[finstell]

Everybody is rushing to develop their own self-driving car, then they change their minds in various ways, then team members change who they work for, some of them quit and start their own company, some fail, some got acquihired. They give deadlines and cannot meet them. Everyone is dying to be first in the market or something like it. Feels there is so much greed. I don't get it, what's the big fuss, why must everyone hurry? Can't they be just calm and progress in a sensible manner. I don't expect this thing to turn out to be winner takes it all. Automative industry is much like a close knit society where they depend on similar resources. It's not exactly the same as posh start up culture.

[emdowling]

Simple: everyone is looking for the next big platform. The smartphone revolution came by surprise to most and it has taken years for companies to catch up. Everyone is racing because they don't want to be caught with their pants down again.

[shostack]

The size and value of cars as a platform can't be emphasised enough.

It goes way beyond hardware. Whoever controls the car as a platform in theory controls the services layer that would be built on top of it (ride sharing, delivery etc.) and also the captive attention of the cars' occupants outside their phone. Then there is the data that comes from a device that takes you wherever you want to go, with tons of internal and external sensors to get data on occupants (unique identification will be trivial for audio ads, etc) and their external environment.

My guess is Google sees the rather low margins of cars and would much rather focus on what is likely substantially higher margins (and far less capital outlay) on the software side which is much more closely aligned their core strength.

[bzbarsky]

And thus we continue to build the panopticon around ourselves. It's a 20th century dictator's wet dream, but self-imposed.

[dx034]

Exactly that. If one company (especially one with high production capacity) would now release a fully functional self driving car, they could dominate the car market for the next decade. Orders for trucks, delivery vehicles, uber etc would alone max out production for several years. If done by a major player (e.g. Toyota, GM, VW), they could probably get legislation in some states changed quickly if they can proof that their system is safe.

Catching up isn't easy, you won't be able to just reverse engineer it.

[__derek__]

What you're suggesting is that the first self-driving car manufacturer (as long as its cars are rolling off the lines of one of the major players) will dominate (parts of) the fleet sales market. That's not insignificant, but it's not the car market. I think you're underestimating how resilient the existing driving paradigm will be among consumers (i.e. the people buying personal cars and driving fleet vehicles).

> Catching up isn't easy, you won't be able to just reverse engineer it.

No, but the California-based employees with an understanding of that system will be able to get a nice payday when they start working for a competitor.

[netsharc]

So if Google has this tech, isn't this move basically saying "Hey boys, I've got this sexy stuff. Who's going to be the highest bidder?".

Is there still a no politics policy on HN?

-- POLITICS BELOW-- Also it'd be interesting to see in the Age of Trump if Google would be red-taped if tries to sell the tech overseas, Trump would probably solicit bribes from the Detroit people to put up such a law.

[nradov]

No, all of the major auto manufacturers have pared down their spare capacity (and dealership networks) to save costs. So even if Toyota, GM, or VW invented an affordable fully functional self driving car tomorrow they wouldn't be able to dominate the market. It would take years to build up additional factory capacity, giving competitors a chance to catch up.

[blahfuk]

The first person to succeed will usher in a new transportation revolution and become insanely rich.

But I think a certain amount of hubris is what's causing all these companies to take a swing and then back out once they realize what they're up against. It's not an easy nut to crack. The software in itself is incredibly complicated, not to mention the politics involved, and all the special interests you have to win over before you can even pass go. But then the fact that you need to attach it to a car means you also have to compete in one of the largest and most competitive industries that ever existed.

[jahnu]

Big risk/reward situation. Althoug the reward is possibly gigantic. It's not even clear that it's possible as described yet, imho. i.e. Fully autonomous vehicles. It's somewhat likely to be a winner takes all situation in terms of market share, patent ownership. I'm exaggerating slightly but I can't see there being more than a couple of big winners in this space.

[ma2rten]

Google has been working on this for over a decade, I think it's fair to say that they are taking it slow.