3 ms·

I'm deeply skeptical about the viability of these sort of car sharing networks (autonomously driven or otherwise) for this reason. This doesn't scale when ever

I'm deeply skeptical about the viability of these sort of car sharing networks (autonomously driven or otherwise) for this reason. This doesn't scale when everyone is trying to use the service and the street capacity for cars is fixed.

aub3bhat 10y ago
NO that's not how it works, in fact the street capacity of cars being fixed is the very reason why car sharing networks are awesome, otherwise you are stuck in current situation where each individual has his/her own car leading to traffic and congestion.
- Tiktaalik 10y ago
  There's a decrease in traffic from car sharing from what you state, moving from a single individual in a car to several in a car, however the overall capacity of the road remains fixed. If car sharing networks become popular in comparison to public transit, then the amount of cars will increase and this will eventually come in conflict with the physical limits of the road network.
- lloyd-christmas 10y ago
  > otherwise you are stuck in current situation where each individual has his/her own car leading to traffic and congestion. This is a terrible assumption. The deal being offered is Manhattan -> Manhattan, and only below 125th street. That's a population of ~1m people[0], which doesn't even factor in age demographics. There are 1.6m commuters into manhattan each day and only 225k cars registered in manhattan[1]. Only 30% of employees in manhattan live in manhattan[2]. Basically, the share of people using uberpool will almost exclusively come from pre-existing public transportation. [0] https://dmv.ny.gov/statistic/2015reginforce-web.pdf https://dmv.ny.gov/statistic/2015reginforce-web.pdf [1] https://www.health.ny.gov/statistics/cancer/registry/appendix/neighborhoodpop.htm https://www.health.ny.gov/statistics/cancer/registry/appendi... [2] http://www.census.gov/population/www/cen2000/commuting/index.html http://www.census.gov/population/www/cen2000/commuting/index...
  - aub3bhat 10y ago
    As someone who lives in Manhattan below 125th street, you could not be more wrong. What you have is transition from Cabs with single rider to Uber Pool, not Subway to Uber Pool. Also you are using figures of user who live "outside" Manhattan which do not at all benefit from Uber POOL.
    - lloyd-christmas 10y ago
      > As someone who lives in Manhattan below 125th street, you could not be more wrong. I've lived in NY for 30 years. I've lived below 125th street in manhattan for a decade of that. The wealth of statistics available to defend your point is at your fingertips. And... you went with "nuh-uh". That doesn't really convince me of much. For some anecdata, I've only known 1 person to take a cab in both directions every day. The only reason was because they got a new job. 3 months later their lease was up and they moved to a more convenient spot. I know plenty of people who take cabs several times a week, but that's also not a demographic it will be taking from. > Also you are using figures of user who live "outside" Manhattan which do not at all benefit from Uber POOL. That was my point. I'm pointing out that the majority of the traffic comes from elsewhere or non-commuter traffic, meaning that the impact of this would be insignificant in the big picture.
mason55 10y ago
I think the idea is that if all cars are automated then the street capacity grows. Imagine if all the cars on the road are talking to each other with swarming algorithms. Now you don't need stop lights, you don't have people cutting each other off, you don't have gridlock, you have seamless merging. You can also have the cars intelligently route around congestion to balance flow (similar to what Waze does only much more efficient since all cars on the road could do it). Sure, there's still a limit to what the road can carry, but it would be much higher.
- Tiktaalik 10y ago
  I get this idea but the gains from this sort of "perfect drivers" scenario are pretty fuzzy. Maybe the biggest gains for this concept would be on the highway, but in a city with pedestrians and cyclists, where there certainly does need to be stop lights, and much, much more cautious automated drivers, I'm skeptical the effect would be large.
- ufo 10y ago
  Even if the road capacity increases with self driving cars, I find it hard to believe that its ever going to reach the same densities as other forms of transportation. Pictures like this one[1] come to mind. [1] http://urbanist.typepad.com/.a/6a00d83454714d69e2017d3c37d8ac970c-800wi http://urbanist.typepad.com/.a/6a00d83454714d69e2017d3c37d8a...

aub3bhat 10y ago

NO that's not how it works, in fact the street capacity of cars being fixed is the very reason why car sharing networks are awesome, otherwise you are stuck in current situation where each individual has his/her own car leading to traffic and congestion.

