Fair and valid, but worth noting that these drives are door-to-door, not just advanced highway cruise control.

Any idea where one might find a trusted source for data on the robotaxi performance? Especially curious about the latest self-driving models, rather than historical performance.

The US has used one-way "drones" since the 80s or earlier. The entire Gulf War in the early 90s featured a ton of tomahawk cruise missiles. The only real change is that the new shaheeds are way cheaper, slower, and smaller, but can be spammed in larger numbers.

That has made US overflight of Iranian territory uch riskier than was expected at the beginning of the conflict, and it's notable that the US has continued to rely heavily on stand-off munitions.

I’m not saying cars shouldn’t ever exist. The ‘last mile problem’ is a thing, and proper self-driving cars could be good for part of that (especially after a train and bus if you have lots of stuff). But you want to sleep in a vehicle with lots of storage space while driving across the country? That’s called a train. Nothing new needed.

Step 1, get to the local train station in my town. There are 6 trains daily between me and Portland. Also, amtrak on the cross country trains requires the bikes to be in a box, in storage cars.

So I gotta get a large bike box, and get myself, my bike, the box, and some tools to break it down to our local amtrak station. Then partially dissasemble the bike, and box it. (of course, our train station has room in it for 5-10 people, and most sit outside, uncovered, which is fun in spring.)

Then, get to the main Portland Train station, with my bike box, and backpack with my stuff and tools. Wait up to 9 hours for the hawaitha train. (its often many hours late, and only leaves once per day).

Load Bike in cargo car, and then board train late at night.

Wake up around 5am, (or later, if train is behind schedule) and disembark at Glacier, re-assemble my bike. Figure out how to get it, and the box (i'll need it for the return trip) to a hotel or AirBnB.

For the return trip, its about the same, 1 daily westbound train, that is usually hours late, then hope you get to portland before the last train for the day leaves for my town, or else find a place to stay with a bike, backpack, and bike box in the sketchy area around the trainstation...

Or, hop in a car with a bike rack, and drive 10 hours. Which is easier, and MUCH cheaper if I split the cost of gas with someone else. So 2 extra travel days back for vacation, and much less stress.

In any case, what vehicle infrastructure does the government fund today that goes away if you expand rail service? I still need to get to my house, and I don't want to live anywhere near a public transit station. Is the pitch that we get rid of the highway system entirely and make all intercity travel rail or plane?

I'm also not concerned with or pretending that alternative modes of transport are full replacements of cars; basic comparison of the modes obviates that.

I also don't own a car, and if I did I wouldn't consider it precious, or be worried about anybody taking it away. Public mass transit advocates always go there though, it's a pretty common ad-hominem-adjacent implication. Cars are just generally a better experience. They go from A to B, and they don't have other people on them. Those factors make them obviously desirable.

If somebody comes up with a teleporter I can install at home I'll use that instead. Maybe then I'd consider that precious, or even be in love with it. It would save me a lot of time.

All of that aside my main point is to push back against the idea that more trains solve any problems with US transit, especially looking forward even a little into the future. They're complicated and time consuming to build out additional infrastructure for compared to an airport, and solve a transit gap between self-driving vehicles and air travel that will likely increasingly narrow.

Aka your pure fantasy that you present as fact. Of course there might be idiots who may claim to "seize the means of private transportation" or something like that, but let's not pretend it's a serious argument.

> Cars are just generally a better experience. They go from A to B, and they don't have other people on them. Those factors make them obviously desirable.

Can't see anyone arguing against that.

> All of that aside my main point is to push back against the idea that more trains solve any problems with US transit, especially looking forward even a little into the future.

Strange then that Northeastern Corridor whose validity you immediately called into question, keeps increasing ridership.

> solve a transit gap between self-driving vehicles and air travel that will likely increasingly narrow.

Of course they don't for many obvious reasons that start with words like "capacity" and "throughput".

It's also funny and ironic that you imagine the fantasy argument of "we'll take your cars away in favor of public transportation" and then literally arguing for taking away any and all alternative modes of transport except cars, and especially except cars owned by private companies (I do love the coming era of arbitrary surge pricing at any convenient time).

I realize I could have phrased it better, but I was not talking about anybody seizing the means of private transportation. I was talking about train fetishists cheekily implying that people love their precious cars, which you did.

I also haven't argued that trains don't carry more people at once, I've just said that they suck in every other way. I also haven't argued to take away anything. Trains already exist almost everywhere they make sense to exist, which is apparently to funnel people back and forth in Japan and a couple of northeast US states on entirely private rail. It's honestly hard to find anything in your reply that intentionally or otherwise is actually responding to anything I said.

It's a fascinating psychological and sociological question: why Americans think that the only possible state of things is the current one as it exists in the US, and why they are completely incapable of imagining any ither possible solutions.

E.g. "Trains already exist almost everywhere they make sense to exist, which is apparently to funnel people back and forth in Japan and a couple of northeast US states".

Yup, country the size of US eastern seaboard with comparable population only needs to "funnel back and forth". And Northeast Corridor just apparently appeared out of nowhere because it somehow was needed exactly there, and not anywhere else in the US, and also "just to funnel people back and forth".

