliver it. This will only become more common when the cost drops by not having to pay a human being, and so we can expect a much greater volume of traffic from this one example use.
Other currently-marginal use cases will see equivalent growth due to the similar economics.
Re traffic: with robot-driven cars, congestion pricing would be technically easy. This is an example of not considering possibilities created by the new tech.
You are comparing with the least efficient form of transit, and looking at cities that are not really designed for transit anyway. So yes, bad transit in cities not designed for it isn't that efficient. Note, on top of this, that you are purely looking at energy efficiency, when the key problem of individual robotaxis is space efficiency.
Realistically, what breaks efficiencies in the US is city design: Places of work mostly detached from activities and housing, pretty much guaranteeing inefficiency. The robotaxis don't make a difference here: The max number of people looking to travel at once doesn't change, and neither does their destination. This makes the number of vehicles actually on the road not change very much at all: The best you can do is hope for people being comfortable (and not inconvenienced) by carpooling with strangers, not unlike what lyft and uber offer already. In the middle of the day, there's no new demand for robotaxis: They'd be parked somewhere, and every mile they move is a car/mile of congestion they create. So you either park the cars where the jobs are, getting, in practice, the same results as private vehicles, or you park them somewhere else, increasing total congestion. It's not just robotaxis that would park: See what happens in all the train commuter lines in American cities, where most trains just get parked in a yard.
The miracle of cities with top transit (See, for instance, Madrid), is that a relatively high percentage of the network is about equally busy in both directions. Trains might stop in rush hour, but route 6 will be useful in either direction most of the time, with little waste compared to US cities. This is the real weakness of the argument of just adding more transit: What we need to make it work is basically urban rebuilds. IMO still a good idea, bit it's a far slower, and more expensive problem than it might appear.
Autonomous vehicles will not make cycling or walking next to much faster multi-ton vehicles more pleasant, even if those vehicles are less likely to kill you. I mean, have you walked along even a 35mph road with significant traffic? It's extremely unpleasant due to the noise and air pollution.
In fact, today the area around elevated freeways is often much cheaper than the surrounding area for several blocks.
They were wrong and their argument was bad, but it remains a fact that America has dismal public transit compared to other developed nations (which is why their argument was bad).
Robo cars are more expensive during peak usage, so they can form habits where public transit is both encouraged because I took a robocalled to work and the bus home because then it was rush hour, or Vice versa. Or I took the bus to work and left late at night after the buses stopped running via a robo taxi…etc. All this occurs with Uber already, I don’t bother renting a car because even though it sucks to take the bus from the airport, I can Uber that and use more efficient transit when available, since I have something that fills in the gaps. Robo cars just make that link more efficient (lower cost eventually as humans become more and more expensive).
They make not taking your car with you a bit more feasible, and that’s a start at least.
Traffic isn't a function of road space. The true throughput of roads is immense when utilized by better drivers (one extreme example: https://www.youtube.com/watch?v=4pbAI40dK0A). A lot of traffic is caused by bad human drivers: accidents (and associated rubbernecking), shockwaves, running red lights, double parking, circling around looking for parking. Additionally entire additional lanes of traffic (for cars or personal mobility vehicles) will be available when streetside parking isn't necessary. Personally I look forward to the possibilities of turning all that unecessary parking space into something useful ("we’re the Saudi Arabia of developable land in cities").
Good public mass transit presumably involves cops kicking out the smelly homeless people, right? I don't think that would fly in most places these days though, it isn't humane. Yet it's the only way to make public transit tolerable.
Some of them aren’t merely lacking access to resources. Mental illness is part of the picture too. Trains and busses are warm comfortable places for such, as well as the criminal element. People think of bus and subway stations as dirty and often dangerous places, not just in the USA.
I don't know if they're un-homed or not, but I've encountered a lot of people smoking, shouting, or making a mess with food in rail cars, making them unusable for everyone else.
Cars take up too much space. Relying on cars instead of mass transit means everything has to be more spread out, which makes walking and biking a lot worse.
If everyone can share a small fleet of self driving vehicles, they don’t need to take up a lot of space at all. I’m assuming you were talking about parking space; I don’t see how self driving cars would take up significantly more space than mass transit while on the road.
> If everyone can share a small fleet of self driving vehicles
They cannot.
> I don’t see how self driving cars would take up significantly more space than mass transit while on the road.
They would take up more space because they're cars. Cars can't get to similar efficiencies of space compared to buses unless you compared packed private cars with sparsely ridden buses. And then of course, there's trains.
Perhaps someday we'll have all self driving cars that can safely coordinate and pack themselves into smaller spaces on the road, but that's decades away at a minimum, since you'd first need to get rid of all manually driven cars.
Trains are a very inefficient use of space. Trains often run nearly empty, and the tracks can only be used by trains. A road (utilized by autonomous vehicles) can be used at any time and any schedule by any vehicle, including the possibility of giving automatic intersection right of way to pedestrians, or safely sharing road space with personal transport (bikes, scooters, etc.)
If a thousand people try to go to work at the same time, the car being driverless or not will not result in less than a thousand cars being driven, and the number of cars in the road is the biggest cause of congestion.
However, a bus can carry a hundred people comfortably. Ten buses instead of a thousand cars results in much leas congestion.
