15 ms·

Yeah this is all very odd and everyone is lapping it up. He claims to want to reduce tunneling costs by a factor of 10, but starts by quoting a very high benchm

by erentz 9y ago

Yeah this is all very odd and everyone is lapping it up. He claims to want to reduce tunneling costs by a factor of 10, but starts by quoting a very high benchmark of $1 billion per mile which is quite easy to beat. Then he reduces diameters to something much smaller than all the expensive mega projects, into the range of something where technology is well established and costs are already 1/10th of his quoted $1 billion per mile. But then for some reason the solution has to involve putting cars into these tunnels. If what he thinks he is on to is viable with cars in those tunnels, it's even more viable with trains carrying 1,000 passengers each coming every 3 minutes. So why aren't we building these subways with trains? The largest part of the problem to solve here for the US is not the technology of tunneling. It's the process of engineering and building infrastructure like this that seems to become way more expensive in the US compared to places like Spain. I do believe he could reduce costs by vertically integrating the whole thing. But then we should ask why don't we do that anyway?

CodeWriter23 9y ago
He's trying to hit a price point. We're not completely allergic to the $20M freeway bridge that's about 100ft long, and comes with heavy upkeep costs over the long term. So, $100M per mile with Upkeep Lite™ sounds good when making conversation about the topic. Also the $1B/mile is the cost of the Purple Line extension in Los Angeles. Worst case cost no doubt. A real figure nevertheless.
- objclxt 9y ago
  > So, $100M per mile with Upkeep Lite™ sounds good when making conversation about the topic. Also the $1B/mile is the cost of the Purple Line extension in Los Angeles. Worst case cost no doubt. A real figure nevertheless. The figure is way out. Crossrail in London has a $20 billion total budget (£15b GBP), with 26 miles of tunnel[1]. But the tunnel is only part of the project, several billion is being spent on new regular track and station rebuilds, so the tunneling figure is really closer to $15 billion. That's about $570 million per mile, but the Crossrail tunnels are just shy of 50% bigger in diameter than what Musk wants to build (20.5' vs 14'). And as the FAQ says, "reducing the diameter in half reduces tunneling costs by 3-4 times". LA's tunnel might be costing $1B/mile, but how much of that cost is actual construction versus bureaucracy and mismanagement? [1]: http://www.crossrail.co.uk/construction/ http://www.crossrail.co.uk/construction/
  - tomjakubowski 9y ago
    I don't know where this "$1B/mile" just for tunneling figure comes from. That's certainly not the tunneling cost for any current subway extension project in Los Angeles: all-in (i.e. not only tunneling) estimates for the first phase of the Purple Line extension are at $2.8B for 3.9 miles of heavy rail subway; the full Purple Line Extension is estimated at $6.3B for 9 miles of heavy rail subway. The closest thing is the Regional Connector project, which will be 1.9 miles of light rail and has a budget of $1.8B. But that's the entire project budget. Tunneling costs are only part of it; it's hard to find a finely grained budget, but total construction (including four new underground stations) is budgeted at $1.2B here [1], which is still "just" $600M/mile. [1]: https://mtadash.mlmprojectservices.com/projects/5773#/cost https://mtadash.mlmprojectservices.com/projects/5773#/cost
    - kuschku 9y ago
      In one of the subway projects in Hamburg, a cost of 1.1 billion EUR per kilometre was reached. But that project was building a subway including station directly below a riverbed.
      - Piskvorrr 9y ago
        Well, that's some adverse conditions - a silt riverbed behaves quite like water, so you're actually building a submerged part of the city. Look at London's early tunnels and their cost overruns (and the maintenance due to leaks).
  - tigershark 9y ago
    No, crossrail tunnels are WAY less expensive than that: http://www.crossrail.co.uk/news/articles/crossrail-awards-major-tunnelling-contracts-worth-125bn http://www.crossrail.co.uk/news/articles/crossrail-awards-ma... It's about 145M$ per mile that is well in the range of 10 times less the outrageous cost that they pay in US for a single mile.
prostoalex 9y ago
> But then for some reason the solution has to involve putting cars into these tunnels. If what he thinks he is on to is viable with cars in those tunnels, it's even more viable with trains carrying 1,000 passengers each coming every 3 minutes. So why aren't we building these subways with trains? Because with trains you never build just a tunnel - there are train stations, escalators leading in and out, ticketing booths/machines, electrical work to power all that fun stuff, connectors to existing depots so that maintenance workers can come and service the cars and the engines. His idea is closer to Madrid, which has dug tunnels to relieve the city streets of the traffic. The tunnels themselves are very unassuming and look like onramps and underground parking garage entrances.
- djsumdog 9y ago
  But even with the stations, escalators and everything else, you're talking about orders of magnitude between trains and cars. London during peak hour has trains arriving less than 1 min 30 sec apart. They move an absolutely insane amount of people. Singapore's trains are entirely automated and arrive at <2min intervals at peak. Everything about Musk's idea seems absolutely terrible. If America had better trains and subways, it could move massive amounts of people. Train automation is a solved problem, and currently implemented in cities around the world. There is absolutely no possible way in any reality where individual car/pod transport could even approach the capacity of real mass transit: http://penguindreams.org/blog/self-driving-cars-will-not-solve-the-transportation-problem/ http://penguindreams.org/blog/self-driving-cars-will-not-sol... Americans need to get over our wasteful individualism, our "alone time" in cars and realize you can read a book, watch a movie or do something not involving driving on real public transport. Buses need to stop being just a means of transport for the poor.