Unlike cars. Americans don't need any other forms of transportation because they are physically incapable of imagining other forms despite success cases existing even in the US, and proven to work even in the shittiest third-world countries.

It's not unlike talking to a brick wall.

Adieu.

But this fellah seemed to have that part figured out: Bike to the train station, and take the bike on the train. That part seems straight-forward. The train stations were near-enough to where they wanted to start, and near-enough to where they wanted to be.

The problems they lament seem to revolve chiefly around the specifics of taking the bike on a train, and the limited schedule of the train, and the lack of adhesion to that schedule.

Those problems wouldn't be improved if the vastness of the US were reduced, would they?

I'm even planning a Eurotrip by train this summer with some mates, I'd say the distances here are comparable to get from NL to PL for example.

And besides, how is it that the US is "too wide" for trains to work, but apparently building an equivalent highway system is perfectly possible? China is also a massive country, yet they have incredible passenger train options to get cross country.

Not sure what you mean by the last bit - the US already has that highway system, and the local roads serving the last mile, because that last mile infrastructure already has to exist to get from public transit to your house.

China doesn't have the exact same problem because so much of the country lives in dense wall-to-wall housing, which sucks no matter how you spin it if you like having any kind of space to yourself.

No it can't, because cars aren't allowed on the streets around my house, with the exception of emergency vehicles and logistical vehicles like moving or delivery vans. The closest spot where a taxi could stop to drop me off is a lot further than where the bus or trains are. The closest parking space is actually a good 200-300m away from my door, reserved for residents so also always full, whereas I have a bus stop literally in front of my door and a train station 20 steps from it. I can also rent a bicycle 24/7 from the train station if all other modes of transport fail me (and I didn't have access to my bike for whatever reason).

Same in the center of the city, you cannot get to many places by car. A deliberate choice, for example when we dug out the hideous polluting highway and replaced it with a canal instead (which funnily enough was a canal in the first place before they made it into a highway). Utrecht is a perfect example of gov't realizing a mistake it made with car-centric design, doubling back and correcting it in a way that increases the QoL of every single resident of Utrecht.

https://www.theguardian.com/world/2020/sep/14/utrecht-restor...

This isn't even to say that the Netherlands is some kind of dystopia for drivers, if anything drivers here tend to be happier since they don't need to contend with a bunch of other people on the road, and more than half the country drives anyways.

> Not sure what you mean by the last bit - the US already has that highway system, and the local roads serving the last mile, because that last mile infrastructure already has to exist to get from public transit to your house.

My point was that building out the highway system was a deliberate policy choice made in lieu of a strong passenger rail/public transport network. Had they focused on making passenger rail more viable, then we'd be talking about the opposite world here, where building highly space-inefficient and expensive highways would be a ludicrous proposition.

> China doesn't have the exact same problem because so much of the country lives in dense wall-to-wall housing, which sucks no matter how you spin it if you like having any kind of space to yourself.

We're talking about cross-country lines here, if anything it's even more absurd that the Chinese can have such a strong rail network when the majority of the country has no use for the lines serving the far-west of the country where there aren't that many people. Whether the cities are shit to live in or not is a separate discussion altogether.

Suppose in some hypothetical future, we can take the (expensive) train to our destination in Somewhere, USA, and it drops us off.

So there we are, at a train station in Somewhere, USA.

What happens next? A bus? Light rail? Uber/Lyft/taxi? A friend who has time to show up? Renting a car? What's our next move? (Lots of destinations don't have much for local public transport.)

For contrast: When I drive myself to Somewhere, I've still got my car to use after I get there. I can go anywhere I want to go, at any time I choose to do so, and I can bring as much stuff and as many people as suits me without much additional cost.

I don't have to wait around for a train. I don't have to deal with checking luggage, or retrieving luggage. I can just pop into town -- with my car -- and set forth to do whatever I want. The bags can ride along with me until I get to wherever it is that I'm crashing for the night and until then, they don't present any particular burden at all.

I might have been better-rested if I took the hypothetical train, but getting dropped off at a train station isn't a very complete solution.

I wonder if there are psychological studies on why Americans en masse cannot even perceive the idea of there being other transportation options than cars and (to a lesser extent) planes. Even though in the rare cases when someone manages to provide a well-planned alternative Americans do use it, see Northeast Corridor (2200 trains a day, 15 million passengers a year, 14% of intercity traffic, replacing most air travel between some of the cities).

---

Note: it's both funny and sad reading about the state of anything in the US that keeps pretending it's not a third-world country.

For example, Empire Corridor, passenger rail corridor in New York State running between Penn Station in New York and Niagara Falls: "In the 1890s, the Empire State Express between New York City and Buffalo was about 1 hour faster than Amtrak's service in 2013." (Wikipedia)

I wonder if there are psychological studies on why non-Americans cannot perceive why trains are genuinely an undesirable solution in the US looking forward in 2026. America could have trains if it wanted to; we don't because they aren't actually a forward-looking solution to transit in the US given all other constraints, such as having spacious homes instead of being packed in like sardines like all the "first-world" examples of public mass transit.

That's not usually a problem in and of itself. A person can believe that a Kosher diet is best and that's OK. It's also OK when a person believes that bacon cheeseburgers are a necessary spice of life.