Well, not exactly. The good thing is that automated, centrally booked autonomous systems can run pretty good car pools. Sure human drivers can do this but it's just not as practical. Not at all. So it's possible that those 1000 people might be served by 500, 300 or maybe even 250 autonomous cars.
From an energetics perspective it's incredibly hard to beat walking, biking or rails. If your occupancy on the car is reliably greater than 1, that is an improvement over single occupancy cars, but those boxes are just huge compared to a bicycle.
I spouted this line for a long time, but was recently corrected by commenters pointing out that 2-3 occupancy electric cars actually beat the vast majority of subways and trains in practice, because you don't need to accelerate and decelerate every passenger at every possible stop. Cars are designed to be remarkably aerodynamically efficient, especially compared to motorcycles or bicycles, and an electric battery+motor is much more efficient at producing power than your muscles.
Delivering on this requires some sort of "Uber Pool" like algorithm to put a few people together in a vehicle, but doesn't require large busses.
In cities where space is at a premium, a bunch of people standing, or on a bicycle, will always beat out people sitting down with room for an engine and a trunk.
Average occupancy of a car is 1.5 in the US, and that number shrinks as household size declines.
(It’s also worth noting that there’s no reason that automated technologies wouldn’t also spread to buses and trains. In fact, you can already automate trains. The big win there, is that buses represent massive fleet commonality.)
> In cities where space is at a premium, a bunch of people standing, or on a bicycle, will always beat out people sitting down with room for an engine and a trunk.
It depends. For example, you can theoretically fit more people into crowded bike lanes than you can into crowded streets. But around here at least, bike lanes don't get a lot of people. I lived by one for years that I don't think I ever saw a bicycle on once. But even when not in use, all of that space is still being set aside for cyclists.
Most of the time the issue with bike lanes is really an issue of a bike lane network.
Where a wide high-quality network is deployed quickly, you do often see high ridership increases, but usually in the US what you see is a bunch of scattered lanes that barely reach anywhere in an unbroken link, and no one wants to ride a journey that is even 5% dangerous.
As a regular, enthusiastic, urban cyclist, there is nothing I want more than widespread use of autonomous vehicles. With well-behaved (i.e. not human) drivers, every street becomes safer and less stressful than a dedicated bike lane.
Also, there was that case with the Uber self-driving car hitting the person crossing the street with a bicycle in Phoenix, because they had turned off the bicycle detection resulting in automatic braking.
Keep going with that thinking. What if trains were autonomous, electric, exactly/efficiently sized to what they were transporting, didn't require expensive/dedicated tracks, and could go directly between any 2 points? You end up with autonomous vehicles.
Trains are nice because they don’t require massive reconfigurations to hold varying crowds of people and the things they carry. People move furniture on subways all the time. There isn’t really a ‘perfectly sized’ personal vehicle; even now, just the simple bifurcation between a normal and XL taxi leads to longer waits for the latter, and god forbid you need special accommodations like for a wheelchair.
Car-like vehicles have massive inefficiencies, because the act of merging into and out of other lanes is inefficient.
A train line can carry up to 80k people per direction per hour. The FHWA’s estimate of a car lane car capacity is 2k people per direction per hour. And because cars have to eventually dump onto a surface network that requires timed cycles so that pedestrians and cyclists can cross the street, these low capacity segments are actually the bottleneck.
Once it makes economic sense to design car models specifically for robotaxi use, small light one-passenger electric models may well become the most common type. In the graph I linked from my other reply, a one-person E.V. like this beat all of the transit systems and came in second only to an electric scooter/trike.
Still more cost efficient to run a single bus than separate vehicles, just in maintenance alone. Also we already have buses, these single occupation EVs that still have enough space for a decent battery size don't exist.
These seems reasonable, but I don't think the math holds up. A 2016 study (http://www.columbia.edu/~ja3041/Electric%20Bus%20Analysis%20...) showed the lifetime cost (including maintenance) of a bus for its 12 year useful lifespan to be $1.4 million. Without even doing the inflation adjustment, that could buy at least 12 electric vehicles. Those 12 vehicles could be much more efficient even running on a single fixed bus route (i.e. more frequent schedule), or (more likely) operating like taxis by taking people to and from exactly where they want to go.
That's a bit extreme. I lived in a city with a similar climate for over a decade and had no problem walking, running, biking, and playing basketball in 110+ degree dry heat. I now live in a relatively humid area and I wouldn't go hiking when it gets over 85 degrees.
Like LA: Large, sprawling, full of concrete, car-dependent
Without the good parts: No media industry, very little tech industry, much less diverse, far fewer good restaurants, no ocean
As an employer, the flipside of this is that there needs to be some value in having someone there that you know is leaving. If someone has been halfway checked out because they've been interviewing elsewhere, it might not be worth keeping them around for six weeks. They can't start any long-running work, or anything with any dependencies, and many people's work quality drops significantly after putting in their notice.
It's a nice idea, for sure; but most jobs can be transitioned in 2 weeks. Anything longer than that and all sense of urgency is lost ("We can transition that in a month", etc).
I've quit both ways -- with a long notice period and a short one, and short notice periods are the only times that there's been an actual transition plan.
