Hacker News new | comments | ask | show | jobs | submit login
What a Real Train System Looks Like (2009) (newworldeconomics.com)
240 points by apsec112 84 days ago | hide | past | web | favorite | 199 comments

That's 2009. The Shinkansen goes to Hokkaido now, via underwater tunnel.[1]

Japan has a plan to run a superconducting maglev train between Tokyo, Nagoya, and Osaka. It's a huge job. The "Japanese Alps" are in the way. Most of the route would have to be in tunnels cut through hard rock.

It's not just talk. The first section is already running. At 500km/hr. There are working trainsets running on 42km of track, mostly in tunnels. Stations are under construction. Tunneling continues. Construction started in 2014. Completion in 2027, to Nagoya. Osaka 10 years later.

Many Asian cities now have large new monorail systems. Daegu, 30 stations. Mumbai, 17 stations. Chongqing, 64 stations. Many smaller systems, too.

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

For comparison: here in West Seattle, we will get what essentially is a tramway in 2033. It's less than 15km from downtown.

Thier map is also missing the Kyushu shinkansen down to Kagoshima which opened partially in 2004 and fully in 2011

When you look at the Tokyo railway map it looks like a totally unnavigational maze.

Actually, and despite the fact that services are provided by more than half a dozen independent companies, it's quite easy to navigate. Even when you run into vertigo when seeing it for the first time.

Each line, no matter if Tokyo metro, private metro, JR, private railway is color coded and has a fixed one or more unique letter designation.

Say, you want to go to Ginza, to just take one example. You don't even have to remember the station name. You just need to know the station codes. In this case M16 (red), H08 (gray) or G9 (orange). The letter designates the metro line, while the number represents the station for the respective line. And each combination is unique.

That, in combination with a PASMO or SUICA smart card makes orientation very straight forward and seamless.

I was able to navigate the Tokyo system on my first visit better than I could navigate the New York system, even though I had lived in New York for years. I credit the high quality signage.

Yes, 100%. That, combined with Google Maps for routing made any sort of train or metro interaction in all of Japan utterly painless.

The only time I got lost was when I couldn't immediately find a Tokyo Metro station and Google Maps sent me to the JR station on the other end of the mall. Even then, it only took me a few minutes of wandering around to find what I was looking for.

Second this. Tokyo was extremely easy to navigate considering I had done no planning until I arrived.

It's SO good, in fact, that when I was visiting Tokyo with my 12 year old a few years ago and he was bored of looking at guitars with me in Ochanomizu, I just let him take the train back to our accommodations without me.

Tokyo is so safe, friendly and easily navigable I didn't give it a second thought.

I was curious what the map of all Tokyo train lines overlapping might look like, and found this post on Reddit:


Quite impressive! It even includes the Disney Resort Line (small orange loop toward the right-hand side).

> it's quite easy to navigate

After a recent visit to London I've really come to appreciate how fantastic the signage is in the Tokyo railway system. In London I often had trouble finding exits or connections (often following signs only to seemingly run out of directions), but in Tokyo it always seems seamless.

I’ve got the opposite impression coming to London from Brussels. London is so much better. When you arrive in London by Eurostar, you walk on a nice wooden floor. St. Pancras station is wonderful, clean and easy to navigate. When I come back in Bruxelles Midi, everything is dirty, mechanical stairs are broken, nobody can find the metro entrance, most exits are closed at 22:00...

It seems that the grass is always greener on the other side.

I’d like to try Tokyo some time though, it seems so nice.

(I’m from Belgium)

I think it's fair to say that St. Pancras is uniquely fantastic, and Bruxelles Midi is unusually terrible. I suspect stations in the UK and Belgium are much the same once you average them all out.

That being said, I do think that the Tube has generally very good signage and ticketing. Its main issues are being unpleasantly loud and hot.

You are right. In London as a foreigner I’ve always found the Tube easy to use. Even when there is an interruption there is always someone to help and give guidance. The Oyster card is convenient and cheap for one day tourists.

In train I’ve only ridden on the GWR Paddington-Bristol route. I found it very comfortable although it was still diesel powered. I’ve heard the line now has been electrified up to Swindon I think. I’m a railway worker so I can ride first class for free. It’s great because with GWR you get free drinks and snacks and the seats are really comfy. But if you have to pay it’s really expensive in the UK.

Brussels has the only subway I've yet encountered that I could not figure out without local help. (And that was partially because I couldn't figure out how to buy a ticket from the machine and there was no booth with a person to sell me one.)

Notably, London's Underground is already an example for navigation designers, at least in the West. Personally I'd like to see a detailed longread about navigation in Tokyo―haven't seen anything yet while London is featured regularly in design circles, and I guess that's why people mainly have London to look up to.

Jane Jacobs talks about how cities generally fall into one of two different kinds of equilibria: a public transit equilibrium, and a car equilibrium. In each one, the preferred kind of mobility further shapes land use to make that kind of mobility even more preferable.

Most of North America is so stuck in the car equilibrium that it seems hopeless. There's not even a handful of cities that would be nice to get around or live in without a car. This seems like it would take decades to correct, even if most North Americans actually wanted to. As an semi-extremist urbanite I'm incredibly pessimistic about this.

I'm somewhat hopeful about new mobility methods like bikeshare or even rideshare, but none of them seem to be actively driving land use changes.

You don’t need a huge public transit system to make the jump. You need high density + mixed use over a relatively small area with good public transit in that area. Which lets two car families drop to one car family’s. This reduces the need for public parking pushing up density and property values. It also makes links to that core of good public transit more valuable.

The main problem in the US is you don’t get both high density and mixed use.

PS: His harping on Vienna Fairfax completely misses the point, it’s the last stop on the orange line, the parking allows a vast area to commute via the subway including many people just visiting DC. The Pentagon City and Crystal City stops for example directly connect to shopping next to huge residential sections and have minimal parking.

I’m amused by the Fairfax bit. I used to live within walking distance of that station. Almost all of the green space in the article’s image is now designated for commercial use (the remainder has housing), has plans ready to go, and is just waiting for the developer to start putting shovels into the ground. But the developer apparently doesn’t think it would be worth the cost of construction, so it just sits there.

I’m not sure what else you could do there, unless you want to have the government heavily subsidize commercial development, which seems like a recipe for empty, decaying buildings in a decade or two.

The American fascination with single use zoning is so counterproductive. If commercial doesn't work, just put housing next to it! Or build the housing first, or something. Pair with a tax designed to disincentivize land squatting.

And it doesn't need to be eight story apartments with elevator, garbage, the whole shebang; a four-family house like this would still be a massive improvement over most American suburbs. https://activerain-store.s3.amazonaws.com/image_store/upload... Plus with how expensive DC is, you could do worse than selling houses.

This one is the opposite. The developers would be happy to build nothing but housing, but the government wants offices and retail too. The developers built as much housing as they were allowed to without doing the rest (my house was one of them) and then basically bailed out.

But what the government wants isn't necessarily what the market will support. Fairfax has a very high office vacancy rate, particularly in newer buildings, so I'm not surprised that developers don't want to build buildings that will stay vacant. [1]

When I say single use, I'm not talking about the neighborhood, I'm talking about the land itself; it sounds like the developer is restricted from building anything other than commercial or retail on that particular portion of the neighborhood. Even in the overheated property markets of New York, the new all-office World Trade Center is struggling with vacancy, and Hudson Yards had to be revamped from all-office to mixed-use because there's simply too much office space on the market. And even after they did that, the city had to offer tax breaks so that companies would populate the new district.

[1] - https://www.bizjournals.com/washington/news/2018/04/23/in-fa...

That portion is supposed to be a mix of all three. It’s possbible that some modification of the mix would be profitable, but supposedly the retail is non-viable without a population of office workers to provide steady traffic.

Oddly, the massive success of mixed-use development at the nearby Dunn Loring station is being used as an excuse to sit on this project, because they say it’s too close and couldn’t compete. (Meanwhile, a massive new mixed use development is going up just half a mile down the road.)

In Seattle there are lots of buildings going up that are residential with commercial space on the ground floor. Residential is so profitable that the commercial space can sit empty for quite awhile and still make a profit. They then cover the vacant storefronts with colorful graphics.

And eventually businesses do move into those vacant storefronts. Even places that are deserts are starting to develop a little character. South Lake Union isn't exactly where I'd like to spend my time, but it's getting more and more interesting as all those vacancies fill in with little businesses.

Well, that does fit into what I meant -- Outside North America, it seems to be generally (more) taken for granted that public transit and density/mixed-use feed into each other (such that either can be a prerequsite for the other), whereas in North America, because we're in such a car equilibrium, public transit mostly just spends most of its time apologizing for the poor land use.

