Why not a 105 mile perimeter circle? would that not potentially cut the average commute time between random points on the line?
I guess that might change some dynamic related to "interior" vs "exterior" properties and potentially make the bullet train slower to account for curved tracks.
I'd guess that being coastal is pretty important here as well. But definitely interesting to think about in the context of a hypothetical new landlocked city.
Pedantically, I think you probably mean 105 mile/(2*pi) radius circle.
I think the idea is you have a subway that runs the length. One set of trains in both directions that are express and another set with lots of stops when you only want to go a couple of miles up the road, say to work. You wouldn't have to tunnel, just dig a ditch. Then have a park in place of the street, perhaps with bike lanes. Any cars would be parked just outside the city but still within walking distance. I personally think it's an interesting idea and hope it gets built.
Detroit is rebuilding the city around neighborhoods. The idea is you reinforce neighborhoods around the idea that most of what you need is no longer than a ten minute walk. They've built greenways connecting places to make this possible. Kind of ironic for the city of cars?
Interesting thought experiment. Assuming the line is 1/4 mile wide, it would have the same area as a city with a 5 mile diameter, so slightly smaller than SF.
I would like to hear the proposed advantages of the design.
One advantage that comes to mind is that the entire city could run off of a single transportation corridor.
Also a line has the maximum perimeter to surface area. Maybe this is relevant for expansion?
It’s one arterial road. I have lived in a city like that. The first arterial road has homes and businesses and then as population grew, new parallel roads are built.
But now we have to have inter city and intra city transport as two diff systems. What I suggested was to have inter city mobility underground and intra city transportation over ground..and many cities can be joined and networked like so.
So a sci-fi arcology, but along a line? It would be an interesting thing to give a try- mix it with some hyperloops to give it an even more modern sheen.
Also a 20 minute commute across this thing is ridiculous - that wouldn't just require you to go 317 mph, it would require you to average 317 mph. With everything along the same path, your high speed rail needs to stop at every city node. If the video is accurate to scale, these nodes are only a half mile apart, so you're talking over 200 stops along your journey. Even if you had a train that could teleport instantaneously between stations, that would leave only 6 seconds at each station for embarking/disembarking. With everyone travelling on the same path, you can't skip stops even if you know for a fact no one on your train is getting on or off there without risking a collision with another train that does have a stop.
You could do a little better with parallel rail lines - for example an express line connecting every 15th node and a local line servicing the nearest 15 nodes so in the worst case scenario you only have 30 stops, but that's still only 40 seconds between stops on average and that has to include switching trains. Best case is 5 parallel lines which gets you down to a worst case commute of 15 stops over 159 total miles. Assuming a minute at each stop on average (which note includes 4 different transfers), your train only needs to go 1908 mph (not including accelerating or decelerating) to get you home in 20 minutes (not including any walk to or from stations).
God forbid one of these lines requires maintenance.
If we assume a high speed train that can average 200 mph not including time at stops, your commute time across the city is about an hour. Not bad when you consider that this is the equivalent of traversing every street in a city, but abysmal when you consider that if this city was a solid circle you could get to your destination in half the time on a bike at a leisurely pace.
There are 210 stations spaced a half mile apart each. You are starting out at station 1, you want to get to station 186. The train starts at station 1, then stops at 2, then 3, then 4... eventually you get to station 196 - you've stopped 195 times and travelled 98 miles.
Now consider if you had two lines that split the work. The first line goes to every 15th station, and the second line connects each 15th station with the 14 local stations preceding it. So you get on the first train at station 1, it stops at station 15, station 30, station 45... eventually you get to station 210. You have stopped 14 times and travelled 105 miles. Then you get on the local train, it stops at station 209, 208... until you eventually get to station 196. You have stopped an additional 14 times, and travelled an additional 14 miles, for a total of 28 stops and 119 miles travelled.
If you add more lines, there are fewer stops on any given line, but you potentially have to travel a greater total distance as you repeatedly "overshoot" your target. Once you get to 5 lines, each line is just adding one stop between those of the earlier line - you've essentially reduced the commute to a binary search, and adding any more lines won't get you home in fewer stops.
Since writing my initial comment, I have realized that in reality you could improve on this worst case scenario further. Instead of having a single local line that just connects a single hub to its local stations, you could have two lines connecting two hubs to the local stations between them, so that a traveler would at most travel half the distance between the hubs and never overshoot. This doubles the number of lines, but halves the worst case number of stops and substantially reduces overshoot. Thus you could reach any node in the city within 10 stops and a maximum trip length of 109 miles. Thus with the same parameters you could conceivably commute make the commute in 43 minutes, which is more reasonable though still double what Salman claims, and it would still be faster to bike if the city was circular. These 10 rail lines would run a total of 1050 miles, which at an extremely generous estimate of $300 million per mile of underground rail is about $315 Billion dollars, over 1.5X the maximum expected cost according to the article.
