4 ms·
While other cities and regions may poach auto manufacturing jobs and plants from Detroit, the idea that any significant amount of manufacturing will be done in
by armenarmen 12y ago
While other cities and regions may poach auto manufacturing jobs and plants from Detroit, the idea that any significant amount of manufacturing will be done in the SV is unlikely, based on real estate prices alone. Not to say that a large part of research and innovation cannot be done here rather than in Detroit.
- Zigurd 12y agoIf the Silicon Valley mindset wins, cars will be made by contract manufacturers, not by vertically integrated manufacturers. As others have pointed out, there is a lot of expertise in Detroit. That expertise will have to go obsolete. Self-driving "cars" could mark their obsolesce. A self-driving "car" is going to be nothing like current cars, and 100 new business models will bloom alongside their new forms.
- sliverstorm 12y agoWhy would it necessarily be contracted? All the big players in SV are currently hell-bent on vertical integration. SV is not immune to it.
- jonmrodriguez 12y agoIn the past 20-30 years, automotive engineers have invented a lot of ways to make cars much safer (e.g. crumple zones). It sounds like you're saying the designers of self-driving cars will throw all of that expertise out the window -- I assure you that they will not.
- rstupek 12y agoNothing like current cars. Except they'll have wheels, suspensions, door, bodies, propulsion systems, etc. Don't believe that a self driving car changes much beyond who is driving the car
- Zigurd 12y agoSo do buses. Busses differ from cars by all these components being significantly different. Self-driving cars will be at least as different. > Don't believe that a self driving car changes much beyond who is driving the car Except everything. Current cars have one occupant a high percentage of the time. That occupant drives the car. Change that and every design parameter changes.
- potatolicious 12y ago> "Current cars have one occupant a high percentage of the time. That occupant drives the car. Change that and every design parameter changes." I doubt this. The American car (distinct from cars in general) is not only a utilitarian tool but a cultural artifact. For huge portions of the country it is the manifestation of freedom - especially now that we've demolished all of our urban areas in favor of freeways and strip malls. Look at American culture and you'll see that personal car ownership is intrinsically associated with freedom, coming-of-age, and a huge host of cultural values that Americans take very seriously. If you think making driverless cars is going to suddenly tilt us away from single-occupant SUVs and into little self-driving pods (Minority Report style), IMO you're sorely mistaken. At best we're going to get SUVs without a steering wheel. Cars will not get substantially smaller without regulatory intervention, people will continue to own personal automobiles, and they will still have more seats than are needed 99% of the time. Driverless cars will be great, but I think you're severely overestimating their impact on car design/ownership/usage patterns. I know it's popular to imagine a society where people stopped owning cars, where cars are always right-sized for the situation, where cars show up on demand and parking lots made obsolete, etc etc, but that's a 1950s Popular Mechanics style of sci-fi pipe dream.
- AnthonyMouse 12y ago> I know it's popular to imagine a society where people stopped owning cars, where cars are always right-sized for the situation, where cars show up on demand and parking lots made obsolete, etc etc, but that's a 1950s Popular Mechanics style of sci-fi pipe dream. It does seem kind of plausible. Right now you need a car for everything. To get to work, to buy groceries or anything else from a store, to get to any kind of recreation, etc. Now suppose there exists an autonomous electric taxi that you can hail from your phone and use to get from one end of the city to another for around $1. You no longer need a car for anything but getting to work. Rush hour is the peak demand so the price of a ride during rush hour would be prohibitive for commuting. But the same reason the car operators can't justify buying more cars just for rush hour is the same reason commuters won't want to do it. Once commuting is the only reason you own a car you start looking for housing and employment near functioning mass transit. Which is going to bring people back into the cities. And once you do that and people give up owning a car the rest of it follows.
- greglindahl 12y agoAn amusing prediction, given that Tesla and SpaceX are both vertically-integrated manufacturers.
- simulate 12y agoTo wit, Tesla selected Nevada as the site for its first $5 billion battery factory. Alternative sites included Sacramento. http://www.latimes.com/business/autos/la-fi-hy-tesla-nevada-gigafactory-20140903-story.html http://www.latimes.com/business/autos/la-fi-hy-tesla-nevada-... It's certainly not necessary to have R&D and manufacturing in the same location.
- msandford 12y agoOn the surface I agree with you, but it's really dumb to keep design & production separated by any distance. Tesla can site battery production away from the car plant because for them batteries aren't the car. They're a part of the car but having a battery plant isn't a core competency. Musk is building one because that's the fastest way to bring prices down. Not because they have to have it in order to continue churning out the Model S. http://qz.com/32575/ge-a-pioneer-of-outsourcing-is-bringing-american-manufacturing-back-to-life/ http://qz.com/32575/ge-a-pioneer-of-outsourcing-is-bringing-...
- bdcravens 12y agoAll of Google's data centers aren't in Mountain View. Wherever said manufacturing takes place, I'm pretty sure it'll be in a location where union labor is nonexistent. (Consider that a large number of Toyota's trucks are produced in San Antonio)