The real problem with the cigarettes analogy is that cigarettes were not regulated based on harm to the user. They were regulated based on harm to others.

I remember this because I lived through it. Nobody ever got anywhere telling people cigarettes were bad for them or that they were going to cause them cancer. Making those arguments just made people smoke even more resolved to smoke, because nobody wants to be as square or a health-nut. Nobody cared, and there would never have been any regulation on that basis.

Cigarettes were finally regulated because they *cause harm to others*. And they didn't even go after the smokers, but they regulated on behalf of the non-smokers. That's why you got non-smoking sections, and no smoking in public places like airports, but still allowed in private business, then eventually you got no smoking indoors, then you got laws against smoking when kids are in the car, and finally you got laws against smoking around pretty much anybody. It was all based around harm to others. This is a much harder argument to make about "ultraprocessed" foods, whatever that even means.

There's also the fact that cigarettes are PROVABLY addictive, so marketing them was seen as exploitive, especially to children, so the taxes and advertising regulations were based on that percieved exploitation of getting somebody addicted to something. This is also hard to make for "ultraprocessed foods", whatever that is.

This. It's important to make a distinction between what feels unsafe and what actually is unsafe. Any driver that's honking at you, by definition sees you. The bad accidents happen by people who are on their phone and don't even see you, much less have time to be angry at you or honk at you. It's always important to drive, walk and roll in such a way as to be seen. Between being honked at and being dead, I'll always take being honked at.

At a recent conference in Phoenix, I watched a coffee place from a distance. I observed people lining up in a line to order coffee on a touchscreen. There was no human interaction. But there was still a human; he was behind the touchscreen wiping down machines, cleaning, and juggling supplies.

Robots were being used, but instead of using robots to do the drudgework of cleaning, prepping, and fairly mechanistically brewing coffee, and having a human up front interacting with customers, instead we had a human in the back doing drudgework and a robot up front interacting with the customers.

I understand the human is the expensive part of the operation, so companies just want to replace humans with robots wherever they can, however, I don't think society is thinking it through very well. Humans also add value that robots can't. Instead of using robots to replace low-human-value tasks (automated dishwasher, automated coffee brewer), and letting humans keep doing human things, they are replacing the human things with robots.

When we invented dishwashers and coffee machines, it was OK, because it was robots helping humans be a barista. Now, it's humans helping robots be a barista. The first one seems fine; the second one seems pretty doomed. Robots serving humans can work. Humans serving robots just doesn't seem likely to ever work out.

On two occasions I have been asked, — "Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out?" In one case a member of the Upper, and in the other a member of the Lower, House put this question. I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question.

Charles Babbage, Passages from the Life of a Philosopher (1864), ch. 5 "Difference Engine No. 1"

This is bot correct at all. You may be thinking about a different project.

California's high-speed rail is first-class HSR, using normal HSR trains from Siemens, and it has a good alignment that directly connects multiple population centers. There are many ways you can criticize CAHSR, but it's definitely HSR.

The problem with CAHSR, to be fair, is that they bit off a huge megaproject all at once, without a clear funding path. Argument 1 is that they should have started by building a single link between two biggest cities, like the French advised them to do at the beginning, then built it out from there.

The counter Argument 2 is that there are already rail links between those two cities, the land and existing infrastructure problems between those two cities are by far the hardest of the entire alignment, plus, the real problem in California is they need to provide alternatives to I-5, which is a traffic crisis impacting a lot of the state. Also, the marginal benefits per dollar go up the more cities you connect due to network effects. So, like it or not, California voters approved CAHSR only under the condition that it connect the major CA cities in the central valley and provide an alternative to I-5. When it's built, it will be brilliant. But they didn't go about it in the "how to show earliest results possible" way. The built in the "what makes the most long-term sense for the state" way.

It's a really good megaproject. It's main problem is that until now, it hasn't had megaproject funding. The election of Trump counter-intuitively gave CAHSR a huge tailwind. For one thing, they stopped even asking for federal money, and have opened up other funding sources to keep construction going. For another, they are in way too deep to stop it now. They have no choice but to complete at least the initial operating segment.

As of mid-2025 official CAHSR report
Structures: 55 complete (60%), 32 underway, 5 not started.
Guideway: 72 miles complete (61%), 28 underway, 19 not started.
Right-of-way parcels: 99% delivered overall for CP1–CP4 (2,276 of 2,294).
463 miles of the entire 494-mile Phase 1 SF–Anaheim system are environmentally cleared (construction-ready).