Although it is probably worth noting that e.g. Los Angeles has some places (kind of) like what you're talking about [1], but it's basically a drop in the bucket.

[1] https://pedestrianobservations.com/2018/11/11/meme-weeding-l...

Any urban change takes decades, especially if you have to build things. For example it takes years from proposal to completion of a single bike lane. Digging subway tunnels is much slower. Here in Berlin we're still debating plans that were originally drafted in the seventies.

I think the most effective tool cities have is zoning regulations. Getting them right enables the city to harvest organic change as opposed to having to plan every single project. Reducing free parking and allowing mixed commercial and residential use seems to me like the easiest way towards a public transit centric city. In Tokyo for example there is basically no curbside parking and shops are always within walking distance.

Totally, but in the North America it seems like we are culturally nowhere near aligned on changing this for the better. The cultural opinion seems to be slowly improving, but it still seems like most of North America has very agrarian/suburban tendencies.

P.S. Japanese zoning is fascinating [1], but not without its faults. For example, the relatively laissez-faire zoning policies can be seen as to blame for the lack of bike lane infrastructure in most of Japan [2]. That said, 東京がとてもすきです :)

[1] http://devonzuegel.com/post/north-american-vs-japanese-zonin...

[2] https://books.google.com/books/about/The_Making_of_Urban_Jap...

The other problem with zoning is that it is not a decision made in isolation.

When governments are under financial pressure (aren't they all?), why not zone for what makes the most short-term or long term cash? The environmental benefits of mass transit or even the idea of high density living doesn't necessarily appeal much financially.

How do gigantic parking lots make most cash in the short term?

Daniel Herriges wrote some great articles about the "activation energy" of walkability, and what significance that has for suburban America:



> There's not even a handful of cities that would be nice to get around or live in without a car. > As an semi-extremist urbanite

Don't call yourself an extremist. There's plenty of people that live without a car in big cities (outside of the US).

In these discussions there is always a lot of griping about the US’s poor rail infrastructure when the US actually has the 2nd most used rail networks in the world. It’s just almost entirely used for freight. An equally valid question is why does Europe transport almost all of its freight by truck?

Obviously geography and population density - network effects scale with the square of population in reasonable service area - are huge factors. A multi-day freight trip is quite common while only a tiny fraction of passenger trips are more than a few hours - the reasonable service areas are vastly different.

In Europe you have multiple overlapping country sized rail networks. While this gives the impression of a continental sized network (especially to Eurail pass riding in American tourists) very few people are taking trains journeys more than a few hours even in Europe.

Another big issue is the incompatibility of freight and passenger rail services. It is very difficult to run both services on the same tracks at high service levels. The US focus on fright has hampered long distance and even regional passenger rail services. While in France the government decided to focus on passengers and eliminated all but a few freight routes. It would take an additional complete parallel rail infrastructure to do both well which would be quite expensive. So far the US and Europe haven’t thought this was worth it. China might be going this way.

Perhaps this simply makes sense given the differing situations.

It's not a huge difference. By tonne-km, the EU uses rail for 17%, and the USA for 31%.

So I accept that the USA uses the rail network almost entirely for freight, but it's not true that the EU transfers "almost all" freight by truck.

Freight trains in Europe tend to run at much higher speeds than in the USA. For example, in Britain container trains can run at up to 75mph, and 140km/h in Germany. But compared to the USA, Europe already has a complete parallel infrastructure: pretty much every route has two tracks. I've only been on one long-distance passenger train in the USA, but there were long periods where there was only a single track.




By your own references the EU28 transports around 75% of its freight by truck while the US is around 40%. Seems like a huge difference to me.

And EU28 stats are distorted by Northern and Eastern Europe where rail freight is big. Italy and France are over 80% truck and France and the UK are over 90%. Even Germany - famous for its freight rail network 150 years ago - transports only about 19% of freight by rail and 72% by truck.

While I understand why US doesn’t have a national rail passenger network - the distances are simply too far between most population concentrations - it is somewhat of a mystery why so little freight travels by rail in Europe and so much by truck. Did the passenger rail network cannibalize the previously existing freight network?

If passenger trains just replaced freight trains and trucks replaced cars, is this really the model to emulate?

>> Perhaps this simply makes sense given the differing situations.

I think that's the only reasonable conclusion. There are so many differences: in economies, wealth distribution, consumer preference (e.g. for local produce), population density, landmass shape (international shipping isn't in the EU statistics, but is at least as significant as coastal shipping in the USA), infrastructure (EU has no problem building new pipelines), ...

I can't find a good source of statistics, but it seems very possible that the USA simply transfers far more goods for more km (overland) than the EU.

> Did the passenger rail network cannibalize the previously existing freight network?

I'm aware of a shortage of rail capacity in particular places, e.g. traversing/avoiding London, but I don't think that's a general problem.

More likely, the same types of journey on both continents moved to road (e.g. farmer to local market/distribution centre, factory to another factory in the same general area), but in the US long distance transport of imports from Asia increased.

That kind of statistic reminds me that Great Britain, for example, has 20% of the population of the United States, but almost 20% less land than the state of Oregon.

I'm not sure if Russia's rail meets your criteria of proper service, but it has both passenger and freight transport in abundance, and I haven't heard of problems with that. Both long-distance and commuter passenger services are used extensively. However, freight service is mostly used for industrial materials while finished goods are mostly delivered by trucks (afaik), so that might be the difference.

Out of curiosity: 2nd most used by what metric? Ton-miles? And who has the most-used?

Lets get real here, its partially a chicken and egg problem, but the population and density has to be there to support such a dense and useful train system. Parts of my city (Chicago) support the kind of rail stations he talks about. Some parts do not as the density here varies greatly but generally speaking...

Tokyo has nearly 10 million people in about 1000 square miles of land

Chicago has nearly 3 million people in about 10000 square miles of land

Tokyo probably has at least 5 metropolitan cities of greater than 1 million people within 600 miles.

Chicago has none, New York is about 730 miles away.

Clearly a world of difference in the sort of density and distances that just don't necessarily support the same kind of rail system.

Even then parts of Chicago closely match his experience in Tokyo, though certainly not to the same quantity and quality.

A picture of a far north chicago L stop I lived near at one point in my life, right near the end of the line.


Nearly every stop along that train route is in the center of a neighborhood at least going north.

Where Chicago screwed up was putting L stops in the middle of freeways rather than in the center of neighborhoods for certain sections and lines. It honestly makes those areas feel like the heart of the neighborhood has been cut out of it, there is no central spot where people have a reason to be. And who really wants to live right next to a 12/8/6 lane free way? No one, and certainly that's where Chicago neighborhoods no longer feel like the cozy place where people hang out, eat, drink, and catch their ride into work.

> Lets get real here ... > Tokyo probably has at least 5 metropolitan cities of greater than 1 million people within 600 miles.

Greater Tokyo is about 38 million people within 50-80 miles (5.200 sq mi)


It's too bad he never mentions the Netherlands. We have few metros but take his 'walk to the station' idea and introduced 'cycle to the station'. You can always store your bike at a Dutch trainstation and coupled with good cycling infrastructure this increases the area served by a station.

We also have cross-country trains every 10 minutes and if you get to a station where you don't have a bike you can rent one from the train company. Or rent a car if you're a bit further away.

There's always room for improvement but right now they have a 90% on time record on an open rail system they share with competitors, foreign train services, and freight trains.

Adding bikes to the equation could solve America's "but our urban communities are too sparse to be served by a metro/train". A bike gets you quite far in 10 minutes. Let alone an e-bike or an electric scooter.

In Utrecht, the main station in our star-grid, they just opened a bicycle parking with 22.000 spots! https://www.youtube.com/watch?v=L3M_GM_MDg8

edit: this is what our national rail looks like: https://www.spoorpro.nl/wp-content/uploads/2017/12/spoorkaar...

The model is indeed great, but the situation in Amsterdam is not that good. Metro lines are crowded daily, not that clean (especially 51), often delayed, sometimes out of reach for commuting even with cycling. The 52 line is a dream but at that pace it will take half a century to be half as good as Tokyo’s.

Intercity (or even longer distance trips within the city) are not available nearly often or connected enough. The Ring roads are chock-full of cars even with massive 5-6 lane express ways all around.

> A bike gets you quite far in 10 minutes.

Not when there's a foot of snow on the ground, it doesn't.

The Dutch and Danish approach to cycling is just from a different world. Even cities that I consider decent for cycling, like Berlin, pale in comparison to any larger Dutch or Danish city. And once you introduce cycling into commuting systems at that scale, yes, it enables you to design those systems in new ways.

It also requires a shift in focus away from prioritising cars to prioritising pedestrians and cyclists. In the Netherlands, you can be sure that every town planner owns a bicycle and commutes on it frequently. I live in Ireland. I've met regional town planners here. None of them own a bicycle.