Maybe MIT president Rafael Reif could contribute some talent from the Media Lab for this project ... along with some gifts delivered courtesy of Epstein's fellating chick Neri Oxman!
I'm sure the Crown Prince could use a little more "reputation laundering" from his friends and supporters states-side after the CIA announced that it was the Crown Prince himself who ordered the assassination of dissident journalist Jamal Khashoggi.
Epstein apparently was a good friend of the Saudi Crown Prince and visited him there on the eve of Mr. Trump's election:
15 comments
[ 2.6 ms ] story [ 47.0 ms ] threadI guess that might change some dynamic related to "interior" vs "exterior" properties and potentially make the bullet train slower to account for curved tracks.
Kind of an interesting thought experiment.
Pedantically, I think you probably mean 105 mile/(2*pi) radius circle.
Detroit is rebuilding the city around neighborhoods. The idea is you reinforce neighborhoods around the idea that most of what you need is no longer than a ten minute walk. They've built greenways connecting places to make this possible. Kind of ironic for the city of cars?
I would like to hear the proposed advantages of the design. One advantage that comes to mind is that the entire city could run off of a single transportation corridor. Also a line has the maximum perimeter to surface area. Maybe this is relevant for expansion?
SF is proverbially 7x7 = 49 mi^2.
But now we have to have inter city and intra city transport as two diff systems. What I suggested was to have inter city mobility underground and intra city transportation over ground..and many cities can be joined and networked like so.
Also a 20 minute commute across this thing is ridiculous - that wouldn't just require you to go 317 mph, it would require you to average 317 mph. With everything along the same path, your high speed rail needs to stop at every city node. If the video is accurate to scale, these nodes are only a half mile apart, so you're talking over 200 stops along your journey. Even if you had a train that could teleport instantaneously between stations, that would leave only 6 seconds at each station for embarking/disembarking. With everyone travelling on the same path, you can't skip stops even if you know for a fact no one on your train is getting on or off there without risking a collision with another train that does have a stop.
You could do a little better with parallel rail lines - for example an express line connecting every 15th node and a local line servicing the nearest 15 nodes so in the worst case scenario you only have 30 stops, but that's still only 40 seconds between stops on average and that has to include switching trains. Best case is 5 parallel lines which gets you down to a worst case commute of 15 stops over 159 total miles. Assuming a minute at each stop on average (which note includes 4 different transfers), your train only needs to go 1908 mph (not including accelerating or decelerating) to get you home in 20 minutes (not including any walk to or from stations).
God forbid one of these lines requires maintenance.
If we assume a high speed train that can average 200 mph not including time at stops, your commute time across the city is about an hour. Not bad when you consider that this is the equivalent of traversing every street in a city, but abysmal when you consider that if this city was a solid circle you could get to your destination in half the time on a bike at a leisurely pace.
Sorry but I am having hard time grasping math on this, can you explain a little bit more?
There are 210 stations spaced a half mile apart each. You are starting out at station 1, you want to get to station 186. The train starts at station 1, then stops at 2, then 3, then 4... eventually you get to station 196 - you've stopped 195 times and travelled 98 miles.
Now consider if you had two lines that split the work. The first line goes to every 15th station, and the second line connects each 15th station with the 14 local stations preceding it. So you get on the first train at station 1, it stops at station 15, station 30, station 45... eventually you get to station 210. You have stopped 14 times and travelled 105 miles. Then you get on the local train, it stops at station 209, 208... until you eventually get to station 196. You have stopped an additional 14 times, and travelled an additional 14 miles, for a total of 28 stops and 119 miles travelled.
If you add more lines, there are fewer stops on any given line, but you potentially have to travel a greater total distance as you repeatedly "overshoot" your target. Once you get to 5 lines, each line is just adding one stop between those of the earlier line - you've essentially reduced the commute to a binary search, and adding any more lines won't get you home in fewer stops.
Since writing my initial comment, I have realized that in reality you could improve on this worst case scenario further. Instead of having a single local line that just connects a single hub to its local stations, you could have two lines connecting two hubs to the local stations between them, so that a traveler would at most travel half the distance between the hubs and never overshoot. This doubles the number of lines, but halves the worst case number of stops and substantially reduces overshoot. Thus you could reach any node in the city within 10 stops and a maximum trip length of 109 miles. Thus with the same parameters you could conceivably commute make the commute in 43 minutes, which is more reasonable though still double what Salman claims, and it would still be faster to bike if the city was circular. These 10 rail lines would run a total of 1050 miles, which at an extremely generous estimate of $300 million per mile of underground rail is about $315 Billion dollars, over 1.5X the maximum expected cost according to the article.
I'm sure the Crown Prince could use a little more "reputation laundering" from his friends and supporters states-side after the CIA announced that it was the Crown Prince himself who ordered the assassination of dissident journalist Jamal Khashoggi.
Epstein apparently was a good friend of the Saudi Crown Prince and visited him there on the eve of Mr. Trump's election:
https://www.insider.com/epstein-riyadh-saudi-arabia-private-...