  - philwelch 9y ago
    If you own a car company, you want cars to be a big part of the transportation system.
  - rurban 9y ago
    Absolutely. And the biggest problem is what to do with all those cars in the city. It's an enormous waste of space to park all those cars, besides the enormous hazards of the moving cars, not being to be able to walk. It destroyed cities. That's why public mass transport was created and was so successful in the big cities. US big automotive destroyed most of this working infrastructure in the last century (E.g. Los Angeles, Houston, ...) but it's still cheaper to build mass transport than to support individual transport. Just ask a New Yorker or a foreigner. Tunnels are a hazard for individual traffic. That's why pods do help. But trains are still preferred. It's a good alternative idea to build long range tunnels. He is right. But the focus needs to be mass transport, not individual. It's a matter of costs.
josephg 9y ago
> But then we should ask why don't we do that anyway? Sure. Why didn't we have electric cars before tesla? Why didn't Boeing or Lockheed Martin make reusable rockets? Alan Kay said something[1] recently that has been bouncing around my head for the past few days. He said that many people think darwinian processes (like the economy) optimize. "One my degrees is in molecular biology, and I can tell you, any biologist would say they're absolutely not optimized. The whole point is to fit into some niche in an environment. And if the environment isn't the right kind of environment evolution isn't going to give you something interesting". In the context of the economy, if there's no innovation, and not enough competition, and no new players entering a market, I don't think its all that surprising that the existing companies might stop reaching to be more efficient. It seems strange to think that companies would leave money on the table, but in older industries I think this is exactly what happens. The money just requires too much organizational churn for their corporate appetite. [1] https://youtu.be/DIR6Rmhm3To?t=54m17s https://youtu.be/DIR6Rmhm3To?t=54m17s
- cel1ne 9y ago
  Off topic: The more I think about organizations the more I they seem to be the key players preventing change to me. (Which is not necessarily a bad thing) Business says: We had this great idea, now let's freeze everything in time and let's make money of this snapshot forever, because who knows when we'll find gold again. Entertainment industry likes to do that especially.
  - ajdlinux 9y ago
    The phrase "exploitation versus exploration" seems appropriate here.
  - Angostura 9y ago
    That's not businesses - that's humans. Or do you apply for a new job every two weeks whenever you see a higher salary advertised?
    - cel1ne 9y ago
rmason 9y ago
One of the reasons that subways are built only in very large cities is that you need a very high population density to make it worthwhile because of the cost of digging the tunnels. But what if the cost were a tenth as much? I posit that for a city of 500,000 - 1,000,000 it might become practical to have subways and that radically changes the transit equation. So I am very bullish on Musk bringing down the cost of tunneling. It all depends on how low he's able to reduce the cost.
- prawn 9y ago
  I've got confidence in them too. I think internet commenters are far too quick to overlook the fact that Musk is not going to push forward a venture like this without loads of research from a team of really smart people. It's not like it's just the one Elon with a fantasy about tunnels. It'd start with a bunch of talented people brainstorming and then, early in the process, they'd surely be speaking with really knowledgeable people who know the industry and pitfalls. If smart and motivated teams can't do something, can anyone? With something like soil, maybe they figure the technology will be useful even if it only services projects with easy terrain and environments and avoids areas with swamps.
  - aedron 9y ago
    > Musk is not going to push forward a venture like this without loads of research from a team of really smart people. It's not like it's just the one Elon with a fantasy about tunnels. It'd start with a bunch of talented people brainstorming and then, early in the process, they'd surely be speaking with really knowledgeable people who know the industry and pitfalls. Actually, the story is that Musk made it up on a whim while stuck in L.A. traffic, founded the company the same day and bought a used drilling rig which started digging in the Tesla parking lot.
    - prawn 9y ago
      OK, it'd start with a user frustrated by a problem, then go to brainstorming with colleagues/employees/peers/friends and then go to industry experts and hires with direct experience. ;) Really the main difference is that someone without money and follow through would've just sat in traffic and complained.
Tomminn 9y ago
I think the sled idea is really interesting. Imagine a conveyor belt like you have in airports. Except it moves cars at 150 mph through an underground tunnel instead of people at 15mph. Traffic is often massively asymmetric so you can just reverse the direction it travels at noon. It seems plausible that keeping a conveyor belt in good service is cheaper than keeping a fleet of railway cars is. It's also significantly less hassle as you don't need to figure out where to put railway cars when they get the end of the track. Finally it's costs would scale fairly linearly with distance, making short belts totally doable in a way that short railway systems are not. The natural throughput of the system will be high. Like a new car on the belt every x seconds, where x is not a big number, seems totally plausible.