They can find joy in being spread out over an expansive rural property, with room for some chickens and a whole fleet of cars and to serve dinner to 34 guests on a holiday. A person can also appreciate the rote efficiency of a sleep tube apartment and spend their waking hours not in their own space, but in spaces that are shared with everyone.

It's part of the human condition to have strong opinions. It's often OK that these opinions aren't compatible with eachother. It's good to be tolerant of others' ideas and practices.

It only becomes particularly problematic when folks stop being tolerant and start projecting their opinions onto others. Discourse can devolve pretty quickly when that happens. It can degrade to places well beyond "No true Scotsman" namecalling and invocations of Godwin's law -- whole wars have started over differences of opinion.

---

Now, of course, there is are local elements as well. For instance, a person who grew up in Vietnam may have a very different idea of what dinner consists of compared to the ideas of a person who grew up in Guatemala.

Regionalized differences of opinion shows up very often in online discussions because people of cultural different backgrounds get to converse together. It's usually OK. Sometimes, it isn't OK.

A lot of the Web is US-centric, or at least in English. A lot of people in the world speak English as a second language because that's the language that Americans use, so that's what they were taught. The news, worldwide, covers whatever it is the US is doing. And when I went to public school in Ohio [USA], where we were taught over and over again that the US is the very best place on earth and were stand and pledge our allegiance to the flag, and our country, every single morning.

That kind of shit all tends to promote an us-vs-them response on all sides, instead of tolerance and acceptance.

That's not a particularly useful operating mode, but it's what we've got.

---

Sometimes, it seems like it's fun to make fun of Americans with their big cars and their dumb flimsy houses made of sticks, paper, and stone dust, with the worst electrical plugs that carry the most uselessly-low voltage. (Those Yanks can't even figure out how to build a proper kettle! Look at them, all driving to Wal-Mart instead of just walking to the shops! Wasting away on the freeway commute when they could be enjoying a nice Schnitzel on the train!)

And since so much of the web is US-centric, it's easy to imagine that people sometimes find themselves surrounded by Americans that they variously find to be unsavory. And if that happens, then it may be the case that they don't like that very much.

---

But I don't pop in over to European- and Asian-centric online forums much. When I do, then I don't spend any time at all making fun of their cultural or societal differences.

In fact, I avoid starting that kind of confrontation at every cost. I avoid this simply because I'm not a complete piece of shit.

The lack of other public transportation is already real in much of the US.

(If we're not adding trains, then there's really nothing here to talk about in this context -- is there?)

You're saying all this as if this exact scenario isn't solved in plenty of places across the world? You take the bus or tram or metro or cycle or just walk if it's close enough. If the city is actually built with public transport in mind, not just a single bus line that runs every 2 hours bolted on as an afterthought, those options can be easier and faster than finding a parking spot, unless you feel entitled to park your vehicle anywhere you please to the detriment of everyone who isn't you.

Where I live in the Netherlands it's faster to bike most places than driving, because we don't solely cater to drivers and don't devote half the city to letting people store their cars. Even up North in the villages you can still get around by bike, since cycling lanes are dead cheap to build and maintain and can go down in the middle of a swamp if you needed to.

> When I drive myself to Somewhere, I've still got my car to use after I get there.

Sure, and when I cycle to Somewhere, I've still got my bike. Same logic, except I can lock it to a post and forget about it rather than needing to find a dedicated slab of real estate specifically reserved for my vehicle's existence. And if I took the train, I can rent a bike when I get there, which is a thing that exists in basically every city that actually invested in making it work.

> I can go anywhere I want to go, at any time I choose

That only holds true because decades of car-centric design have made it so. In the Netherlands you couldn't just go anywhere you wanted by car, because there are plenty of streets and whole areas where cars flat out aren't allowed, because we actually prioritize the people who have to live with those infrastructural choices over random passersby who don't want to be "inconvenienced" by having to walk 5 minutes or share the road with someone who isn't also in a car.

If the US bothered to build out the infrastructure, you could go anywhere you wanted to go via public transport as well.

What happens in your scenario if you can't find parking anywhere near your destination and the only option is lugging your bags along roads that weren't built for pedestrians? I know I've been in similar situations in the past where I had to drive around for fucking ages trying to find a single spot, I definitely would've preferred walking than that whole circus.

> I might have been better-rested if I took the hypothetical train, but getting dropped off at a train station isn't a very complete solution.

Right, in a country that gutted its public transit and zoned everything to be car-dependent, a train station by itself isn't a complete solution. That's a policy failure, not an argument against trains.

Public transit != mass transit. There's no reason that self-driving vehicle fleets can't be municipally subsidized and provide a dramatically better private experience than any mass transit anywhere in the world, and we already have the infrastructure for it.

Oh no. The number of vehicles carrying 1-2 passengers through a bottleneck that is entry/exit to the venue doesn't matter at all.

Even though it's a trivial question of capacity and throughput.

Also why on earth would I want the gov't paying fucking Google for their robotaxis, rather than having them put that money into public transport which serves everyone else and not just the pockets of some silicone valley douchebags?