Forget the cost in terms of dollars...the traffic congestion and parking won't sustain economic growth, not at any cost, past a certain point, by pure simple geometric math. If every apartment in Manhattan had a single surface parking spot, the resulting parking lot would be bigger than Manhattan itself. The corollary: you simply can't have a strong urban economy if you rely on universal car transport. There are two choices for cities: Hard-cap your economic growth at a low level, at the same time permanently putting your city at the limit of traffic misery it can never escape from, or take mass transit seriously as the transportation backbone, and not as an afterthought or a charity effort.

As long as there are roads and cars, people are going to drive cars, but that can't be your only option, or even the primary option, in any healthy city economy.

The ease of doing this is not just a theory; Prime Minister Abe was assassinated in 2022 in Japan by a homemade shotgun and homemade black powder. No 3D printer was used, needed or wanted; just metal pipe from the hardware store. Such basic firearms are just as effective as they were thousands of years ago...and clearly effective enough to assassinate a prime minister in the 21st century.

I just did a quick search and nope, US prepaid cards are still a ripoff.

Basically the cheapest prepaid card available is the Walmart Green Dot card, and even that one has a $6 monthly fee unless you get $500 or more in direct deposits in the previous month.

All the other cards cost as much as $10 to buy, or they have high fees to reload like $3 or more reload fees, and some of them have several dollars per transaction fees too. Prepaid cards can be good options for buying a 1-time $100 gift card for somebody, but worthless as an account-free secure cash card like Japanese IC cards. I've had the same Suica card for over 10 years and just keep putting cash on it when I go to Japan.

I will definitely look into this. I tried visa gift cards in the past but it was not reliable...pin codes didn't work on some terminals, awkward to reload value (no cash reload), minimum spend and balance requirements. But it might be getting better now that tap-to-pay is almost everywhere.

Sounds like you are describing IC cards, like the ones they have in Japan. You buy them for cash from vending machines, reload them with cash in vending machines, and they are completely untraceable, and widely accepted for payment nationwide. Also accepted to pay bus and train fares nationwide, so you only need one card for everything. Oh, and aside from purchasing the original card, they are also free (in terms of there are no fees or charges for use).

I have 4 kids and the only way for me to give them non-cash payment is to sign them up with a bank and get a debit card account. Or I can give them cash, but they don't want it and some places don't accept it.

About the train speeds, the benefit of a train is that it will always run on time and to a timetable regardless of time of day. Putting more people on the train doesn't make it slower. In the corridors we are talking about, the interstate isn't running 80MPH constantly; during rush hour these stretches of interstate are crawling whereas the train will still be running at full speed on the same timetable. The peak throughputs of a highway is something like 17mph, but trains always run at peak throughput.

About the land near cities, you answered your own question. When approaching cities, trains can use existing highway ROW because they start slowing down miles outside the city anyway, and nearly all US cities (regrettably) already have interstates cutting right through them.

About stations, train stations are tiny compared to things like airports or even compared to highway interchanges. A single highway interchange is many times bigger than a big train station. No even to mention the size of airports. That's how Penn Station in Manhattan handles more daily passengers than any airport in the world, despite being smack in the middle of Manhattan. You don't need acres of land to build a train station; you can build it in the space of one moderately large parking lot, and almost no city in America is so dense that you can't find such a patch of land for a train station. And in cases where it makes sense, building a station on the edge of town is perfectly viable. That's what we do with airports already, and people manage to get themselves to airports. In fact you might be running the train straight to an existing airport for the obvious connectivity benefits.

So you use those same highway ROW for the trains. They have to slow down miles before they get to the city anyway. In cases where there's no urban freeways, which is really rare in the US, you can build a terminal on the city edge instead, exactly like they did with shin-osaka in Japan.

That's the same situation for interstates. There's no "at grade" road crossing on any interstate highway, by design. Every junction is grade separated, universally, by design. Yet we have a nationwide interstate system. Why is this seen as a barrier, when it's literally how we build anyway?

1) We already have public ROW all over the country for interstates and utility corridors. Not all of it is suitable for true HSR, but then again you don't need true HSR on every mile. Mere "HSR(tm)" is perfectly serviceable. Even most true HSR systems only hit peak speeds in short stretches. This is the approach you would use in built up areas.

2) the vast majority of the mountain West is actually public land anyway. You would build it exactly how we did it in the 1800s when we built the transcontinental RR...the government uses its own land and grants it to private railway companies or the public railway company.