4 ms·

I'm not saying there was conspiracy but there was complacency. The reasoning "batteries are expensive" is circular. As Musk said, technology doesn't advance a

by kjksf 9y ago

I'm not saying there was conspiracy but there was complacency. The reasoning "batteries are expensive" is circular. As Musk said, technology doesn't advance as a function of time but as a function of R&D investment. See atomic bomb or space race for the most dramatic example of how throwing money at the problem compressed the timeline. Car batteries did not get cheaper because 2 years passed but because Tesla and Panasonic invested 5 billion dollars in a giant batter factory. And by creating a demand for lots of batteries, there's incentive for many others to invest in researching better batteries. And before that, cellphones and laptops were driving investment in battery research. I see no reason why all this couldn't happen 10 years ago if car makers invested a sliver of their profits into electric car R&D instead of, you know, putting it all in Diesel and then writing software to cheat emission numbers. And as a corollary, if Tesla didn't push the whole field forward with giant investments, BMW would be "objectively" saying in 2030 that batteries are too expensive to make electric cars viable.

aeturnum 9y ago
I think all of these things are relative. Imagine if Tesla had started in 1983 instead of 2003. We wouldn't have started mass producing electric cars in 2000 - but we'd probably get there earlier than 2020. I don't think the question of how much to invest and where to focus R&D is easy or straightforward. A lot of technological development is the story of unexpected jumps from one area to another. Imagine an alternative reality where fast wireless communication was omnipresent and laptops developed as dumb terminals which used little power. Battery technology might be where it was 40 years ago. It's also possible that, in 50 years, people will look back at the Tesla of today as stupid and foolhardy - striving to make electric cars with stone age battery tech, when a revolution in 2025 changes everything.
- fragmede 9y ago
  Even that's idealized - it's not likely a production ready revolution in battery tech simply appears out of thin air. (In fact, there have been a number of revolutions that haven't managed to scale up to production quantity.)
philwelch 9y ago
> As Musk said, technology doesn't advance as a function of time but as a function of R&D investment. Sometimes. Technology is very path-dependent; to invent z, you need to have invented w, x, and y first. And even with infinite resources that takes time. We’ve been working on fusion for half a century and it might be ready in another half century. Will more investment help? Yeah, but tokamaks take time to build, and you don’t know how to build the next tokamak until after you’ve done all your experiments with the current one. You can’t have iPhones in 1986 no matter how much money you throw at it because the underlying processor and battery tech isn’t there.
- itp 9y ago
  Without necessarily (or completely) disagreeing, I'm curious if you can further justify your assertion of the strong path-dependency. I am personally amazed when I contemplate the Apollo program or the Manhattan Project, and specifically the number of technologies that did not exist at the time. It strikes me that there's a clear observational bias here. Particularly in the modern era, we really only have a single data point around the development of any given technology (given how trivial communication and coordination are). Are there any examples from history of the same technology being developed independently with different dependencies? Possibly relevant, given your choice of fusion as an example, is the somewhat famous graph from the 1976 U.S. Energy Research and Development Administration report on nuclear fusion[0]. We're obviously looking at a number of projections and can't know whether they were right or wrong, but it is noteworthy that we're not even pursuing a plan we believe will ever lead to nuclear fusion ("fusion never"). [0]: https://commons.wikimedia.org/wiki/File:U.S._historical_fusion_budget_vs._1976_ERDA_plan.png https://commons.wikimedia.org/wiki/File:U.S._historical_fusi...
  - davewritescode 9y ago
    You have to remember there’s a huge gulf between building something once and mass producing something, putting it in market and having it produce profits. Electricly powered vehicles existed before the American civil war, it’s taken until now for them to have any hope of success in the market.
maxxxxx 9y ago
"technology doesn't advance as a function of time but as a function of R&D investment" Car companies are not in the business of advancing technology but making money. They will advance technology if they can sell more cars. Now with Tesla they have to respond and are slowly doing so. Musk's motivation is technology for technology's sake. That's a totally different motivation than traditional car companies and that's the reason why Tesla is ahead in technology. It remains to be seen who ends up ahead in terms of business success.
zjaffee 9y ago
I'd argue that there are some conditional factors here and they largely revolve around human capital. In fields with a very small number of people with prior experience, throwing money at the problem won't necessarily accelerate the path as fast as one where there are a large number of qualified people to tackle the problem. This said, continuing to throw money at a problem such that it leads to more and more people developing a skill set in the space would overtime help, but then again, they could reduce the amount of investment down the line, and you'd still have people who were highly specialized in said domain to target that problem anyways (abet being paid less).
rohit2412 9y ago
Except batteries have become cheaper because of consumer electronics, not Tesla. https://assets.bwbx.io/images/users/iqjWHBFdfxIU/imonS5R0.ZBo/v0/1000x-1.png https://assets.bwbx.io/images/users/iqjWHBFdfxIU/imonS5R0.ZB... If what you said was true, only Panasonic would have cheap batteries. Not lg and Samsung.