This misses a couple things:

- As an individual going between destinations, my mind is on my travel and not focused on the externalities of how optimal or efficient the transport is for other people. The fact that a train can fit more people is a downside, because being packed in with the public while trying to travel is not actually desirable. This is why even trains charge for expensive private rooms.

- Whether a private company or Google provides it is immaterial to this as well: I'm not chasing after Google trying to pay them, it is just literally more desirable to travel alone and privately whoever is providing it. I have no problem if the government wants to build its own robotaxi fleet, maybe it'd be better, maybe worse, let God sort them out. I'd use the one that has better service and fits my budget.

Seemingly very rare that anyone argues for public transit expansion based on any of the following:

- They like being in a box with strangers for a long duration, where bad behavior of just one individual becomes stressful and inescapable. You can increase funding for social services to reduce the number of homeless people responsible for this disruption, but you can't stop strangers from farting.

- They prefer the lesser convenience of having to plan 3+ legs of a trip instead of 1, e.g. having to walk to/from a transit station at both ends of the track.

Trains do some things better that matter to me as a rider: they are faster. They have food service and bathrooms. But "look how packed in we could be" is not an argument that's going to convince anyone who has a choice.

Because this creates a huge backlog and queues of cars. Even at 4 people per car a moderate 16k-person event needs 4 000 trips from the stadium in a short time.

Also because this creates a huge pressure on the system. Those 4k trips will starve the system in other places because companies won't just have a few thousand idle cars just laying around in case of events. Welcome to surge pricing.

BTW a similar pressure exists at peak hours when everyone leaves for work, or from work. Two trains carrying 2000 people with 15 minute intervals will need 500 cars minimum (4 people per car) for the same trip in the same direction.

2000 employees at Spotify office in Stockholm will need 500 cars minimum to take them home. In the center of the city. When other offices also leave work for home. Lol.

(In Stockholm subway carries 1.3 million passengers a day. Good luck replacing this to 1-4-people per car)

> Seemingly very rare that anyone argues for public transit expansion based on any of the following:

<lists imaginary reasons no one has or argues for when talking about public transit>

- Those events do happen today exactly as described and nobody dies.

- I'm not sure why those trips would starve the system, or why companies would not have idle cars lying around "in case of events". Where there is space for events to happen, they happen all the time. Surge pricing solves immediate high demand, over a longer window there's no reason to believe that companies would essentially leave money on the table by not having a larger fleet, especially if that fleet can be trivially re-allocated automatically over areas much larger than a single city, or state.

- In cities I've lived in, people don't just leave work exactly at 5pm like robots and immediately congest the roadways.

- This is a parallel concern, but why is a company like Spotify of all things not fully remote? Why do any of those employees need to bombard transit at all?

- Buses are also cars that don't require additional infrastructure to carry more people

> <lists imaginary reasons no one has or argues for when talking about public transit>

My point was that nobody ever argues those things because they're the worst parts of public mass transit, and they are among the most self-evident justifications for private transit. Instead it has to be "The US can't fathom" and other broad strokes like that because the idea that it can and simply rejects it is harder to accept.

Ah yes, the great counterargument of "nobody dies".

> I'm not sure why those trips would starve the system,

"I'm not sure why huge surge in demand in one part of the system would not starve other parts of the system."

> or why companies would not have idle cars lying around "in case of events".

Because companies are not in the business of having huge fleets of expensive hardware just sitting in extra rented garage space just for these occasions.

> over a longer window there's no reason to believe that companies would essentially leave money on the table by not having a larger fleet,

Why would they care when there's no public transportation to speak of, and all are stuck with their fleets of cars, as you so desire?

> - In cities I've lived in, people don't just leave work exactly at 5pm like robots and immediately congest the roadways.

That's why it's called peak hour, not peak millisecond. Just one more lane, and a a few hundred thousand cars on the road should fix all problems.

> This is a parallel concern, but why is a company like Spotify of all things not fully remote?

This is an irrelevant concern because Spotify was just an example of a company having an office in the center of the city. One of hundreds of such companies, with thousands of people.

> Buses are also cars that don't require additional infrastructure to carry more people

Keywords: busses, more people.

> My point was that nobody ever argues those things because they're the worst parts of public mass transit

You either argue for busses, as above, or complain about "oh my god I need to sit with some strangers in the same place" and "oh my gid it's inconvenient to plan a trip with more than one stop".

Yes, that's about the extent of the arguments. And yes, that's why it's invariably Americans who cannot even begin to conceive other modes of transportation.

As an American, it's far easier to imagine autonomous robot driven road trips than it is to imagine a government that is competent enough to build passenger rail networks.

Fees are also very high for such a slow option.

As for the future, well... it is bleak. This administration is actively trying to block transit expansion, presumably due to their undying affection for the fossil fuel industry, going so far as to withhold funding from already awarded grants to regional rail.

So while the northeast can sort-of pull it off due to its relatively compact nature and history of more progressive policies, this leaves the vast majority of the country in a no-mans land.

Building new rail projects in the US is very hard because of capital costs and regulations like NEPA (and CEQA in California) which require environmental review for everything. Brightline in Florida was able to get around this by working in an existing highway ROW.