Tiktaalik 10y ago
There's a decrease in traffic from car sharing from what you state, moving from a single individual in a car to several in a car, however the overall capacity of the road remains fixed. If car sharing networks become popular in comparison to public transit, then the amount of cars will increase and this will eventually come in conflict with the physical limits of the road network.
lloyd-christmas 10y ago
> otherwise you are stuck in current situation where each individual has his/her own car leading to traffic and congestion. This is a terrible assumption. The deal being offered is Manhattan -> Manhattan, and only below 125th street. That's a population of ~1m people[0], which doesn't even factor in age demographics. There are 1.6m commuters into manhattan each day and only 225k cars registered in manhattan[1]. Only 30% of employees in manhattan live in manhattan[2]. Basically, the share of people using uberpool will almost exclusively come from pre-existing public transportation. [0] https://dmv.ny.gov/statistic/2015reginforce-web.pdf https://dmv.ny.gov/statistic/2015reginforce-web.pdf [1] https://www.health.ny.gov/statistics/cancer/registry/appendix/neighborhoodpop.htm https://www.health.ny.gov/statistics/cancer/registry/appendi... [2] http://www.census.gov/population/www/cen2000/commuting/index.html http://www.census.gov/population/www/cen2000/commuting/index...
- aub3bhat 10y ago
  As someone who lives in Manhattan below 125th street, you could not be more wrong. What you have is transition from Cabs with single rider to Uber Pool, not Subway to Uber Pool. Also you are using figures of user who live "outside" Manhattan which do not at all benefit from Uber POOL.
  - lloyd-christmas 10y ago
    > As someone who lives in Manhattan below 125th street, you could not be more wrong. I've lived in NY for 30 years. I've lived below 125th street in manhattan for a decade of that. The wealth of statistics available to defend your point is at your fingertips. And... you went with "nuh-uh". That doesn't really convince me of much. For some anecdata, I've only known 1 person to take a cab in both directions every day. The only reason was because they got a new job. 3 months later their lease was up and they moved to a more convenient spot. I know plenty of people who take cabs several times a week, but that's also not a demographic it will be taking from. > Also you are using figures of user who live "outside" Manhattan which do not at all benefit from Uber POOL. That was my point. I'm pointing out that the majority of the traffic comes from elsewhere or non-commuter traffic, meaning that the impact of this would be insignificant in the big picture.

mason55 10y ago

I think the idea is that if all cars are automated then the street capacity grows. Imagine if all the cars on the road are talking to each other with swarming algorithms. Now you don't need stop lights, you don't have people cutting each other off, you don't have gridlock, you have seamless merging. You can also have the cars intelligently route around congestion to balance flow (similar to what Waze does only much more efficient since all cars on the road could do it). Sure, there's still a limit to what the road can carry, but it would be much higher.

Tiktaalik 10y ago
I get this idea but the gains from this sort of "perfect drivers" scenario are pretty fuzzy. Maybe the biggest gains for this concept would be on the highway, but in a city with pedestrians and cyclists, where there certainly does need to be stop lights, and much, much more cautious automated drivers, I'm skeptical the effect would be large.
ufo 10y ago
Even if the road capacity increases with self driving cars, I find it hard to believe that its ever going to reach the same densities as other forms of transportation. Pictures like this one[1] come to mind. [1] http://urbanist.typepad.com/.a/6a00d83454714d69e2017d3c37d8ac970c-800wi http://urbanist.typepad.com/.a/6a00d83454714d69e2017d3c37d8a...