- kuschku 9y ago
  The throughput of "car directly after car" is about 10 times lower than that of a railway with a frequency of 90 seconds.
  - Robotbeat 9y ago
    How about bus pods directly after bus pods? It'd be comparable or even higher (20 passengers/pod).
    - ancientworldnow 9y ago
      Congratulations, you just reinvented the train.
      - marcosdumay 9y ago
        Yes, but it's a train that also carries cars. And quite possibly freight. Those aren't new either, but not common for urban transportation.
- ascorbic 9y ago
  Generously assuming that it averages 150mph, and very generously assuming 1.5 people per car and estimating from the video a gap between sleds of 40m gives a capacity of around 9,000 pax/hr per line. The Tube in London has trains with around 900 pax capacity, and they run at around 90 second intervals at peak time, giving 40 trains per hour and 36,000 pax/hr.
Cthulhu_ 9y ago
With the capacity in mind, this project will end up having much lower throughput than a regular road, even when taking congestion in mind - unless they build dozens of parallel tubes, in which case the costs will become very, very large. I also don't see it becoming a long-distance alternative, again because tunneling is a slow and long process. It would be interesting I guess if they combined it with the Hyperloop concept, board the tube system in the city, transition to a (cheaper?) above-ground vacuum tunnel system at a transit station. But it's all very science-fiction and, if anything, it's not going to be cheap. Like, at all. At least it's a commercial project, not something a government has to pay for.
- icc97 9y ago
  It's not science fiction, it's just drilling holes in the ground vs waiting for flying cars. It may be expensive - but you only have to build each tunnel once. There's a startup that got some hype recently for flying an electric airplane between London and Paris - when as someone commented on the HN story [0] you can just take the Eurostar through the tunnel - which is electric and takes just as long and doesn't cost any more. [0]: https://news.ycombinator.com/item?id=13930880 https://news.ycombinator.com/item?id=13930880
- haihaibye 9y ago
  I'd guess 1 lane at 200km/h without stopping is ~7-10 lanes rush hour traffic.
  - ryandamm 9y ago
    In traffic, you have to look at both throughput and carrying capacity. Traffic jams usually result when the carrying capacity gets strained, not throughput. (The throughput of stop-and-go traffic is lower than the throughput of 70 mph traffic, but the road can carry more cars in total, so that equilibrium dominates when the load coming from surface streets exceeds some threshold.)
  - Piskvorrr 9y ago
    Without stopping and without merging. The proposal acts as if entering/exiting a tunnel at any depth is zero-cost and zero-size.
bamboozled 9y ago
"If what he thinks he is on to is viable with cars in those tunnels, it's even more viable with trains carrying 1,000 passengers each coming every 3 minutes" When it's all said and done, Elon is a car salesman and entrepreneur, I guess that's why?
godzillabrennus 9y ago
It's mostly due to how much people prefer personal transportation over public transit. There are obvious benefits of being able to come and go as an individual instead of relying on public transit. Hygiene, safety, and storage all pop to my mind.
jlebrech 9y ago
the smaller diameter is because he's building for cars and not roads, and as for trains I think he need the early adopters aka, the rich to invest. mass transit is for the poor and doesn't attract investors. but in the long run it's probably better. but if you look at the mockup video there's an mpv that enters the tunnels, not just teslas. I think it's move polivalent to transport vehicles (personal or mpv) that way the last 1-5 miles can be resumed by the vehicle and they don't have to stop everywhere (which is the main crutch of trains), you want to cover as much area and not have "why don't you stop near my house" syndrome. there's also nothing stopping bicycle riders from using the system provided there are seats to strap into and goggles to wear.
IshKebab 9y ago
> but starts by quoting a very high benchmark of $1 billion per mile which is quite easy to beat Now I know why big companies do those reading comprehension tests... He didn't say that was a benchmark at all. He explicitly said that was one of the highest costs for tunnels. > with some projects costing as much as $1 billion per mile
TeMPOraL 9y ago
> If what he thinks he is on to is viable with cars in those tunnels, it's even more viable with trains carrying 1,000 passengers each coming every 3 minutes. A few reasons off the top of my head: - flexibility; car infrastructure is much more flexible than train infrastructure - incremental construction - cars already exist, you're only solving the problem of moving them around (building train infrastructure means building more trains with additional train-supporting infrastructure and more railroads outside of tunnels) - meshes well with electrification of car transportation - allows to reduce fossil fuel polution with minimum impact on people's habits - instead of asking someone to start using trains, you just ask them to drive into a tunnel, and for the duration of tunnel transit, the pollution (and fuel costs) magically disappear - related, it's easier do incremental changes than all-out revolutions - R&D in efficient boring could be of vital importance to future Mars colony - last but not least, electric PRT system (aka. "packed-switched" public transport) could be, and probably will, incrementally implemented with self-driving electric cars; this again meshes nicely with that future (benefits of PRT over other forms of public transit are a longer topic)