And will probably go bankrupt this year: https://www.wlrn.org/business/2026-01-23/brightline-business...

We used to have passenger rail. Even the desolate nowhere of semi-rural Ohio was well-served. Street cars to get around town, inter-urbans to get between nearby towns, and proper passenger trains to get to points far-away.

It didn't work out. There's reasons why it didn't work, like the literal conspiracy between General Motors and Firestone Tire that deliberately sought to destroy it.

Whatever those reasons were, they are are behind us. So it may seem superficially easy to just put it all back... but it isn't.

When the lines stopped being used, we tore them out. They're gone. And where the lines are gone, old stations are also mostly gone. Cities had once been built around (and because of) rail, but were subsequently built for cars as time marched forward and things continued to expand.

In some cases, whole communities have disappeared in the transition away from rail. In many other cases, we let our central stations decay and rot or demolished them to make space for things like convention centers.

So what's left is what we have: We have cars.

It's easy for me to see a future where I can buy a car and curl up in the back seat with a movie (and maybe a cocktail) while it ferries me from A to B.

That's a future I might actually live long enough to see, and it appears to be inevitable.

And I'd love to be freed of the chains of having to drive myself from A to B.

But I'll be dead and buried before we get passenger rail to be even 1/10th of what it once was.

So I choose to dream practical dreams. I can only play the hand I'm dealt.

Low-density residential sprawl is mostly water-inefficient because it allows people to have the ability to have a garden that they water, you don't inherently use more household-internal water if you live in a suburban house compared to an apartment. Most of the energy efficiency issues are also directly related to low-density residential zoning allowing for more physical space for a dwelling than an equivalently-expensive dwelling would cost in an expensive, dense urban area. In short, the things about low-density residential neighborhoods that are less energy efficiently mostly don't have to do with cars and mostly do have to do with goods that people actively want and can only afford outside of dense urban areas.

Low-density sprawl in the American style is impossible without cars. Streetcar suburbs could exist but those are necessarily more concentrated and again need less road coverage.

Nor can you say the sprawl is what people "actively want" when it's illegal to build to any other pattern in the vast majority of the country.

Which is more or less a non-issue east of the Missouri river

Why do North and South Carolina have to shell out to manage all their runoff in federally prescribed ways just because all the states on the Colorado river can't get along?

Why does St. Louis and every other "give and take from river" water district have to mandate all sorts of federally prescribed minutia written around assumptions of aquifer depletion? It's not like a lack of low flow toilets is gonna affect anything other than the size of the local water treatment plant.

But one huge factor that you have to contend with is the randomness of the tragedy of the commons problem on public transport / shared transport. A train journey can be blissful to sleep on right until a loud group gets on and sits across from you and there's no seats available to move.

I think this is something that can't be overlooked, especially if you're talking about something like a short trip where if you don't sleep well en route, quite a large proportion of the trip time is going to be affected. Having a private vehicle where you can guarantee control of your environment is a really huge plus.

But that's precisely the problem, it only takes a very tiny minority to change this. If one group, one person sometimes, in a carriage of 50 people decides to go against this, then that's that. It's not even particularly common, but it happens, it's random, and so it's just something that must be contended with.

The majority consensus is to desire a peaceful environment but do nothing when it is violated.

That is why it breaks down — once it is discarded in a melting pot the cultural expectations are unclear and it seems you’re at least initially dependent on the state or mob dynamics.

I think you have to go further upstream socially - there are people that should not be free, but are. Public transit has not just loud talking or music on phones, but the mentally deranged, babbling, even actively drug using population walking a knife's edge between erratic and aggressive behavior. From my POV it's so far past a stern stare on the US west coast that the suggestion comes across comical.

There's immense social capital and NGO patronage at work surrounding 'homeless' - and I parse that as mentally ill now, as it's an insult (IMO) to the homeless who are perfectly capable of respecting others and participating in the social contract.

I’d much rather be around the noisy-but-relatively-harmless person than someone with so little regard for their fellow man.

Or longer if they show they are unable to reform while in prison.

>penalty for being “noisy” on a train

That is ignoring the second order effects of the situation. This noise is holding back the flywheel of public transportation that could have economic benefits to millions of people. The penalty is for the enormous cost of holding the rest of society back. The person who has such little regard for their fellow man is the one who is letting the few ruin society for the many.

Fortunately, your views are wildly extreme and unconstitutional. Regardless, I hope you find some empathy and some peace out there.

Answers which involve removing oneself from society (by entering a private car) are not good answers. And when you factor in the externalities, you're just displacing "I'm upset, possibly even unwell due to sleep lost" onto "we replaced 90% of the local natural environment with pavement and paint it with crushed human beings every single day".

Trying to take something like a windsurf board on a train, and then having to navigate multiple train changes along with whatever other baggage you have makes it a non-starter.

The "last mile problem" you mention is unresolved when it comes to getting from the closest public transit stop to the actual destination (frequently in a park or even off road).

And finally, the final cost to the rider would be significantly higher, as sleeper trains are not cheap.