Traffic planners here call the traffic state when pedestrians are crossing the road "dead time".

One of these days when I've got nothing better to I'm going to put together a little photo essay of the bike racks I know of in my city that don't even work. Using a conventional U-lock you can't even lock your bike to them.

Some are too fat for a U-lock to fit around. Some are placed so close to a wall that you can't fit your bike in. Others have these vertical slats that you can stick your front tire in between, but you can't lock your frame to anything secure.

The people who designed and implemented these things must have never used a bicycle for practical purposes. The stupidity is mind boggling. It's not a difficult thing to get right.

> Others have these vertical slats that you can stick your front tire in between, but you can't lock your frame to anything secure.

This is probably copied from Copenhagen, where it's by far the most common type of rack.

Bicycles simply aren't locked to anything secure, unless you bring your own cable lock. That's something you might do with a flashy, new racing bike, or any other very new bike left outside e.g. into the late evening or night, although insurance should cover theft anyway.

If you have a kickstand, you don't even need to use the rack.

Fancy example: https://pricetags.ca/wp-content/uploads/2015/07/norreport-2....

(The blame is still with the designer; it's obvious this isn't appropriate in many places outside Denmark.)

> Others have these vertical slats that you can stick your front tire in between, but you can't lock your frame to anything secure.

If this is what I'm picturing, you can often just lift the bike over so the rear of the front tire is in the vertical slats, instead of the front of the front tire. Then the frame is close to the rack, and you can lock your frame to the rack.

Stick your rear tyre into those slats, usually this allows locking frame and rear wheel. Wirelock on the frontwheel if wanted. Also, the foldable locks are much more practical than the U-style locks.

I'm pretty loyal to the U-lock, it's the fastest. When I was a messenger, locking and unlocking 50 times a day, we all used them. It takes 5 seconds when you got your system down.

Its really sad, not too long ago (60's) most of us didn't have cars and cycled. Heres a great example of how badly we've fucked ourselves up in the last few decades. https://www.flickr.com/groups/dublincyclelanes/

That's a good point. When I look at some stuff that's done for cyclists here in California I also have my doubts if any of the planners ever have used a bicycle.

I feel like the argument weakens when he compares countries. I'm looking at the railway maps of France and Japan thinking: 'this entire country is the size of a single, largeish US state'.

Total land area is a problem to be solved in the US for train travel, and we only have Russia and China to compare to.

City to city comparison makes much more sense

> this entire country is the size of a single, largeish US state

So why is there no comparable railway network in those largeish US states that are comparable to EU countries? Or in the many (almost all of them) states that are smaller than France and Japan?

Because there would be nowhere to go. Imagine building a rail network in New York. Where would it go? How often do people need to get from NYC to Syracuse? And if you built a train to Syracuse, what would you do once you got there (without a car)? The city is the centerpiece of a region with a million people, less than 1/6 of whom live inside the city?

The northeast US conurbation is one of the densest conglomerations of urban centres in the world. A proper high-speed rail system would take you from NYC to Philadelphia in less than an hour and to Boston or DC in a couple of hours. The ”US is so big and sparsely populated” excuse really doesn’t hold water.

The NEC is the one exception to the rule that the US is too big and sparsely populated to support dense intercity rail. (For comparison, the distance between NYC and Chicago is longer than the distance between Copenhagen and Zürich, which is to say, it's longer than the longest axis of Germany.)

The reason why we have such shitty HSR on the NEC boils down to two reasons: one, the corridor is too congested with commuter rail (particularly in Connecticut), and two, the route was rebuilt to "modern" standards in the 1930s. Rebuilding it again to truly modern 220 mph HSR standards involves needing to find virgin right-of-way in Connecticut, which is neither easy nor cheap.

The US is designed differently than the other heavily developed areas of the world. It's not about being "too big" or "not dense enough", it is more complicated than that.

The Northeast US has a similar statistical density as areas of Europe but looks so much physically different. Take some time on Google Earth and check it out. There is a nonstop blob of suburbs that go almost all the way from DC to Boston. In a proposed train network, most of the population has a long trip to the central station before they can take the train to the other city. They still need a car, and need somewhere to park it. In Europe or Asia cities are more concentrated, and the countrysides are less populated.

Building a center-city to center-city train network in the US is not impossible, but it simply wouldn't be as efficient as it would be in Europe or Asia because of the way the cities are designed.

Sure. Obviously it would require a new approach to planning and zoning. As the article says, rail corridors are obvious places for high-density mixed-use development. That aside, driving to a station and taking a train from there is still better than no trains at all.

The article presents DC's park and ride station as "a perfect example of WHAT NOT TO DO". But it's really all they can do if they must force a train system into an area without the right kind of density.

I very much agree we need a new approach to planning an zoning. We need to focus on that first. It could take longer than a generation to transition away from the suburban model.

Its really insane people defend shortcomings in the US by either saying US is big or saying US is too diverse. Very depressing.

The argument is that building train lines isn't the answer because the transport needs in the U.S. are different than Japan, Indonesia, Germany, Russia, etc. Having a cookie cutter "trains are the answer to everything" answer is shortsighted.

Can you please tell me what are those different needs are?

Have you ever been to Japan, Germany, etc? Cities there are shaped completely differently to cities here. The article uses the Vienna Metro as an example of what not to do, and in a way it is. But it’s that way because Vienna is almost entirely low density sprawl, with homes sitting on a quarter of an acre or so. The whole DC metro area is shaped that way, and that drove the structure of Metro. The DC Metro area has 6 million people, about the same size as the Berlin metro area. But 3.7 million Berliners live in the city itself, while just 700,000 Washingtonians do. Berlin and DC are similar of a similar density, but almost all the DC Metro population lives out in the low density suburbs. A lot of the jobs are out there too. A lot of tech jobs are in Reston and Dulles, 20-25 miles west of DC. Do you know what’s 20 miles west of Berlin? Nothing, it’s farm land. The kins of transportation network a city like DC, where the population and jobs are spread out among low density suburbs, and a city like Berlin, with population and jobs concentrated in the core city, needs are completely different.

Primarily: the need to get around in whichever city/area you are in once you get there. In all but the most dense cities (NYC/SF/Chicago), you have to have a car to get around. If I had a high speed train to get between Kansas City and Houston it still wouldn't do any good because you need a car to get around when you arrive. This is true for nearly every area of the USA. We tend to build out, rather than up, making public transport impractical and expensive. Building out does have the advantage of keeping land/home prices relatively low, though.

Many Americans prefer the freedom of being able to go A to B at any time without needing to wait at a bus/subway stop or hoping the transportation system is still running for the day. The amount of money it would cost to get the entire country walk able or reachable by public transit would be astronomical.

> Primarily: the need to get around in whichever city/area you are in once you get there. [...] If I had a high speed train to get between Kansas City and Houston it still wouldn't do any good because you need a car to get around when you arrive.

That's probably true (modulo taxis, even if you rule out public transport), but it's just as true when you fly from Kansas City to Houston. Which is why airports always come with car rental, and it works out for those who want a car. If a car is really needed at the destination, arrangements could and should be made to have car rentals available at the train station.

> A proper high-speed rail system would take you from NYC to Philadelphia in less than an hour

I mean... the shitty high speed rail we have (Acela) does it in an hour and 8 minutes. Making it 55 minutes probably won't change travel patterns substantially. (Although route upgrades to support that is the medium term plan)

The problem with Acela is that vast stretches of it aren’t actually high-speed because freight is in the way. And even at peak speeds, it’s relatively slow wrt other high speed rail.

Nope. The NEC corridor, on which Acela runs, does not carry freight traffic (at least, not except for some rare trains at night). Acela's conflicts are with all of the commuter rail systems that share the track (MARC, SEPTA, NJT, Metro North, Shore Line East, and MBTA commuter rail), plus the regular intercity Amtrak service.

Also, think about California's Amtrack - going from San Diego via Los Angeles to San Francisco. Currently trip from San Diego to LA takes 2.5 - 3 hours. With proper bullet train SD-LA should take no more than 45-60 minutes, making it much more convenient than now, and more convenient than 2 hour drive via highway.

And SD - SF should be less than 4 hours, making it faster than a flight (if you count all the time wasted at the airports).

But what the hell do you do once you're dumped in the middle of SD or LA without a car?

You rent one :)

Or, better yet, you use (now non-existing) public transportation to go around.

Metro, Uber, etc.

>How often do people need to get from NYC to Syracuse?

For people living to Syracuse, tons of times. Equally many times with any EU train that connects e.g. Paris to a city within 50 or 100 or 200 miles.

>And if you built a train to Syracuse, what would you do once you got there (without a car)? The city is the centerpiece of a region with a million people, less than 1/6 of whom live inside the city?