- mikepurvis 9y ago
  > you just ask them to drive into a tunnel Presumably whether this system is privately or publicly operated, there would be a major toll associated with its use. So the ask isn't just "drive into a tunnel", it's "drive into a tunnel and pay $$$", compared with a new subway line integrated with the existing mass transit system (trains and buses) and under the same fare.
  - TeMPOraL 9y ago
    I suppose you could let people buy monthly tickets the same way they do for public transit, and handle the rest the same way automated ticketing of speeding is handled.
Touche 9y ago
I think you're missing the point. Of course trains are better. So why don't we have more trains? Because people have opinions on trains. So instead of building train tunnels you debate it for 20 years in town halls and NIMBY protests and nothing changes. This is clearly a hack around that issue. The article mentions a Hyperloop several times, that's the ultimate goal here.
- noobermin 9y ago
  As interest in car ownership drops as the Millennials age and take the older generations' place, that problem may fix itself.
  - Touche 9y ago
    What do you mean by interest in car ownership?
    - noobermin 9y ago
      While this article[0] demonstrates that numbers have improved slightly for actual car ownership as millenials age, it still contains other tidbits and statistics (like the having driver's license by 18 stat at 60% vs. 80% in 1980) that reflect views on car ownership amongst millenials today. Anecdote, I don't have a lot of mileage (pun unintended), but every girl I've ever dated did not have license, and neither do I, and I am 27. I live alone and bike and bus ride for the most part. [0] http://www.latimes.com/business/autos/la-fi-hy-millennials-cars-20161223-story.html http://www.latimes.com/business/autos/la-fi-hy-millennials-c...
      - Touche 9y ago
        Don't make the mistake of making macro predictions based on a niche segment. I think generally people aren't "interested" in car ownership; they are interested in fulfilling their transportation needs as best as possible. Car ownership just happens to fulfill that the best right now. There is the possibility of increased urbanization and improvements to self-driving car technology will make it economically feasible for car ownership to be unnecessary for most, but I highly doubt that happens within the millennial generation. Probably the next one.
nix0n 9y ago
> If what he thinks he is on to is viable with cars in those tunnels, it's even more viable with trains In Europe, that would be true, but I think in Southern California he can charge exorbitantly per car and people will pay it.
mikeash 9y ago
An interesting aspect of carrying cars rather than trains is that it makes an incomplete network far more useful. A public transportation network's usefulness is roughly quadratic in proportion to its size, because it needs to cover both your origin and your destination. A small subway system is nearly useless, because even people who live or work near it probably don't live and work near it. You can mitigate this to an extent with busses, taxis, letting people park at stations, etc., but there's still a lot of friction. By carrying cars, these tunnels will act as an extension of the road network, which almost certainly covers your origin and destination already. The tunnel part doesn't need to serve the ends, it just needs to exist somewhere in the middle to be useful. If you build one subway line, it's pretty much "great, let's finish the system so we can have something useful." If you build one Boring Tunnel, it would be immediately useful for anyone whose car trips are vaguely in that direction. I'm really skeptical it'll all work out, but I don't think it should be thought of as a lame subway.
- ambicapter 9y ago
  I was thinking the other day how sweet it would be if it was easy to take your bike on the subway.
  - mikeash 9y ago
    The DC Metro lets you do this, and I've taken advantage a couple of times. It is pretty nice. It vastly increases the effective reach of the network. Unfortunately, they don't allow it during rush hour due to space constraints, so it's not nearly as useful as it could be.
  - dsr_ 9y ago
    The MBTA allows this during non-rush hour times, and most buses have fold-down bike carriers in front.
  - Piskvorrr 9y ago
    Indeed. This was implemented here ~15 years ago. The initial fears "Just ride the damn bikes! Trains full of cyclists! Cats and dogs living together!!!" did not materialize, and it's a major improvement on a bike's radius.
futun 9y ago
Not to mention that (like other Musk projects) scalability is a huge issue. What's the throughput of cars on electric sleds through tunnels? Crickets...
adamsea 9y ago
I kinda wonder if the tunnel-borers are meant for Mars.
ruste 9y ago
It's already been mentioned that putting cars in the tunnels increases the usefulness of the tunnels quickly without requiring a large network of tunnels right off the bat. I think it's always possible to convert the tunnels to trains at a later date with a much lower cost than the initial investment. I think having the tunnels be designed for cars makes a lot of sense for Elon in particular because he can design the tunnels as an ideal environment for the first generation of self driving cars. This would give him a good start getting self driving cars on the road.
jpalomaki 9y ago
In order to be useful trains need to form large network. Few mile subway route is not so interesting. A tunnel for cars is different. Even a single and relatively short tunnel is useful and people would probably pay something for access. This means you would start making money on the investment quite quickly. I agree that trains are the right thing to do, but that is much larger project and likely too big to be an option for private company.