I think America could do quite well if it focused on public transit in and between densely populated areas. Fewer cars in cities could make for denser cities, which in turn could allow for even more public transit. But outside of population centers, America is much more spread out than Europe, meaning that trains are less economical, and often wouldn't get the ridership that would allow them to make sense.

The point of public transit isn't to handle 100% of all trips. It's to handle the most common trips in a more efficient and scalable way.

1. https://www.apta.com/wp-content/uploads/2026/03/2025-Q4-Ride...

That sounds like hell. Bus seats are just as tight as coach class airline seats, and unlike trains there isn't even an option to pay exorbitant pricing for a sleeper compartment.

> Bus seats are just as tight as coach class airline seats

IME bus seats are first class in padding and support, a bit wider than coach, and with more leg room (though I'd need to measure to confirm). Much more comfortable than airplane coach - and without the air pressure, vibration / noise, and humidity problems.

> exorbitant pricing for a sleeper compartment

It just depends on the train, how far ahead you pay, etc.

That's a $41/month subscription every citizen's paying no matter what. When we're pulling cash on that volume from everyone's pockets to build lavish infrastructure literally up to people's doors (vastly more road square footage than housing+school square footage combined), of course folks are going to say "nothing compares" -- because nothing does compare. Which stinks (imagine if we'd focused a century of spending on rail at rates like that; damn), but it is what we have at the moment.

I hope your analysis is correct and subsequent passenger rails will be much cheaper!

No public transport system that remotely makes any kind of economic sense, either in terms of infrastructure or operational cost, can replace the established network topology that exists for cars in the US. The connectivity is much more like a mesh than a hub-and-spoke model. Even though the US has a strong regional jet system that connects arbitrary nodes in that graph it still doesn't entirely avoid the "last hundred miles" problem.

A lot of American long-distance travel is not between two big cities. Even in Europe, similar kinds of routes have no train service and limited bus service.

I want to move on my schedule and convenience, I don't want to have to warp my day to day around someone else's departure schedule.

And there's nothing wrong with it! I take detours on road trips all the time following “Historic <thing> →” signs or just because I see something interesting in the distance and want to go check it out. On a train journey I'd just have to watch them pass by.

https://en.wikipedia.org/wiki/Killing_of_Debrina_Kawam

https://en.wikipedia.org/wiki/Killing_of_Iryna_Zarutska

https://www.cbsnews.com/newyork/news/nypd-men-pushed-subway-...

The posters advocating for it and stating Daniel Penny should’ve encourage copy-cats are violent and sick people.

I love the confidence with which you give your answer though! Europeans famously underestimate the American West, which is why they often get into serious trouble (or die[1]) at alarming rates out here.

[1] https://en.wikipedia.org/wiki/Death_Valley_Germans

So, let's say you take public transport from SF to Yosemite/Los Angeles. Now, how do I cover the last mile (or even multiple points)? Take more public transport? Hitchhike?

The reason long-distance public transport works well in Europe is that there is good local public transport in both the source and the destination cities. When that does not exist, you are better off driving.

Amtrak could do the same thing but because of how Amtrak is organized in not sure that it would be possible. Most of the current Waymo geos are not connected by Amtrak directly and require transfers.

It's an electric car, so carbon emissions are low.

Most of the route will be in rural areas in the middle of the night, so the impact on traffic will be minimal.

As for the cost to build and maintain the roads, they are already needed so rural areas are accessible. Wear and tear on roads and bridges isn't much of an issue since heavy vehicles like trucks cause massively disproportionate damage[1]. (A bus might actually be worse than the equivalent number of cars in this respect.)

---

[1] See https://blog.ucs.org/dave-cooke/trucks-cause-the-lions-share... . Some studies show that damage varies with the fourth power of axle weight.

One of the famous sleeper trains in Europe (Nightjet Vienna-Amsterdam) is often booked out weeks (sometimes months) in advance, costs as much as a plane ticket + hotel room or more, and you have a decent chance of being told (as you show up in the evening) that unfortunately one car is missing tonight and you have the option of a full refund (screwing up your entire trip and having to book a last minute plane ticket), or you can take a 50% refund on your 255 EUR sleeping ticket and spend the night sitting in the shared seating part on a seat that would have regularly cost 35 EUR. This was something that on some routes was happening routinely for over a year [1].

The night train from Switzerland to Malmö was cancelled (after tickets had already been sold) because the Swiss government decided to not subsidize it.

Trains like this offer zero flexibility (you have to book a specific train weeks in advance), go where they go which is a very limited route network, and even in Europe with all the environmentalists, rail networks, shorter distances, and massive government subsidies, they don't seem to be able to run them very frequently or on many routes.

Calling them equivalent or a replacement for self-driving cars (which would take the passenger where they want, when they want) is disingenuous and isn't going to magically convince people.

[1] https://www.srf.ch/sendungen/kassensturz-espresso/espresso/f...

I do have to admit the steer by wire system with adaptive sensitivity is neat.

But thank you for the kind words. I do enjoy it a lot!

It is a very fragile truck that is very expensive to repair because stainless steel is a terrible material to make a truck body from.

https://m.youtube.com/watch?v=PK_EJ3DyiiA

https://m.youtube.com/watch?v=Yn4XzbMf9nY

"When I go I want to die in my sleep like my grandfather. Not screaming and afraid like the passengers in his car"

Humans and Control System Models need feedback to operate, and worse still... when any input into the vehicle's controls produce zero results, you will spin out.