If it's "a million people" it should be able to afford its own public transport connected to the train station, which answers the question "what to do there without a car".

It's not like there aren't places in Europe, Asia etc that are exactly like this...

Are there places in Europe and Asia like this? Syracuse doesn’t have a transit network because nobody lives in Syracuse, they live in the sprawl around it.

What shocked me about my recent trips abroad (to Munich and Tokyo) was that those cities were completely different. You drive a bit out of Munich and the city just ends and becomea farm land. Even with Tokyo there ia farmland between the city and Narita. In the US, it’a sprawl in every direction, and most of the population lives in that sprawl, and not in the city.

>Are there places in Europe and Asia like this? Syracuse doesn’t have a transit network because nobody lives in Syracuse, they live in the sprawl around it.

Well, not quite: "At the 2010 census, the city population was 145,252, and its metropolitan area had a population of 662,577". There are places like that in Europe.

In Greece Athens has a population of 600,000 people and the surrounding metropolitan area is 4 million -- of continuous sprawl in every direction.

> How often do people need to get from NYC to Syracuse?

Far more often than you'd think. If such a system existed people would use it all the time.

Look at the metra lines that go from Chicago to the suburbs in the north/northwest. People take those all the time because living in those areas is significantly cheaper.

People would be able to visit families long distance without needing to own a car or cough up for a plane ticket. That land in the US is vast is not really a good excuse.

Syracuse isn't a suburb of NYC. The distance between those cities is like the distance between Chicago and Iowa City. Nobody commutes to Chicago from Iowa City.

Commuting for work, probably not. But along that route I’d imagine people would utilize it, similar to how metra is used for people coming from Wisconsin.

Being able to conveniently take the train from Chicago -> Iowa would be great, especially if the route went further.

Being able to take a train from anywhere to anywhere is else is great, ceteris paribus. That doesn't make it practical.

Forget about just inside NY. What about between NY, MA and PA. Are you going to say not many people will travel to/from DC-NY, Boston-NY, Philly-NY? Arguments about distance and space or no passenger will use it are really excuses that was pushed down our throats.

Aren't those places that are already well-served by rail? Aren't these just the exceptions that prove the rule?

No, they aren’t served well by rail. At least not at the scale of major European cities.

Yes, Amtrak runs from DC, Baltimore, Philly, NYC, and Boston. But, it’s not high speed (except a few sections on the Acela train). And the subways and neighborhood lines are lacking.

Take DC for example... the Metro single tracks and stops at every station. No express route across town. And trains only run every 20 minutes off-peak (and that’s assuming they haven’t shut the whole thing down because it’s caught on fire again). So, getting from Tyson’s Corner or Bethesda to Union Station is not fast.

An regular train from the inner business centers to Union Ststion, plus a proper bullet train to NYC would make a suitable substitute for flying. Right now, rail just isn’t a workable option for many business travelers.

Acela carries (per Wikipedia's cited source) around 3.5M passengers per year.

From Washington Dulles airport, approximately 265k passengers per year fly to Boston. Another 715k fly DCA-BOS. DCA-LGA accounts for 318k. DCA-JFK, IAD-LGA and IAD-JFK do not crack top-ten lists of busiest domestic routes for their origins.

To get a feel for just Acela (which is not the only Northeast Corridor train service), it's slightly less than the annual passenger volume by air SFO-LAX, SFO-JFK and SFO-BOS combined.

The entire Northeast Corridor carries approximately 12M passengers per year. If it were operated by air, it would be the second-busiest route in the world by passenger volume, slotting between CJU-GMP in Korea (13M/year) and MEL-SYD in Australia (9M/year).

In 2012, the New York Times reported that 75 percent of travelers between New York and Washington used one of Amtrak's Northeast Corridor trains (Acela isn't the only one), and between New York and Boston 54 percent used a train:


There certainly are things that could be improved. But Amtrak's operations in the northeast are not some hypothetical potential maybe-someday competitor to air travel. They are and for years have been a real and significant competitor to air travel.

And for what it's worth, Amtrak also has a farebox recovery ratio for its entire nationwide network that hovers around 95%, second in the US only to BART Oakland Airport Connector (96%).

Thanks for the numbers. I hadn't realized Acela was so highly utilized in the DC-NYC route.

There used to be. Americans, for the most part, prefer drive their own cars for short to moderate trips, or fly.

Prefer is an interesting choice of words in this case. Its like me prefering comcast as an ISP.

Sort of off-topic, but every time someone says this, I am reminded of the subprime car loan market where the same car is sold at inflated prices and usurious interest rates to people who have bad credit, then repossessed and sold again. If we had decent public transport, that wouldn't happen.

This argument doesn't hold water. Why is there no decent regional train system in the US then? Your point explains why there is no coast to coast NYC-SF bullet train, fair enough. Each US megaregion [0] could have a decent train system like France, Germany or Japan.

[0] https://en.wikipedia.org/wiki/Megaregions_of_the_United_Stat...

The short answer is that not enough people want that. The argument seems to be that they would if only they knew how awesome it was. But that argument isn't sticking, so the political will to make it happen is not there.

It seems a bit more complicated than that to me. In many cases Americans in at least some regions do want it, and even vote to fund it, but nonetheless the political & civil service machinery seems unable to actually deliver it. Especially, unable to deliver it for anything approximating a reasonable cost. See, for example, the way over budget and delayed NYC 2nd Ave Subway, or the California HSR. Some countries are able to build a single 400-mile HSR line in less than 30 years.

I take the Acela from Boston to NYC often. It's okay but not great. Nothing like the TGV which I've also ridden.

If the trip could be two hours instead of 3.5+ that would be great but as it is it is barely justifiable over flight (being dropped off midtown is a big bonus tho).

The benefits for even longer travel just don't exist. Once you start getting into the 4+ hours on a train it's almost always better to find a flight.

Japan's shinkansen travels at 2-2.5x the speed of the Acela. At the maximum speed of current bullet train technology, Boston - NY could likely be done in about 1.5 hours, Boston to Washington in about 2-3, and even Boston to Cleveland in around 3-4. That basically covers the entirety of travel in the Northeast corridor.

This doesn't even cover the new maglev trains that are expected to go into service sometime in the late 2020s in Japan, which have a maximum speed of 550km/h, expanding the range of that 4 hour "maximum" significantly.

The major problem with the Acela is that it mainly runs on older infrastructure that wasn't designed with HS trains in mind. This severely limits its speed on most of its route. Most HS systems in other places around the world are on newly developed HS lines that were built up specifically to support the much faster speeds and requirements of these trains, and are completely separate from the regular rail lines (which might be running in parallel nearby).

It is also that many people actively oppose trains because they are afraid that the trains will bring in undesirable people (poor people, criminals, hipsters from New York) to their neighborhood. When New York MTA tried to extend Metro North service 25 miles beyond Poughkeepsie on already existing tracks, residents of the surrounding towns opposed the extension claiming that NYC commuters will move to their rural towns.

>I feel like the argument weakens when he compares countries. I'm looking at the railway maps of France and Japan thinking: 'this entire country is the size of a single, largeish US state'.

Not that old tired chessnut again.

First, all of Europe is connected by trains. You can go from any country to any country in EU by train -- if you wish so. That's equally large to the US (and the train lines even continue down to Moscow and further down to Beijing).

Second, it's not like you need to hook up all of the US or nothing. Where is a comparable and equally nice metro/train solution within a single small state (dozens of which are smaller than most European countries) or even within a city?

Even Chicago or Manhattan are worse than anything shown train/subway wise.

> You can go from any country to any country in EU by train

Surely you mean "continental Europe" unless there's a tunnel to Ireland

> That's equally large to the US

Uh, no. The southwestern tip of Portugal to the northeastern tip of Estonia is around 3300 km. The mainland US equivalent is 4400 km. Moreover, unlike Europe, the population centers are mostly along opposite sides.

If you want to do it, the fast train from Birmingham ends at Holyhead station in Wales, and (depending which end of the train you're in) you walk about 100m to the check-in building of the ferry.

I don't know for this place, but it's fairly likely that the ferry schedule aligns with the train schedule.

Train+ferry connections used to be very common in Europe. Sometimes the train was loaded onto the ferry (still happens between Germany and Denmark). Some have been replaced by bridges, others are defunct, but where they still exist it's usually possible to buy an ordinary train ticket which includes travel across the ferry. Interrail/Eurail passes include the use of these ferries.

>Surely you mean "continental Europe" unless there's a tunnel to Ireland

Surely that's a pedantic objection, when you can get to 95% of EU (or more population wise) by train, and when you can just take the ferry off of Scotland (and France and UK are of course connected by train)...