by kjksf 9y ago

aeturnum 9y ago

I think all of these things are relative. Imagine if Tesla had started in 1983 instead of 2003. We wouldn't have started mass producing electric cars in 2000 - but we'd probably get there earlier than 2020. I don't think the question of how much to invest and where to focus R&D is easy or straightforward. A lot of technological development is the story of unexpected jumps from one area to another. Imagine an alternative reality where fast wireless communication was omnipresent and laptops developed as dumb terminals which used little power. Battery technology might be where it was 40 years ago. It's also possible that, in 50 years, people will look back at the Tesla of today as stupid and foolhardy - striving to make electric cars with stone age battery tech, when a revolution in 2025 changes everything.

fragmede 9y ago
Even that's idealized - it's not likely a production ready revolution in battery tech simply appears out of thin air. (In fact, there have been a number of revolutions that haven't managed to scale up to production quantity.)

philwelch 9y ago

> As Musk said, technology doesn't advance as a function of time but as a function of R&D investment. Sometimes. Technology is very path-dependent; to invent z, you need to have invented w, x, and y first. And even with infinite resources that takes time. We’ve been working on fusion for half a century and it might be ready in another half century. Will more investment help? Yeah, but tokamaks take time to build, and you don’t know how to build the next tokamak until after you’ve done all your experiments with the current one. You can’t have iPhones in 1986 no matter how much money you throw at it because the underlying processor and battery tech isn’t there.

itp 9y ago
Without necessarily (or completely) disagreeing, I'm curious if you can further justify your assertion of the strong path-dependency. I am personally amazed when I contemplate the Apollo program or the Manhattan Project, and specifically the number of technologies that did not exist at the time. It strikes me that there's a clear observational bias here. Particularly in the modern era, we really only have a single data point around the development of any given technology (given how trivial communication and coordination are). Are there any examples from history of the same technology being developed independently with different dependencies? Possibly relevant, given your choice of fusion as an example, is the somewhat famous graph from the 1976 U.S. Energy Research and Development Administration report on nuclear fusion[0]. We're obviously looking at a number of projections and can't know whether they were right or wrong, but it is noteworthy that we're not even pursuing a plan we believe will ever lead to nuclear fusion ("fusion never"). [0]: https://commons.wikimedia.org/wiki/File:U.S._historical_fusion_budget_vs._1976_ERDA_plan.png https://commons.wikimedia.org/wiki/File:U.S._historical_fusi...
- davewritescode 9y ago
  You have to remember there’s a huge gulf between building something once and mass producing something, putting it in market and having it produce profits. Electricly powered vehicles existed before the American civil war, it’s taken until now for them to have any hope of success in the market.

maxxxxx 9y ago

"technology doesn't advance as a function of time but as a function of R&D investment" Car companies are not in the business of advancing technology but making money. They will advance technology if they can sell more cars. Now with Tesla they have to respond and are slowly doing so. Musk's motivation is technology for technology's sake. That's a totally different motivation than traditional car companies and that's the reason why Tesla is ahead in technology. It remains to be seen who ends up ahead in terms of business success.

zjaffee 9y ago

I'd argue that there are some conditional factors here and they largely revolve around human capital. In fields with a very small number of people with prior experience, throwing money at the problem won't necessarily accelerate the path as fast as one where there are a large number of qualified people to tackle the problem. This said, continuing to throw money at a problem such that it leads to more and more people developing a skill set in the space would overtime help, but then again, they could reduce the amount of investment down the line, and you'd still have people who were highly specialized in said domain to target that problem anyways (abet being paid less).

rohit2412 9y ago

Except batteries have become cheaper because of consumer electronics, not Tesla. https://assets.bwbx.io/images/users/iqjWHBFdfxIU/imonS5R0.ZBo/v0/1000x-1.png https://assets.bwbx.io/images/users/iqjWHBFdfxIU/imonS5R0.ZB... If what you said was true, only Panasonic would have cheap batteries. Not lg and Samsung.