by erentz 9y ago

CodeWriter23 9y ago

He's trying to hit a price point. We're not completely allergic to the $20M freeway bridge that's about 100ft long, and comes with heavy upkeep costs over the long term. So, $100M per mile with Upkeep Lite™ sounds good when making conversation about the topic. Also the $1B/mile is the cost of the Purple Line extension in Los Angeles. Worst case cost no doubt. A real figure nevertheless.

objclxt 9y ago
> So, $100M per mile with Upkeep Lite™ sounds good when making conversation about the topic. Also the $1B/mile is the cost of the Purple Line extension in Los Angeles. Worst case cost no doubt. A real figure nevertheless. The figure is way out. Crossrail in London has a $20 billion total budget (£15b GBP), with 26 miles of tunnel[1]. But the tunnel is only part of the project, several billion is being spent on new regular track and station rebuilds, so the tunneling figure is really closer to $15 billion. That's about $570 million per mile, but the Crossrail tunnels are just shy of 50% bigger in diameter than what Musk wants to build (20.5' vs 14'). And as the FAQ says, "reducing the diameter in half reduces tunneling costs by 3-4 times". LA's tunnel might be costing $1B/mile, but how much of that cost is actual construction versus bureaucracy and mismanagement? [1]: http://www.crossrail.co.uk/construction/ http://www.crossrail.co.uk/construction/
- tomjakubowski 9y ago
  I don't know where this "$1B/mile" just for tunneling figure comes from. That's certainly not the tunneling cost for any current subway extension project in Los Angeles: all-in (i.e. not only tunneling) estimates for the first phase of the Purple Line extension are at $2.8B for 3.9 miles of heavy rail subway; the full Purple Line Extension is estimated at $6.3B for 9 miles of heavy rail subway. The closest thing is the Regional Connector project, which will be 1.9 miles of light rail and has a budget of $1.8B. But that's the entire project budget. Tunneling costs are only part of it; it's hard to find a finely grained budget, but total construction (including four new underground stations) is budgeted at $1.2B here [1], which is still "just" $600M/mile. [1]: https://mtadash.mlmprojectservices.com/projects/5773#/cost https://mtadash.mlmprojectservices.com/projects/5773#/cost
  - kuschku 9y ago
    In one of the subway projects in Hamburg, a cost of 1.1 billion EUR per kilometre was reached. But that project was building a subway including station directly below a riverbed.
    - Piskvorrr 9y ago
      Well, that's some adverse conditions - a silt riverbed behaves quite like water, so you're actually building a submerged part of the city. Look at London's early tunnels and their cost overruns (and the maintenance due to leaks).
- tigershark 9y ago
  No, crossrail tunnels are WAY less expensive than that: http://www.crossrail.co.uk/news/articles/crossrail-awards-major-tunnelling-contracts-worth-125bn http://www.crossrail.co.uk/news/articles/crossrail-awards-ma... It's about 145M$ per mile that is well in the range of 10 times less the outrageous cost that they pay in US for a single mile.

prostoalex 9y ago

> But then for some reason the solution has to involve putting cars into these tunnels. If what he thinks he is on to is viable with cars in those tunnels, it's even more viable with trains carrying 1,000 passengers each coming every 3 minutes. So why aren't we building these subways with trains? Because with trains you never build just a tunnel - there are train stations, escalators leading in and out, ticketing booths/machines, electrical work to power all that fun stuff, connectors to existing depots so that maintenance workers can come and service the cars and the engines. His idea is closer to Madrid, which has dug tunnels to relieve the city streets of the traffic. The tunnels themselves are very unassuming and look like onramps and underground parking garage entrances.