>Uh, no. The southwestern tip of Portugal to the northeastern tip of Estonia is around 3300 km. The mainland US equivalent is 4400 km. Moreover, unlike Europe, the population centers are mostly along opposite sides.

That's a difference of 25%, and not the machine scale difference implied at first with the US vs France comparison.

Second, that's off too. Trains continue way past Estonia, (Sweden and Finland go even further north, part of Russia is in Europe etc). Furthermore train connections continue all the way to China (e.g. with the trans-siberian-express) and even Korea and across the sea from Japan.

>Uh, no. The southwestern tip of Portugal to the northeastern tip of Estonia is around 3300 km. The mainland US equivalent is 4400 km. Moreover, unlike Europe, the population centers are mostly along opposite sides.

Besides the fact that the rest of the US is still 150 million strong (and thus could very well have some good lines connecting it), this just makes it even easier to have a great train system in those populous cities in the opposite sides -- one in East and one in the West coast for example.

Heck, the US doesn't have a good train and subway system anywhere (e.g. not even within a single state), so the logistics to have one throughout the continent is a red herring.

That the whole of USA is a large land mass is no excuse for individual cities having bad public transport.

To take the analogy over the top, Greenland is a part of Denmark, I don't hear them saying that because of that, they should start investing only in freeways for cars in Copenhagen?


Only Texas and Alaska are larger than France. I give you Alaska, due to the geography and low population, but Texas is just slightly larger and has half the population of France - not a massive difference.

I think it could be reasonably fair to compare countries to single states at least?

You can, of course, compare countries to states but sadly too many of these articles make sweeping generalisations and don't mention the compromises. Most people would much rather have the convenience of a car if possible and cheap but in Japan it is often neither due to space constraints and generally parking is illegal on most roads.

Compare Texas, lots of open space and historical reasons necessitates car travel except for major urban to urban transport which could be done by train but why bother? As in the UK, the acceptance that cars be supported for some journeys means that on balance, they are better to use for most others. Even for me, the coach runs direct from my town to Heathrow airport in a similar time to a car and pretty cheap but it won't always run when I need it so why not drive (or get a lift)?

This is an interesting comment.

As a Londoner (have lived both centrally and in the suburbs) I think you're right that it's one or the other.

If you design for cars (e.g. M25, UK motorway network and roads, towns like Swindon and MK) then cars are the uncontested king.

In an area like London, it almost becomes an emergent phenomenon. The suburbs have car use because they're not very dense and it still beats sporadic bus routes.

In the centre there's just no space. Forget parking, Uber might be slower than the tube in a no-congestion scenario simply due to traffic lights and giving way.

Germany and California are in the same ballpark when it comes to size and economic power. The state's internal transportation network should be just as developed but it is not. It's not only the size of the country, it's also the philosophy of the people and society.

When they talk about the train in Russia, they really only talk about the train in Moscow. Everyone I know says stations are very nice, and schedule during peak hours is like a train/min. BUT, at the same time, it's much much more packed than here in new york. Even during rush hour I don't feel pressed against other passengers here.

You can't claim that a high-speed line between Boston, Providence, CT cities, NYC, Philly, and down to Washington is not sensible. For many other states like Illinois, where there's one big city without much else near it, yes it's sensible to not have a great network of rails. However, there is really no geographic excuse for not having high-speed rail in the basically linear east coast corridor.

Having traveled many times from Boston to NYC via car, which is about 5x cheaper than Amtrak and yes actually comparable in cost to the bus, depending on your MPG, it's kind of ridiculous to me that we don't have a better version of Amtrak by now.

Both Russia and China have way better train systems than the US, both in the cities and outside them.

TO be fair, despite the US having some large emptier states, the US has nodes of very high urban density - most notably the Northeast corridor. From Boston to Washington there's around 50 million people, with - depending how you define it - a density of around 390/sqkm, three times that of France's 120/sqkm.

Of course not representative, but you can definitely see the influence of the automobile in how these cities have grown. Compare to say, Moscow - another major urban city in a fairly sparse city - New York City has only added a few km of subway lines in the past 20 years compared to 150+km.

Spot on. Public transportation is obviously built where people are.

> These days, it’s easy to make subway systems because we have amazing tunneling machines, which will bore a train-sized hole like a giant rock-eating worm. So, no excuses.

I wouldn't say "easy". It's possible. Here in Seattle, we're building a new 2-mile tunnel (for cars). It was originally scheduled to open 3 years ago, and will finally open in 2019. It was planned to take 29 months, but took 67.

We're not so great at bridges, either. One of my managers observed "We've got 4 floating bridges, and we sank 2 of them." That was back when I was on the Boeing 7E7 project (first flight: 28 months late).

Is it any wonder we can't get excited about the prospect of an "easy" new rail project? We can't estimate big engineering projects to within 50%, and we can't keep existing infrastructure afloat (literally). I always vote for public transit on principle, but I'm also aware it's always going to take longer and cost more than they claim, so I can't really blame people who don't.

Want broad support for infrastructure projects? First, demonstrate you can execute on them reliably. Nobody wants to agree to pay for a tunnel that might end up being the next "Big Dig".

> I wouldn't say "easy". It's possible. Here in Seattle, we're building a new 2-mile tunnel (for cars). It was originally scheduled to open 3 years ago, and will finally open in 2019. It was planned to take 29 months, but took 67.

To add to this, some areas are made of materials that are simply difficult to bore through. Atlanta, for example, sits on top of a bunch of granite which at least partially contributes to the difficulties of underground rail expansion.

Adding one lesser known example of imho outstanding public transportation service is Vienna, Austria - not the largest of cities, at 1.9 Mio inhabitants, and about half the area of Berlin (or 1/4 of London), still the 6th largest city in Europe.

You can get almost anywhere within half an hour tops, the Metro runs till after midnight weekdays and 24/7 on the weekends, there's an extensive system of Tramways & Busses to cover anything a bit farther from a Metro station.

And you can get a yearly subscription, giving you all-inclusive usage of the services for 1€/day, billed monthly.

This is mostly to stress the point of how much you take such a system for granted when it's available and just works - since I've moved here owning a car would be almost an inconvenience, and even London feels like a horrible public transportation experience.

> And you can get a yearly subscription, giving you all-inclusive usage of the services for 1€/day, billed monthly.

This kind of thing is key - make it easy to use transit and lower the marginal cost of doing so. Yearly subscriptions are good. London's approach—unify almost all modes under the same payment system with automatically calculated caps and discounts—also works quite well as you don't have to think about buying passes, individual tickets, etc. Just tap and go and the system takes care of it for you.

Not many places in the US have this level of integration yet. For instance, the NYC Ferry, NYC Subway, LIRR, Metro North, PATH, and regional Amtrak service all use disjoint ticketing methods. It'll slowly get integrated over the next decade, but it greatly increases the mental effort needed to hop on transit. Perhaps the closest in the US is the Bay Area's Clipper Card - they managed to shoehorn over a dozen area agencies onto the same payment method - but it's still not quite as good as something like London's Oyster.

The TAP card in Los Angeles is good for dozens of local public transport companies. For quite a while, this was not the case; you could only use it on Metro (County-level agency). The integration over the years has certainly increased my use of transit in the area; there are many tiny, obscure transit systems in Los Angeles County and I can ride almost all of them now without having to carry any cash or change. Sometimes you can even get cross-company transfer fares with the TAP. It's fantastic.


> 1€/day, billed monthly

Nitpick: AFAIR it's only 1 EUR per day if you pay the whole 365 EUR for the year at once. If you pay monthly, it's 396 EUR for the whole year. This is a pet peeve of mine as it's a good example of how expensive it can be to be poor.

However if you're getting used to the subway system running every 3-5 minutes you get agitated if you have to wait 15 minutes (or even need to plan when to arrive at the station)

The author, Nathan Lewis, makes frequent references to concepts he established in detail in previous articles. For example, "Traditional City", "Place", "19th Century Hypertrophic City".

All of his articles on cities:


It's a shame this article doesn't lay out any plan for how to get from where the US is now to where the author would like it to be. You're not going to dig up the existing train stations, so how could the parking be converted to shops and housing? How can you persuade drivers and politicians to support whatever other steps are necessary? (eg. converting roads to railways)

I wonder what changed in our mindset in the last 50 years that such projects seem impossible now but possible back then. I mean subways were not any easier to build 100 or 50 years ago than today. And it surely was not cheaper.

My favorite theory is that the post WW2-era American value shifted to personal liberty and the American dream: your home, your picket fence, and your car. You choose where you go, when you go, and how you get there. Drive on those new interstates! Take the family for a picnic. Go surfing!

How can you be truly independent and live that perfect magazine cover life if you’ve got to wait on the 9:27 to take you to the B line so you won’t be late to work?