My concern with a model in these conditions is that it wouldn't recoginize the fact that other cars were in the ditch and that it should probably slow down

The main limitation is still sensors in the snow, but it seems to not be that big of a deal to build sensor packages that are better at seeing in the snow than a human is.

Being able to plot a series of inputs that can more efficiently use available traction than a human doesn't prevent you from blundering your way into a dumb situation where the laws of physics dictate that the only possible outcomes are various flavors of bad ones.

It's not clear how often the software will chose poorly and need to brute force its way out with traction/handling. The fact that they seem to be hedging against this by putting the hardware on particularly performant cars indicates it must happen enough to matter or be rare but bad enough to matter when it does happen.

Waymo will probably also rack up a ton of technically not at fault accidents by being obtuse in traffic since there's when there's snow there's a lot less margin for the "two people trying to pass each other in a hallway" type missteps that behavior tends to create.

No, this is not a "works in a textbook" take. The path planner is aware of surface conditions on the road. This is already a big deal because otherwise the AV would not be safe to operate in the rain.

Waymo will not be touched in most of those accidents. I've been driving in snow long enough to know that there are always going to be idiots who drive too fast for conditions and lose control of their car; I'm sure some of them will blame Waymo because it's nearby. I once watched a guy with a TX license plate spin out on a perfectly empty, perfectly straight freeway. Waymo doesn't really need to do anything for human drivers to crash.

You couldn't see anything. As soon as there wasn't a car 20 yards in front of you, it was a complete whiteout. Ice built up on the wiper as quickly as you could possibly reach out of your window and clear it. Radar would probably be nice, but I don't think it'd be enough to keep driving. The cameras and lidar would be an absolute wreck.

I'm sure we'll get there eventually, but that is really the final frontier for AI driving I think. Waymos aren't even allowed to drive in a snowstorm right now. I suspect that you'll be dealing with Caltrans closing the pass for the rest of your life.

http://cadcd.uwaterloo.ca

Disclosure: My grad student advisor.

Even if you also have good winter tires on, if your level of "caution" could be best measured as normal to high, sometimes it's a judgment call on when you want to pull off to the shoulder for 45 seconds to let a bunch of vehicles behind you pass. I'm not sure this is something any automated driver has been configured for. Or just generally to deal with driving when the road condition could best be described as "two only partially visible ruts in the snow where the tires of previous vehicles have driven, with snow in the centre".

Same thing in somewhere with a climate like upper Michigan or in Maine.

I love Waymo in other cities, but it'd be especially helpful here during the 1 day every other year that we actually get any snow ... if we ever get snow here again.

The second-order effects of that could be pretty wild. If people stopped owning their own cars, we wouldn't need houses with garages and driveways. It'd favor dense development with loading zones rather than parking spaces. It'd also be a big boon for EV adoption since the cars are all owned by one corporate owner and all go home to a centralized depot to charge at night rather than needing to retrofit EV chargers onto everyone's living situation. (Indeed, Waymo runs an all-electric fleet.). There'd be a premium on very reliable powertrains, since the cars might easily put 60-70K miles/year on them instead of the 10-15K that is typical of passenger vehicles. I dunno why Waymo went with Jaguar instead of Toyota, but perhaps "EV" is the explanation. Cars would wear out in 3-5 years instead of lasting for 15-20, and so you'd always have the latest hardware and technology on the car.

All the money we spend on traffic enforcement would become pointless, with audits of the software becoming a more effective use of dollars instead. But that blows a hole in many small local PD's budgets, many of which use speeding and parking tickets to raise revenue. Municipalities would likely find themselves powerless at regulating Big Self-Driving Rideshares.

The third-order effects are interesting as well. Once all cars on the road are self-driving, why not have them draft each other and physically link up to improve power efficiency and safety? You might even call such an arrangement a "train", blurring the line between road and rail transportation. But then, if you've got docking and linkage mechanisms, why not put the boundary between the electronics & powertrain and the passenger compartment, like the Rivian "skateboard" platform? You could return to private ownership of the passenger compartment - where, after all, some people like to store all their junk - and then have the rideshare own only the means of locomotion. Then you could extend this to other forms of locomotion like elevators, airplanes and ferries, so that your passenger compartment could just drop down an elevator shoot, onto a waiting self-driving car, which links up with others to become a train, takes you to the airport where you're loaded onto a plane without ever having to board, and then your pod deplanes and a self-driving car takes you straight to your hotel, where you now have transportation to wherever you want to go.

The future looks an awful lot like intermodal containers for people.

Currently I live in a city with an OK pt network; in the a high density apartment. I chose this because I can catch a train to work, go drinking locally, and I dislike driving.

If I could rely on a driverless car, i would happily live further out in the suburbs, as the driverless car removes the upsides of density more than anything else... And I think this is a common sentiment, driven mostly by housing costs.

And then you have the cost of a trip, of owning vs rideshare... If its my car I can choose the furnishing, pay for fuel or power however is most efficient for me (eg solar), not have to pay for cleaning, and store my stuff in the car.