djsumdog 9y ago
But even with the stations, escalators and everything else, you're talking about orders of magnitude between trains and cars. London during peak hour has trains arriving less than 1 min 30 sec apart. They move an absolutely insane amount of people. Singapore's trains are entirely automated and arrive at <2min intervals at peak. Everything about Musk's idea seems absolutely terrible. If America had better trains and subways, it could move massive amounts of people. Train automation is a solved problem, and currently implemented in cities around the world. There is absolutely no possible way in any reality where individual car/pod transport could even approach the capacity of real mass transit: http://penguindreams.org/blog/self-driving-cars-will-not-solve-the-transportation-problem/ http://penguindreams.org/blog/self-driving-cars-will-not-sol... Americans need to get over our wasteful individualism, our "alone time" in cars and realize you can read a book, watch a movie or do something not involving driving on real public transport. Buses need to stop being just a means of transport for the poor.
- philwelch 9y ago
  If you own a car company, you want cars to be a big part of the transportation system.
- rurban 9y ago
  Absolutely. And the biggest problem is what to do with all those cars in the city. It's an enormous waste of space to park all those cars, besides the enormous hazards of the moving cars, not being to be able to walk. It destroyed cities. That's why public mass transport was created and was so successful in the big cities. US big automotive destroyed most of this working infrastructure in the last century (E.g. Los Angeles, Houston, ...) but it's still cheaper to build mass transport than to support individual transport. Just ask a New Yorker or a foreigner. Tunnels are a hazard for individual traffic. That's why pods do help. But trains are still preferred. It's a good alternative idea to build long range tunnels. He is right. But the focus needs to be mass transport, not individual. It's a matter of costs.

josephg 9y ago

> But then we should ask why don't we do that anyway? Sure. Why didn't we have electric cars before tesla? Why didn't Boeing or Lockheed Martin make reusable rockets? Alan Kay said something[1] recently that has been bouncing around my head for the past few days. He said that many people think darwinian processes (like the economy) optimize. "One my degrees is in molecular biology, and I can tell you, any biologist would say they're absolutely not optimized. The whole point is to fit into some niche in an environment. And if the environment isn't the right kind of environment evolution isn't going to give you something interesting". In the context of the economy, if there's no innovation, and not enough competition, and no new players entering a market, I don't think its all that surprising that the existing companies might stop reaching to be more efficient. It seems strange to think that companies would leave money on the table, but in older industries I think this is exactly what happens. The money just requires too much organizational churn for their corporate appetite. [1] https://youtu.be/DIR6Rmhm3To?t=54m17s https://youtu.be/DIR6Rmhm3To?t=54m17s

cel1ne 9y ago
Off topic: The more I think about organizations the more I they seem to be the key players preventing change to me. (Which is not necessarily a bad thing) Business says: We had this great idea, now let's freeze everything in time and let's make money of this snapshot forever, because who knows when we'll find gold again. Entertainment industry likes to do that especially.
- ajdlinux 9y ago
  The phrase "exploitation versus exploration" seems appropriate here.
- Angostura 9y ago
  That's not businesses - that's humans. Or do you apply for a new job every two weeks whenever you see a higher salary advertised?
  - cel1ne 9y ago

rmason 9y ago

One of the reasons that subways are built only in very large cities is that you need a very high population density to make it worthwhile because of the cost of digging the tunnels. But what if the cost were a tenth as much? I posit that for a city of 500,000 - 1,000,000 it might become practical to have subways and that radically changes the transit equation. So I am very bullish on Musk bringing down the cost of tunneling. It all depends on how low he's able to reduce the cost.

prawn 9y ago
I've got confidence in them too. I think internet commenters are far too quick to overlook the fact that Musk is not going to push forward a venture like this without loads of research from a team of really smart people. It's not like it's just the one Elon with a fantasy about tunnels. It'd start with a bunch of talented people brainstorming and then, early in the process, they'd surely be speaking with really knowledgeable people who know the industry and pitfalls. If smart and motivated teams can't do something, can anyone? With something like soil, maybe they figure the technology will be useful even if it only services projects with easy terrain and environments and avoids areas with swamps.
- aedron 9y ago
  > Musk is not going to push forward a venture like this without loads of research from a team of really smart people. It's not like it's just the one Elon with a fantasy about tunnels. It'd start with a bunch of talented people brainstorming and then, early in the process, they'd surely be speaking with really knowledgeable people who know the industry and pitfalls. Actually, the story is that Musk made it up on a whim while stuck in L.A. traffic, founded the company the same day and bought a used drilling rig which started digging in the Tesla parking lot.
  - prawn 9y ago
    OK, it'd start with a user frustrated by a problem, then go to brainstorming with colleagues/employees/peers/friends and then go to industry experts and hires with direct experience. ;) Really the main difference is that someone without money and follow through would've just sat in traffic and complained.