I think the_mitsuhiko is wondering not about rail projects specifically, but about a loss of confidence about performing large-scale projects in the abstract sense, independent of what the project is for.

For example, in 1930 it seemed feasible to build the Hoover dam; in 1950 it seemed feasible to build the interstate highway system; in 1960 it seemed feasible to land a man on the moon.

Now, in 2018, people think it isn't feasible to build some train stations. And those people might be right! What happened to our ability (practically, politically, and financially) to get such things done?

> Now, in 2018, people think it isn't feasible to build some train stations. And those people might be right! What happened to our ability (practically, politically, and financially) to get such things done?

Over time and increasing population there is a evolutionary niche filling kind of process that occurs within social, political, economic, cultural and other axes/planes/spheres. After the density of these filled niches increase sufficiently over some threshold, our civilization experiences a calcification of movement across the niches.

Each niche acts as a middleman to extract some portion of energy to successfully navigate across them. This takes the form of some kind of concession, trade, grant, swap, boon, promise, etc. Fill a path with enough niches that didn't use to exist before, and we experience it as "we can't get it done" because the cumulative energetic cost rises above a tolerable amount.

Historically, the emergent property in most capitalist structures appears to naturally tend towards influencing government agents via the principal agent challenge to support/enforce increasingly rentier behavior over time and deviate from capitalist structures; I personally believe this is due to scaling issues (Adam Smith was not a fan of the kind of large-scale capitalism we take for granted today). This wasn't supposed to happen according to some economic schools of thought, but here we are. And niche filling, and use of our government to enforce stakeholder extraction claims, appear as another form that this behavior takes.

For starters, the political system started caring about the concerns of average people, and stopped letting grand visionaries destroy people’s neighborhoods for the greater good. Having to appease each and every household you affect will put a damper on things.

Half of it is what people wanted, and the other half is what the people were given. The Federal Government directly created today's heavily suburbanised America through its policies: https://www.historysouth.org/wp-content/uploads/2015/12/The-...

> How can you be truly independent and live that perfect magazine cover life if you’ve got to wait on the 9:27 to take you to the B line so you won’t be late to work?

By that stance, no one living in Chicago, NY, or any city with public transit are "truly independent", or whatever. Waiting a few mins for a train or bus is no different than being stuck in traffic.

When I lived in Chicago my quality of life went up precisely because I didn't need to use my car. I never had to worry about finding parking. All I needed to do was learn some of the bus timetables and where the CTA lines generally were.

Agreed! I lived in Taipei for a couple of years, and currently am spending some time in NY. I was being a bit flippant. :-)

It was quite a lot cheaper, AFAIK. To use a recent example local to me, in Portland, Oregon. We built a replacement bridge for an old one that was at the end of it's service life. The new bridge is largely similar to the old one. A little bit wider so it has wider sidewalks and bike lanes, but still it is a two-lane bridge. Adjusted for inflation, the budget for this bridge was 10x what the first one cost. And that doesn't include any overruns that may have occurred, that was just the sticker cost.

For various reasons that could probably be debated for a long time, big project construction now is much, much, much more expensive than it was a century ago.

I assume you're talking about the Sellwood bridge. If so, you're being disingenuous. The new bridge is more than twice as wide as the old one (64' vs. 28'). It's also an entirely different design (deck arch vs. truss).

The work included moving the existing bridge, demolishing the existing bridge, earthquake resilience, and building a big new traffic interchange on the west side.

The inflation-adjusted cost of the original bridge was $7.5 million. The final cost of the new bridge project was about $324 million. (2016 dollars) The new bridge project was 43x more expensive than the original bridge, but so much is different, I don't think you can draw any conclusions about changes in construction costs.

That is sort of my point, though. We raise the standards for everything we do, from safety during construction, to safety during use (bicycle lanes, wider pedestrian sidewalks, earthquake resistence, etc). This all bumps the price up, and significantly so. Whether that is good or bad is a separate issue, but it's definitely a major factor in why we don't do more big infrastructure projects now.

Heck, I wonder if it would have been cheaper to build a more comparable direct replacement bridge for vehicle traffic, and then a completely separate walking/riding bridge. The new Sellwood bridge cost nearly three times what Tilikum Crossing did, and that has light rail and buses in addition to bikes & pedestrians.

Personally, I'm all for improved safety standards. And I very much appreciate the wide pedestrian and bike lanes on the new Sellwood bridge. I used to bike across the Sellwood regularly and it was a nerve-wracking experience.

It's interesting that the Sellwood bridge was so much more expensive than Tilikum (2.4x as much, not 3x, according to my quick search). I wonder why that was? I'm not a civil engineer, but if I were to guess, I would guess that it's a combination of the new Sellwood interchange and the need to deal with the existing bridge.

The one that always gets me for sheer ambition is the raising of Chicago.

"By 1860, confidence was sufficiently high that a consortium of no fewer than six engineers—including Brown, Hollingsworth and George Pullman—took on one of the most impressive locations in the city and hoisted it up complete and in one go. They lifted half a city block on Lake Street, between Clark Street and LaSalle Street; a solid masonry row of shops, offices, printeries, etc., 320 feet (98 m) long, comprising brick and stone buildings, some four stories high, some five, having a footprint taking up almost one acre (4,000 m2) of space, and an estimated all in weight including hanging sidewalks of thirty five thousand tons. Businesses operating out of these premises were not closed down for the lifting; as the buildings were being raised, people came, went, shopped and worked in them as if nothing out of the ordinary were happening. In five days the entire assembly was elevated 4 feet 8 inches (1.42 m) in the air by a team consisting of six hundred men using six thousand jackscrews, ready for new foundation walls to be built underneath. The spectacle drew crowds of thousands, who were on the final day permitted to walk at the old ground level, among the jacks."


Regulation. Environmental impact studies. Workplace safety rules. When we first built the interstate highway system, we just started bulldozing a path across the land. Now, there's five years of planning and permitting necessary for even a small bypass project.

Edit: I'll add, also, survivorship bias. We see what was achieved. We don't see what was planned and never came to fruition for lack of political will or whatever reasons.

That's one thing that concerns me about the US in general. Other than military it seems the country has given up on acting on any issues society faces. Be it gun violence, health care, immigration, retirement or infrastructure, issues are just used as material for never ending political fights but nothing gets solved. Pretty sad.

Politicians need those issues to divide and rally their voters. If they actually solved the problems, they would lose those wedges.

To paraphrase Upton Sinclair, "It's impossible for man to solve a problem, when his salary depends upon it."

I don't think it has to be that way. When I compare US politics to Germany Germany politics is much less vicious and they talk more about issues instead of the focus on character assassination and demonization of the "other" side as it's common in the US.

Really? I don't get that impression at all when looking at the AfD, in fact it seems to be playing out more or less identically to US politics:


Yes, there is the AFD but the general tone is much more civilized.

This is pretty much it. The political class needs conflict to have purpose and reason to exist.

Too-low corporate tax rates can't be a reason for this. Until the 2017 tax changes, the US corporate tax rates were (depending on who you asked) either on par with or significantly greater than those of comparable countries. So we can't attribute the US failure in developing public transit to low corporate tax rates.

Here's a nice summary of NY issues with this:


I'm going to give the typical leftie answer: neoliberalism.

Basically, large public works projects still seemed possible up to the early 1970s. The hypothesis is that it's only after two near-simultaneous changes, the rise of Reaganism-Thatcherism and the fall of the Soviet Union, that the state seemed to lose all of its potency. It seems like we placed all of our faith in the wisdom of the market, a machine that can never be stopped once given power.

Citizens United was the nail in the coffin. What potency the state has left is now designed to funnel tax dollars into corporate interests at minimal expense. That precludes any public works projects that don’t directly benefit corporate interests.

I mean subways were not any easier to build 100 or 50 years ago than today. And it surely was not cheaper.

They were much cheaper: http://slatestarcodex.com/2017/02/09/considerations-on-cost-...

I don't think there's much point in thinking about that change, because in the US there isn't the political will to operate and maintain such systems on an ongoing basis.

The NYC subway + DC metro, while far from perfect systems, are basically built on the desirable "traditional city" model. In both cases, actually maintaining and funding the system to operate to a reliable and acceptable standard, instead of running it into the ground, is a controversial idea with limited political support.

Unless and until such things can be operated as a going concern on an ongoing basis, there is no point building any more.

Like you alluded to, it’s mostly about finding, gathering, and organizing the political will to fund such projects. There are, I imagine, plenty of projects archived in some filing cabinets in the basement of some federal building somewhere that had its day in the legislature and didn’t make the cut. My favorite example of this being the 2012 plan for the Northeast Corridor:


Super Express service between New York and Boston in 90 minutes, cutting through central Connecticut with no stops, carried by new EMU-driven rolling stock.

tl;dr: Not happening. Instead, Amtrak’s gonna straighten out a few kinks in the current track so the Acela service can get up to and maintain higher speeds south of New York and in eastern Connecticut. Not sure how they’re gonna add more capacity, if at all.

