Tesla isn't a car company. It's building Rome. All great "startups" are not some good idea executed well. They are companies with a long term vision that generate ideas to execute that will get them there. Anyone might steal an idea or copy a product, but no one can steal a mission or a destination far in the future. Apple, Microsoft, Facebook, Yahoo, Google... all started as or at some point became "build Rome" companies.
Master Plan, Part Deux
291–300 of 705 posts
Re: Master Plan, Part Deux
#292> "When used correctly, [partially autonomous driving] is already significantly safer than a person driving by themselves" If you're an American, you're twice as likely to die with a steering wheel in your hands as you are to die at the hands of a murderer. Human-driven vehicle deaths cause grave second-order suffering for families and friends - and hurt the economy. A shift to technologies safer than human-driven ca…
I don't understand, why would you need self driving cars if you don't want to drive? Public transport is here, is cheaper and safer. Also it allows to multitask.
I'm a huge fan of public transit. It is of limited utility in land-use patterns typical of modern American cities, or, far more so, suburbs.
If you can take a vehicle directly to your destination, or a short walk from it, things works pretty well. If you need to transfer, even once, you can easily double trip time (consider a typical commute is ~20 minutes one-way).
Scheduling transfers with shared-roadbed (e.g., bus, light rail) systems is all but impossible.
Prior to the creation of early transit systems (horse-drawn omnibuses, either wheeled or tracked), few cities had a cross-section of more than 1-2 miles, as 1 miles is about a 20 minute walk. Streetcars (horse, electric, steam) extended that based on their average speed, typically 5-15 mph, with a 5-10 mile commute becoming possible though still usually less, and those commutes forming along rail lines, hence a hub-and-spoke city layout (see for example Chicago). Dense grids such as New York's were unusual.
Personal (or small-ridership) vehicles can go directly point-to-point, and make longer, flexible-route commutes possible (I'm not saying this is a good thing, merely a fact), though requiring parking, chauffers/cabbies, automatic vehicle control, etc.). In an interesting parallel, one reason there were so many horses in late-19th-century America was because of railroads. You had 30-60 MPH transit along fixed routes, but limited movement within cities or towns. So horses, wagons, and coaches to get to and from railway stations for goods and people. Petrol-powered automobiles replaced horses before they substituted for trains, though railroad use was already in decline by 1920 (rising one last time due to fuel rationing in WWII).
Upshot: there's a place for public transit, but almost certainly only with massively revamped land-use within urban and suburban regions.
Re: Master Plan, Part Deux
#293"A first principles physics analysis of automotive production suggests that somewhere between a 5 to 10 fold improvement is achievable by version 3 on a roughly 2 year iteration cycle. The first Model 3 factory machine should be thought of as version 0.5, with version 1.0 probably in 2018." What that really means: Tesla is going to lose a ton of money per car on the Model 3, or raise the price, until at least 2022. T…
I think some of this might be offset by the fact that this plan doesn't seem to focus on just cars. A solar product described in the first step of the plan could help fund the cars.
Solar offers even longer time to payoff/profit. It's a long term bet... even longer than Tesla's current profile IMHO.
Re: Master Plan, Part Deux
#294Earlier quoted context omitted.
There are 30,000 motor vehicle deaths in America each year. In perhaps 30 years, fully autonomous vehicles could cut all of that needless carnage to nearly nothing. And what about the hundreds of thousands of injuries we'll be spared? Worldwide, fully-autonomous vehicles will be a public health victory on the scale of vaccines or antibiotics.
> Worldwide, fully-autonomous vehicles will be a public health victory on the scale of vaccines or antibiotics. While autonomy is great and will definitely save many lives, it is absolutely not on the scale of vaccines or antibiotics. Not even close. You're comparing the deaths of millions to thousands.
https://www.google.com/search?q=how+many+people+die+in+car+c...
1.3 million people die every year. This very well may be on that scale.
Re: Master Plan, Part Deux
#295Earlier quoted context omitted.
That clicked when he talked about Tesla Semi. A roof installation on the trailer would make sense and might generate a 'useful' amount of power.
That's a great point. I'd imagine they need 3-4 times the power of a car but have 100 times the surface area. It seems like that could actually work.
I think that's still two OOM too little.
Re: Master Plan, Part Deux
#296> "When used correctly, [partially autonomous driving] is already significantly safer than a person driving by themselves" If you're an American, you're twice as likely to die with a steering wheel in your hands as you are to die at the hands of a murderer. Human-driven vehicle deaths cause grave second-order suffering for families and friends - and hurt the economy. A shift to technologies safer than human-driven ca…
> Besides direct costs for the car, fuel, and maintenance, the main disincentive to driving is how damn boring it is. What happens when we turn fully-autonomous vehicles into luxury entertainment centers? I suspect that, if we're not smart about this shift, we could see wild sprawl on a scale that would dwarf the mid-20th century sprawl we saw in Los Angeles and elsewhere. I was just thinking about this today (and ha…
I know what you mean about SUVs and trucks feeling safer and more comfortable. Objectively however, their handling characteristics and weight distribution (particularly the very light rear weight in pickups) make them quite a bit less safe in terms of what they will do in a high speed evasive action situation.
Re: Master Plan, Part Deux
#297To anyone who's paid close attention to Tesla and to Elon's various offhand remarks to the press and on Twitter, this was all easy to see coming, every bit of it. But now that he's confirmed it all officially: this NUTS . This is so awesome. The press is going to go crazy with this. I wonder what will happen to Uber...seems like it will be hard for them to compete with the rates of cars that don't have to pay their d…
Do you think they'll have solar panels on the trailer? Would they have to roll them on/off when they switch trailers?
Not very soon, there are better places for Solar Panels than roads...
Re: Master Plan, Part Deux
#298Earlier quoted context omitted.
Also millions of miles of fleet learning a day, which Uber doesn't have, since they don't own the cars. Self-driving cars is a supervised learning problem, and he who has the data wins
Data isn't that hard to gather. Uber has more robotics scientists than all the others combined.
Funny when it started as a GPS tracking webapp.
Re: Master Plan, Part Deux
#299Earlier quoted context omitted.
What does that data give you? Raw sensor data + steering angle + acceleration/braking is not particularly useful in of itself. Nobody is trying to shove large amounts of data through a blackbox model and directly trying to predict car controls from images. This is not MobileEye's approach and almost certainly isn't Tesla's approach. The bottleneck is data that is decomposed and labeled in greater detail (lane marking…
Geohot is pretty much claiming to do that! http://www.theverge.com/2016/6/6/11866868/comma-ai-george-ho...
Sure, some people are exploring the approach. This does not mean it works.
Re: Master Plan, Part Deux
#300To anyone who's paid close attention to Tesla and to Elon's various offhand remarks to the press and on Twitter, this was all easy to see coming, every bit of it. But now that he's confirmed it all officially: this NUTS . This is so awesome. The press is going to go crazy with this. I wonder what will happen to Uber...seems like it will be hard for them to compete with the rates of cars that don't have to pay their d…
Do you think they'll have solar panels on the trailer? Would they have to roll them on/off when they switch trailers?
Almost certainly not. Except for, say, powering auxiliary electronics, concentrations of solar power are far too low for on-vehicle collection.
Put the panels alongside the road. Not on the vehicles. Not on solar freaking roadways, which are the stupidest idea since stupid. But in medians, on extant construction, and in concentrated high-insolation solar farms (deserts, etc.).
We're going to need one hell of a lot of panels.