Tomminn 9y ago

I think the sled idea is really interesting. Imagine a conveyor belt like you have in airports. Except it moves cars at 150 mph through an underground tunnel instead of people at 15mph. Traffic is often massively asymmetric so you can just reverse the direction it travels at noon. It seems plausible that keeping a conveyor belt in good service is cheaper than keeping a fleet of railway cars is. It's also significantly less hassle as you don't need to figure out where to put railway cars when they get the end of the track. Finally it's costs would scale fairly linearly with distance, making short belts totally doable in a way that short railway systems are not. The natural throughput of the system will be high. Like a new car on the belt every x seconds, where x is not a big number, seems totally plausible.

kuschku 9y ago
The throughput of "car directly after car" is about 10 times lower than that of a railway with a frequency of 90 seconds.
- Robotbeat 9y ago
  How about bus pods directly after bus pods? It'd be comparable or even higher (20 passengers/pod).
  - ancientworldnow 9y ago
    Congratulations, you just reinvented the train.
    - marcosdumay 9y ago
      Yes, but it's a train that also carries cars. And quite possibly freight. Those aren't new either, but not common for urban transportation.
ascorbic 9y ago
Generously assuming that it averages 150mph, and very generously assuming 1.5 people per car and estimating from the video a gap between sleds of 40m gives a capacity of around 9,000 pax/hr per line. The Tube in London has trains with around 900 pax capacity, and they run at around 90 second intervals at peak time, giving 40 trains per hour and 36,000 pax/hr.

Cthulhu_ 9y ago

With the capacity in mind, this project will end up having much lower throughput than a regular road, even when taking congestion in mind - unless they build dozens of parallel tubes, in which case the costs will become very, very large. I also don't see it becoming a long-distance alternative, again because tunneling is a slow and long process. It would be interesting I guess if they combined it with the Hyperloop concept, board the tube system in the city, transition to a (cheaper?) above-ground vacuum tunnel system at a transit station. But it's all very science-fiction and, if anything, it's not going to be cheap. Like, at all. At least it's a commercial project, not something a government has to pay for.

icc97 9y ago
It's not science fiction, it's just drilling holes in the ground vs waiting for flying cars. It may be expensive - but you only have to build each tunnel once. There's a startup that got some hype recently for flying an electric airplane between London and Paris - when as someone commented on the HN story [0] you can just take the Eurostar through the tunnel - which is electric and takes just as long and doesn't cost any more. [0]: https://news.ycombinator.com/item?id=13930880 https://news.ycombinator.com/item?id=13930880
haihaibye 9y ago
I'd guess 1 lane at 200km/h without stopping is ~7-10 lanes rush hour traffic.
- ryandamm 9y ago
  In traffic, you have to look at both throughput and carrying capacity. Traffic jams usually result when the carrying capacity gets strained, not throughput. (The throughput of stop-and-go traffic is lower than the throughput of 70 mph traffic, but the road can carry more cars in total, so that equilibrium dominates when the load coming from surface streets exceeds some threshold.)
- Piskvorrr 9y ago
  Without stopping and without merging. The proposal acts as if entering/exiting a tunnel at any depth is zero-cost and zero-size.

bamboozled 9y ago

"If what he thinks he is on to is viable with cars in those tunnels, it's even more viable with trains carrying 1,000 passengers each coming every 3 minutes" When it's all said and done, Elon is a car salesman and entrepreneur, I guess that's why?

godzillabrennus 9y ago

It's mostly due to how much people prefer personal transportation over public transit. There are obvious benefits of being able to come and go as an individual instead of relying on public transit. Hygiene, safety, and storage all pop to my mind.

jlebrech 9y ago

the smaller diameter is because he's building for cars and not roads, and as for trains I think he need the early adopters aka, the rich to invest. mass transit is for the poor and doesn't attract investors. but in the long run it's probably better. but if you look at the mockup video there's an mpv that enters the tunnels, not just teslas. I think it's move polivalent to transport vehicles (personal or mpv) that way the last 1-5 miles can be resumed by the vehicle and they don't have to stop everywhere (which is the main crutch of trains), you want to cover as much area and not have "why don't you stop near my house" syndrome. there's also nothing stopping bicycle riders from using the system provided there are seats to strap into and goggles to wear.

IshKebab 9y ago

> but starts by quoting a very high benchmark of $1 billion per mile which is quite easy to beat Now I know why big companies do those reading comprehension tests... He didn't say that was a benchmark at all. He explicitly said that was one of the highest costs for tunnels. > with some projects costing as much as $1 billion per mile

TeMPOraL 9y ago

> If what he thinks he is on to is viable with cars in those tunnels, it's even more viable with trains carrying 1,000 passengers each coming every 3 minutes. A few reasons off the top of my head: - flexibility; car infrastructure is much more flexible than train infrastructure - incremental construction - cars already exist, you're only solving the problem of moving them around (building train infrastructure means building more trains with additional train-supporting infrastructure and more railroads outside of tunnels) - meshes well with electrification of car transportation - allows to reduce fossil fuel polution with minimum impact on people's habits - instead of asking someone to start using trains, you just ask them to drive into a tunnel, and for the duration of tunnel transit, the pollution (and fuel costs) magically disappear - related, it's easier do incremental changes than all-out revolutions - R&D in efficient boring could be of vital importance to future Mars colony - last but not least, electric PRT system (aka. "packed-switched" public transport) could be, and probably will, incrementally implemented with self-driving electric cars; this again meshes nicely with that future (benefits of PRT over other forms of public transit are a longer topic)

mikepurvis 9y ago
> you just ask them to drive into a tunnel Presumably whether this system is privately or publicly operated, there would be a major toll associated with its use. So the ask isn't just "drive into a tunnel", it's "drive into a tunnel and pay $$$", compared with a new subway line integrated with the existing mass transit system (trains and buses) and under the same fare.
- TeMPOraL 9y ago
  I suppose you could let people buy monthly tickets the same way they do for public transit, and handle the rest the same way automated ticketing of speeding is handled.