> Like you alluded to, it’s mostly about finding, gathering, and organizing the political will to fund such projects. There are, I imagine, plenty of projects archived in some filing cabinets in the basement of some federal building somewhere that had its day in the legislature and didn’t make the cut. My favorite example of this being the 2012 plan for the Northeast Corridor:

To be fair, Hurricane Sandy and the associated disruption of the Gateway Project would have probably done that ambitious plan in on its own.

> Instead, Amtrak’s gonna straighten out a few kinks in the current track so the Acela service can get up to and maintain higher speeds south of New York and in eastern Connecticut. Not sure how they’re gonna add more capacity, if at all.

"Instead"? If I am reading that long-term plan correctly, this was the actual first phase of the project. The increased capacity comes from making the Acela trains longer and more frequent (per the proposal).

Climate, pollution and oil price are the main drivers here. It will settle back with time.

The Westfriedhof station pictured is also in Munich, not Stockholm, guess it's a copy/paste error. It is kinda nice, but it's relatively young (15 years maybe?) whereas most of our subway stations are from the 70s and not half as nice.

Edit: it's 20 years old, not 15: https://en.wikipedia.org/wiki/Westfriedhof_(Munich_U-Bahn) and the others pictured are also on Wikipedia: https://en.wikipedia.org/wiki/Georg-Brauchle-Ring_(Munich_U-... https://en.wikipedia.org/wiki/Candidplatz_(Munich_U-Bahn)

Old does not mean it has to look shabby as the moscow stations demonstrate. Or these from berlin:


No of course not, but some stations here have this "this wall was modern in the 70s, also it's kinda grimy and some of the tiles fell off" look - like https://en.wikipedia.org/wiki/Sendlinger_Tor_(Munich_U-Bahn), but some were worse. They are renovating some - and to be honest if it's structurally sound I couldn't care less if it looks shabby, I was just trying to point out the differences - because some of the newer ones look really nice :)

> This is the train map for Washington DC. As you can see, it is an irrelevant little fart of a train system.

I used to live in Japan, now I live in Southern California near the Gold Line, which is the pride of the Los Angeles County Metropolitan Transportation Authority. The Gold line is approximately as long as the dinky little private Tawaramoto Line that cuts a triangle between JR Oji Stn and Kintetsu Tawaramoto Stn in Nara Prefecture. Except the Tawaramoto Line always ran at least four cars while MTA's Gold Line only runs that many during rush hour. LA's crown jewel Union Station is approximately the size of a mid-size suburban station in Japan. It's incredibly sad.

"These days, it’s easy to make subway systems because we have amazing tunneling machines..."

-spits out coffee comically-

I suppose compared to 1905 it's easier, but it's still not "easy." Tunneling never goes off without some kind of hitch, and almost always goes over budget or off-schedule.




I agree with the general thrust of the piece though.

Nah, the super expensive and always goes way over budget are features of subways only when Americans make them, unfortunately. That's not to say no one else has construction projects that go over budget, but the rest of the world -- more developed countries and less developed countries alike -- all manage to build infra for way less money, and much closer to budget.

If we want nice things -- like trains that work! -- we have to figure this out.

see, for example, https://www.citylab.com/transportation/2018/01/why-its-so-ex...



You're generally right and I'm not trying to be a pain in the ass, but the one example I thought of belatedly AFTER hitting "Add Comment" on the above, just happened to be this... in a country that mostly seems to have its shit much more together than the US, too: https://en.wikipedia.org/wiki/Historical_Archive_of_the_City...

As a DC resident, I partially agree with his conclusion on the DC metro. But, the stations he chose are suburban, not downtown. Tokyo is a much larger city than DC. so, it’s a bit apples to oranges.

It was designed to get people into the city from the suburbs. As such, it completely fails today, where most jobs are also in the suburbs. It fails for reverse commuting as well, because, as noted, most stations are not within walking of commercial hubs.

That said, the expansion to Dulles came with lots of rezoning and denser development. Arlington started this before the expansion with the Courthouse-Clarendon-Ballston areas, with increased density and walkable zones. Reston and Herndon are doing similar things.

Now, if we can figure out how to build a line from Tyson’s to Rockville or Bethesda (without going all the way downtown and back out again).

> But, the stations he chose are suburban, not downtown.

Why? They could have been constructed downtown (assuming you mean in Vienna or Fairfax, rather than in the middle of the motorway not-especially-near either of them). Or they could have been surrounded by the shopping malls that instead are nearish (but too far to walk).

The town (Vienna) is almost 2km to the north, and doesn't have a station — yet it used to! There's a path/trail labelled "Washington and Old Dominion Trail", which is obviously an old railway line, ripped up and replaced with a bike path.

Since the article mentions Stockholm, I've chosen the metro station furthest from the centre of Stockholm [1], and the station before it.

It's a suburban area, but the stations are still in the middle of the 'suburban centres'. They're within 150m of multiple schools/colleges, a shopping mall, a culture centre and a theatre.

Stockholm also has commuter railway lines ("Pendeltåg") extending further from the city, not shown by Google's map. These stop at some fairly small places, but it will still be somewhere in the middle of each place.

[0] https://www.google.com/maps/@38.8758039,-77.2713152,5435m/da...

[1] https://www.google.com/maps/@59.2443707,17.8132863,1092m/dat...

The Orange line in Fairfax County was constructed in the median of I66 to save money. That was a huge mistake. The difference between the Orange line in Arlington where the locals lobbied to have it routed through actual neighborhoods and the Orange line in Fairfax County where it sits in the middle of an eight-lane highway is staggering.

The loss of the W&OD railway was a major failure of foresight IMHO. It collapsed for the same reason most local rail systems collapsed around that time: automobiles ate their lunch.

Now the local government is spending billions to build the silver line expansion more or less parallel to the W&OD trail, except that it doesn't go as far. The stations locations for the Silver Line are not very good. They bypass the downtown areas of Reston, Herndon, and Ashburn (such as it is) so the rail lines can be built in the middle of the toll road. Only a couple of the Tysons Corner stops really make much sense. Even the Dulles Airport stop is a quarter mile from the terminal.

But really they didn't have the option of running it through town without buying out loads of expensive property (and no doubt a whole lot of politically unpopular eminent domain use) and spending decades in court fighting over it.

The US is indeed addicted to cars, and particularly for those living in the suburbs this is a sunk investment. It really doesn't help that the cities are (usually) NOT built to building codes that actually encourage privacy between units, that needs to get fixed too.

However, an upgrade path would be a great idea. That path must allow for the transit systems (either highly regulated utilities or community owned) to spread.

    * New core has the target service level in place.
    * That service links to a PROPER parking structure at en edge.
    * Such structure should be SECURE (unfortunately monitored)
    * Also package transport services in addition to people.
Renting 'a spot' in that area, as part of a residence OR as part of visiting should be low cost, and shouldn't /punish/ owning a car. The idea is you're incentivizing use of the new zone as well as commerce within it. The parking structures become a gateway as well as additional storage (and should convert to that when use for parking decreases).

One thing these articles nearly always gloss over is that these beautiful, clean metro systems stay that way partially because they close down for maintenance at night. The New York system runs 24/7, as does most of the "sorry crap" in Chicago. I presume there's an article here scolding Americans for staying up too late or something.

In Moscow, for example, the last trains from the end stations depart at 0100 and arrive at 0200-0230 or so at the other end, and the first ones depart at 0500 or so depending on the line.

I wouldn't call those 3-4 hours closing down for the night. Even so, this break is used mostly for track maintenance where one can't have trains rolling.

Station cleaning and rolling stock maintenance is ongoing, construction/repairs that do not require cutting electric power to the trains are being done 24/7.

So, no excuse for NY.

You don't need to shut down the system to clean the stations, that should be happening all the time in a properly managed system. And when it comes to the trains themselves, with extra cars, you can easily rotate stock in and out in order to clean them, or have workers get on the trains at non-peak times (including at night) and clean as the train is in service. You don't need to deep clean each train every day to keep it clean, and with only 2 lines running 24/7, it's not like it'd be that hard to have extra rolling stock just for those two lines while cleaning the trains on non-24/7 lines at night.

Simply put, 24/7 is not as big of an obstacle to cleanliness as you make it out to be.

As other people mentioned, too, most of the work done in the evenings is track maintenance.

If you visit London and wander around a station (metro or regional rail), it won't take you long to see a cleaner.

I used to live at the end of a line, and no matter the time of day there was very often a cleaner waiting by the frontmost door as the train pulled in. They could usually remove any litter before the train was due to leave, perhaps 5-8 minutes later.