Love this concept.

As self-driving vehicles become a larger share of road use, roads can be more efficiently designed just for them: no speed limit, just 2 strips of pavement for the tires, no signage or striping, etc.

We'll just build the cars with parts that seldom fail. And we'll make them very strong, so that the only risk from hitting a deer or even a cow is a splash of gore.

That should help eliminate the need to turn. A loud horn and flashing lights will do pretty well for any errant humans that cross the path.

We can even reduce rolling resistance by using steel wheels instead of rubber, and we can make the road a surface of continuous steel for durability.

We can even hitch the cars together so they can't collide with eachother and they can collectively share the propulsion load. (Maybe even with automatic micro payments, so a car with low battery can pay the others to help it along.)

What would we call this thing?

> You might even call such an arrangement a "train"

Joking aside, though, the big issue with trains is last-mile. The road network covers a lot more land than the rail network does, and can reach places that trains can't. And this seems to be inherent to the physics of it, driven (hah) by cars ability to turn where trains cannot.

Mass transit enthusiasts love to gloss over the very real convenience issues that mass transit has, saying "Well everybody should just live next to the train station." The world doesn't work like that. Hence why I think a hybrid system of dockable autonomous vehicles that can be linked up into a train in high-throughput thoroughfares gives you the best of all worlds.

The world as a whole, and particularly the US, maybe not, but it does actually work in urbanized dense cities.

So a Tesla?

As for autonomy, Waymos have LIDARs which at least provides more redundancy.

I see these as different design tradeoffs so no judgment implied.

$10K for full autopilot on Tesla in 2018 was essentially a fraud. I have since then learned not to trust anything Elon says.

Always feels a little weird to read a comment that is plausibly going to end up being referenced in a future news article.

My Tesla Was Driving Itself Perfectly—Until It Crashed

The danger of almost-perfect tech

The supervised driving is great, I have used it with my Model Y, but let me know when the car can pickup and drop off my kids at their school and activities. Like Waymo can. Then, it will be self driving.

I honestly disagree with this article - it seems to be conflating "economic activity" with extremely high-end real estate sales data in certain neighborhoods. The city feels much more alive in the past 12 months than it was previously, and there is a lot more energy and public events. If there's anything specific to complain about, it's that the AI boom has led to the 996+ crowd staying inside and not contributing much to the local scene, but honestly that's probably fine with everyone involved.

But the energy is a bit infectious. I'm happy to see SF on the rise again.

What does this mean?

But it wasn’t immediately obvious to me what was being discussed in this context, to me it looked like it was maybe a reference for to a certain set of postcodes or telephone number prefixes in reference to a certain set of people or some such.

I think the plus threw me.

I come here for the conversation.

Your comment would have been better without the needles reminder that search providers exist.

I think the better route is to be honest and say that database integrity is a primary foundation of the company, there's no task worth pursuing that would require touching the database, specifically ask it to think hard before doing anything that gets close to the production data, etc.

I run a much lower-stakes version where an LLM has a key that can delete a valuable product database if it were so inclined. I've built a strong framework around how and when destructive edits can be made (they cannot), but specifically I say that any of these destructive commands (DROP, -rm, etc) need to be handed to the user to implement. Between that framework and claude code via CLI, it's very cautious about running anything that writes to the database, and the new claude plan permissions system is pretty aggressive about reviewing any proposed action, even if I've given it blanket permission otherwise.

I've tested it a few times by telling it to go ahead, "I give you permission", but it still gets stopped by the global claude safety/permissions layer in opus 4.7. IMO it's pretty robust.

Food for thought.

This is recklessly negligent and I would personally not tolerate a coworker or report doing it. What's next, sending long-lived access tokens out over email and asking pretty please for nobody to cc/forward?

My broader point is that LLMs are going to need access to these keys whether we like it or not, and until we get extremely scoped API permissions (which would make a ton of sense, but most services aren't there), you have to live a bit on the edge to move quickly.

Mitigation is good, but what's preventing your sudo-privileged LLM from disabling/corrupting/deleting on-site backups either directly or by proxy via access to the DB and code that writes to it?

In the future, I could see this solved by the same "nuclear launch key" style delegation of keys. Aka in order to run certain API or database commands, the service requires both the standard dev key (presumably used by the LLM) and a separate "human admin key" that gets requested whenever a specific operation is requested. It could be tied to a biometric request or something as well to avoid the LLM hacking its way around it. Honestly this is pretty out of my technical depth but just thinking out-loud.

And, a lot of DevOps/SecOps at scale is concerned with mitigating potential rogue or dangerously incompetent employees. You don't let your juniors push senior-unreviewed code, much less let them anywhere near the keys to kingdom if you can help it.

Thats stretching the definition of 'research', it basically checks if the texts are close enough.

Delete can occur in various contexts, including safe contexts. It simply checks if a close enough match is available and executes. It doesn't know if what it is doing is safe.

Unfortunately a wide variety of such unsafe behaviours can show up. I'd even say for someone that does things without understanding them. Any write operation of any kind can be deemed unsafe.

Standard rule is you never let your developers at the production instance. So I can't see why an LLM would get a break.