Touche 9y ago

I think you're missing the point. Of course trains are better. So why don't we have more trains? Because people have opinions on trains. So instead of building train tunnels you debate it for 20 years in town halls and NIMBY protests and nothing changes. This is clearly a hack around that issue. The article mentions a Hyperloop several times, that's the ultimate goal here.

noobermin 9y ago
As interest in car ownership drops as the Millennials age and take the older generations' place, that problem may fix itself.
- Touche 9y ago
  What do you mean by interest in car ownership?
  - noobermin 9y ago
    While this article[0] demonstrates that numbers have improved slightly for actual car ownership as millenials age, it still contains other tidbits and statistics (like the having driver's license by 18 stat at 60% vs. 80% in 1980) that reflect views on car ownership amongst millenials today. Anecdote, I don't have a lot of mileage (pun unintended), but every girl I've ever dated did not have license, and neither do I, and I am 27. I live alone and bike and bus ride for the most part. [0] http://www.latimes.com/business/autos/la-fi-hy-millennials-cars-20161223-story.html http://www.latimes.com/business/autos/la-fi-hy-millennials-c...
    - Touche 9y ago
      Don't make the mistake of making macro predictions based on a niche segment. I think generally people aren't "interested" in car ownership; they are interested in fulfilling their transportation needs as best as possible. Car ownership just happens to fulfill that the best right now. There is the possibility of increased urbanization and improvements to self-driving car technology will make it economically feasible for car ownership to be unnecessary for most, but I highly doubt that happens within the millennial generation. Probably the next one.

nix0n 9y ago

> If what he thinks he is on to is viable with cars in those tunnels, it's even more viable with trains In Europe, that would be true, but I think in Southern California he can charge exorbitantly per car and people will pay it.

mikeash 9y ago

An interesting aspect of carrying cars rather than trains is that it makes an incomplete network far more useful. A public transportation network's usefulness is roughly quadratic in proportion to its size, because it needs to cover both your origin and your destination. A small subway system is nearly useless, because even people who live or work near it probably don't live and work near it. You can mitigate this to an extent with busses, taxis, letting people park at stations, etc., but there's still a lot of friction. By carrying cars, these tunnels will act as an extension of the road network, which almost certainly covers your origin and destination already. The tunnel part doesn't need to serve the ends, it just needs to exist somewhere in the middle to be useful. If you build one subway line, it's pretty much "great, let's finish the system so we can have something useful." If you build one Boring Tunnel, it would be immediately useful for anyone whose car trips are vaguely in that direction. I'm really skeptical it'll all work out, but I don't think it should be thought of as a lame subway.

ambicapter 9y ago
I was thinking the other day how sweet it would be if it was easy to take your bike on the subway.
- mikeash 9y ago
  The DC Metro lets you do this, and I've taken advantage a couple of times. It is pretty nice. It vastly increases the effective reach of the network. Unfortunately, they don't allow it during rush hour due to space constraints, so it's not nearly as useful as it could be.
- dsr_ 9y ago
  The MBTA allows this during non-rush hour times, and most buses have fold-down bike carriers in front.
- Piskvorrr 9y ago
  Indeed. This was implemented here ~15 years ago. The initial fears "Just ride the damn bikes! Trains full of cyclists! Cats and dogs living together!!!" did not materialize, and it's a major improvement on a bike's radius.

futun 9y ago

Not to mention that (like other Musk projects) scalability is a huge issue. What's the throughput of cars on electric sleds through tunnels? Crickets...

adamsea 9y ago

I kinda wonder if the tunnel-borers are meant for Mars.

ruste 9y ago

It's already been mentioned that putting cars in the tunnels increases the usefulness of the tunnels quickly without requiring a large network of tunnels right off the bat. I think it's always possible to convert the tunnels to trains at a later date with a much lower cost than the initial investment. I think having the tunnels be designed for cars makes a lot of sense for Elon in particular because he can design the tunnels as an ideal environment for the first generation of self driving cars. This would give him a good start getting self driving cars on the road.

jpalomaki 9y ago

In order to be useful trains need to form large network. Few mile subway route is not so interesting. A tunnel for cars is different. Even a single and relatively short tunnel is useful and people would probably pay something for access. This means you would start making money on the investment quite quickly. I agree that trains are the right thing to do, but that is much larger project and likely too big to be an option for private company.