The service is less frequent overnight (Friday and Saturday only), which leaves plenty of time to clean more thoroughly at the depot.

But I think that's a distraction. In comparison, NYC's subway looks dirty

a) by lack of maintenance. Some areas just need fresh paint.

b) by design. Bare concrete, unfinished, gloomy ceilings, yellow-ish lights, grey trains.

In other situations (shopping mall, fast food) Americans have remarked to me that Europe can seem clinical, due to the high temperature of the lighting. I think those places, and the NYC subway, look dingy. The same light could look cosy, but only in a café or bar.

There's even Freedom of Information requests on the number of cleaners and frequency of cleaning London Underground trains.

Combined, the information shows they do a "deep clean" on every train every 21 days, and it takes four hours per carriage.

Then 15 minutes per carriage just under twice per day per carriage.

I suspect the cleaning I saw isn't part of this, but is the station cleaning staff feeling it's easier to pick the litter from the train rather that wait for wind to blow it out and around the station.



I was thinking something similar, but honestly, getting wtf. Few months back I was traveling from Bronx to Brooklyn, pretty much opposite sides of the area. Normally it should take 2 hours and change. This time it took me 4... and a half. I left earlier, around 10. Didn't help.

What's the point of 24/7 schedule, when it takes an hour and a half during the night for your train to arrive?

Idk, I think I might've preferred them to shut the train down say 12pm-6am to give them time to improve whatever they need to. It would be much more useful to know the exact time it would take me to get places. Trains at night are pretty empty, I think with some adjustment, it wouldn't be a big deal for people, but maybe there is more to it.

Well, the NYC subway is still grossly mismanaged, about 20-30% of it is down even on the best days, with tens if not hundreds route changes and station closures. On weekends and at night up to half of the system may not work. Because the system is poorly maintained, the trains run slower than what they should be, causing even more delays.

EDIT: edited percentage to 20-30%

I've read lots of articles about mismanagement and unions pushing up costs for work but I wonder how much problem the 24/7 mentality creates? Even in London, the Tube is mostly shutdown for around 6 hours per night, not a lot of time but enough to make some fairly big inroads into the maintenance.

I found the NYC subway to be fairly decent but I found the feeling of darkness in many stations to be more of a big deal. I wouldn't mind waiting in stations that were much lit much better.

> about 70-80% of it is down even on the best days,

I’m not usually one to defend the MTA but this is wildly inaccurate.

thanks, I meant 70-80% is working

If shutting down the trains for a few hours at night is what it takes, then maybe they should do that? Run night buses instead.

"Instead of having a reliable system 20 hours a day let's have an unreliable system 24 hours a day" doesn't seem to make sense.

CTA trains are not 24/7, fwiw

Red and Blue Line are, which are about half of ridership.

Ah, interesting. Thanks

Counter point is rats.

In the 50s, the US made a big bet that car based transportation was the future. It was, for awhile, but now it's clear that denser cities are the future, not suburbs. Countries that invested in infrastructure that supported dense cities are providing a higher quality of life compared to the "drive everywhere" culture of the States.

"most of the rest of the developed world has quite wonderful subways and trains. They are clean, efficient, cheap, run on time, have as many as 20 or even 25 trains an hour...."

We used to have a lot of railroads in the US. Then the automobile came along, which started them going downhill, but there were still lots of them ... then the interstate freeway system was built (with planning help from the CEO of General Motors), and the railroad system couldn't compete for passengers. Passenger service was mostly gone 50 years ago.

But, you ask, did noone plan for the future? Well ... this is the U.S. We only build it well once, then we argue about who should keep it repaired. Which is why the majority of our bridges are in a dangerous state.

It's kinda funny seing an outside perspective on this. I live in europe and while we do have a decent rail system I always dream of a truly unified european rail network that can eliminate much of the air trafic. I don't know if that is possible with current technologies but a man can dream. Still it is pretty good what we have as the author states. I will spend an equivelant of about 50 us dollars to travel home and back for christmas, around 700 km total, which is actually cheaper than the equivalent cost in just gas if I was to drive (if I had a car that was, also not factoring bridge toll, which would make it substantially more expensive).

Image captioned `Westfriedhof Station, Stockholm.` is actually `Westfriedhof Station, Munich.`

Does anyone have comprehensive information on typical transportation department budgets in the US vs. some of these countries?

Anecdotally I had a civil engineer friend from Spain who always told me that the transportation budgets in the USA were tiny compared to in Europe.

It'd be great if it was politically feasible for the US government to take some money from the military and put it into ambitious public transportation infrastructure. National rail probably doesn't make sense but there are definitely regional and city systems that need some serious help.

Rail projects have a lot of political pushback in the US. They're expected to pay for themselves and realistically rarely manage to do so. They tend to be costed compared to cars, and somehow the cost of building and maintaining the roads is not included in that comparison, except for the relatively small fraction that comes from the gas tax.

So rail projects somehow never look economically viable compared to a magically free road system.

One problem is that even if they spent as much, infrastructure construction costs in the US are way higher than elsewhere: https://www.nytimes.com/2017/12/28/nyregion/new-york-subway-...

Very cool article. Somewhat surprised China wasn’t included on the list as they have built a high speed rail network in the last several decades that already has more track mileage than Europe’s.

Where there is a will there is a way, as the saying goes, and America, and the UK, with their profit über alles obsession are happy to let infrastructure rot and make people use woefully underfunded and overcrowded rail systems rather than trying something “new.”

One picture is mistitled. The "Westfriedhof" Station is located in Munich, not Stockholm (also it's the station closest to where I used to live in Munich for almost 30 years). It's a very popular place among photographers.

Too bad the article doesn't talk about the not so sexy trains which were numerous and allowed travellers to go from small cities to other small cities.

The big/speedy trains only cover a few paths...

subway stations from around the world, the second one is not Westfriedhof Station, Stockholm, it is in Munich Westfriedhof station!

Lots of truths in the article.


Trains is always a system that picks winners & losers. The closer u r to the nearest station - the bigger winner u r.

I lived in London. Any location not within 1km of a station is basically a prison. Even if u have a car, you cant go anywhere (since no parking), and no one will come to you.

This is a particular problem when you adding trains (and impeding car) in an existing city.

As someone living in London this is absolutely not the case for me. Mutli-mode transport is very easy in the era of Citymapper, I find myself on buses quite often; areas further from tube stations tend to have better bus services to compensate. Literally nobody I know has a car. Many own bikes and use them almost to the exclusion of public transport.

Is it less convenient to live further from a tube station? Sure! Is it a prison? Absolutely not.

Weird, so all the people taking the bus in London to and from stations must be hallucinating while they’re trapped in their prisons?

A train+bus is a much higher threshold for many people.

Will you go to a lawyer that requires a bus+train? A restaurant? A party u r not too sure about ?

Trains are relatively dependable. The last mile is a big problem.

(To be clear : i am pro public transport. I have a bike and ebike which i love, and car i hate.)

Yes, millions of people do every single day in London and other cities across the UK.

Not sure where you're going? Citymapper will route across both tube and bus, it'll even tell you exactly when you get off.

You pay using a contactless card, apple pay or similar, it works.

> Will you go to a lawyer that requires a bus+train? A restaurant? A party u r not too sure about ?

I live in Madrid, which has only a decent public transport system, not a great one like Tokyo, and I would absolutely do all those things you mention. Those thresholds only exist if your public transportation system is really bad.

I'm a zone 4-6 dweller of 40+ years and don't recognise this characterisation, I'm afraid. Three years ago Diamond Geezer published a map of all the places in London's boroughs which are more than 1 mile (yes, not km, but bear with it) from the nearest station, so you can see the "prisons" here: http://diamondgeezer.blogspot.com/2015/07/over-one-mile.html

I agree, trains are not enough to replace cars, you need much more variety. Buses are typical, but still problematic if you're pushing for small streets and pedestrian areas. Bicycles are an obvious addendum, but still not enough to cover all needs. But the evolution of personal, small-scale, lower-pollution vehicles is happening at a good pace, and I'm confident we'll have good alternatives.

Your point stands, though; a city can't think of replacing cars with a train and calling it done. Cars pack a tremendous combination of flexibility, speed, weatherproofing, isolation, etc that no other method of transport can rival by itself.

Your argument is solid, and I can identify with it. But you'd find a better audience if you typed "you are" than "u r".

Isn't it possible to rent a bike at the station?

Yup, apps like CityMapper will even provide direction based on a Train and Bike hire route.

In Japan, suburb where train stations get more sparse is served by sprawling bus routes. Each suburban stations have at least 3-4 bus stops going to various destinations. And parking is plenty in suburb, but a lot of time it's less hassle and faster to use train.

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact