RichardJensen

'commercial grade autonomous vehicle' is code for the following: "It's not going to work very well--if at all--so none of them will be sold to retail customers." Geez. Geez, geez, geez. Just once, before I die, I'd like to see Ford have two consecutive competent CEOs.

What I love about all these pronouncements is that they're all supposed to happen soon, but not so soon that you'll be able to remember the promise when they fail to keep it. Like VW's "Our EV revolution is coming in 2020!!!" And in 2020, when we're all so busy finding out what happened within the last five minutes to the 8000 people we follow on Twitter, we'll have completely forgotten about it. See also: Musk's semi, roadster, and GM's 'dense urban' autonomous fleet.

History tends to repeat itself, and people tend not to learn from history. Possibly, quite possibly, those two things are related.

Heh. Musk has made money too. Lear made money selling planes. Musk made money selling hot air.

An arrogant, fraudulent charlatan?

First, the problem set for flying is so much smaller​ than driving, and 70 years after semi-autonomous flight was demonstrated, fully autonomous flight is not in use. And although equipped with an auto-pilot, the 707 could ​not​ fly itself. An autopilot is basically adaptive cruise control with lane keeping assist. Second, your statement that the technology for autonomous driving "exists" is, frankly, false, in every application except for exceedingly small and exceedingly well-mapped corners of the world.

I didn't know the 707 could fly itself.

And that's the breakdown. You can't come up with a controlled environment that covers millions of square miles, and you can't reliably count on autonomous vehicles to function well outside of their controlled environment.

I've never seen it phrased better.

The key, key, key question, as pointed out in this video, is that of 'enveloping the world'. You absolutely cannot create autonomous vehicles capable of universal problem solving, and as I've said before, electronics have no secondary awareness; they have no idea what they're doing, therefore they cannot evaluate *obvious* problems. So you are, as this video points out, given the task of filling the entire United States, the fourth largest country on the planet by area, with uniform and uniformly maintained markers for autonomous vehicles, as well as perhaps mandatory markers for other vehicles on the road, etc. It's not going to happen, folks.

$200,000,000 PRIVATE PLACEMENT TESLA ANNOUNCES NEW JUNIOR JUNIOR JUNIOR JUNIOR UNSECURED DEBENTURES "FOUNDERS SERIES" 1,000 $200,000 NOTES Payable sometime after 2020, nobody has the slightest idea when, with a new class 8 truck. Maybe. DIVIDENDS None INTEREST None

All of them. It's a de facto Ponzi scheme at this point.

................................................"entire balance due for a reservation".................................................... https://en.wikipedia.org/wiki/Unsecured_creditor If you do this you are basically giving Tesla an interest free unsecured $200,000 loan. You are giving an interest free unsecured loan to the corporate equivalent of that guy who took out a third mortgage and refinanced with an ARM 12 years ago in order to buy a new SUV.

How much money do you think they blew on those two prototypes?

I can just imagine myself back in the 50s having similar arguments with people who believed that flying cars were the future.

I don't think governments 'everywhere' understand the benefits, and even if they did, consider this very simple example: A short lane closure put in place by a private contractor in order to install a driveway cutout. This closure may be in place for less than a full day. How will this be tracked? And if the assumption is that the car will 'detect' the temporary traffic control devices, what will be the fine tuning required to correctly identify the control devices under all conceivable circumstances? You see, the problem with machine learning is that you cannot program a machine to recognize edge conditions. Consciousness is so thoroughly different from machine learning in part because we are aware of what we are doing; a computer cannot be made 'aware' of the context in which it is making decisions, and thus there will always be edge conditions that the computer does not recognize, and who will be responsible for what happens when those edge conditions are encountered?

It's not a question of making the data available. It's a question of keeping the data current.

All of this. Also, Uber isn't supposed to save you money, it's supposed to be like a Keurig machine, in that it appears to be an advance but is in actuality an expensive step sideways.

You're not very old, are you? 3D map data, accurate down to the inch, for all of the US, kept current, is not something that will be available any time soon.

This might be one of the easiest scenarios to manage. The complicated scenarios are the ones that seem 'simple' to us because of how fundamentally different human perception and AI are. Also bad weather. Bad weather is going to be hard for AI just like people (snow covered roads: Where is the center line? Where is the shoulder/curb? What is the maximum safe speed?)

Yep. Bob Lutz pretty much destroyed FoE by betting the farm on RWD BMW fighters and shortchanging the bread & butter Escort/Fiesta. Then, when the Scorpio & Sierra crashed in Europe, he cooked up this idea to send excess production to the states, and, well.... -- As far as Lutz taking *ANY* credit for the Explorer: Lutz left Ford in 1986. The Explorer was launched in 1990.

What you don't know about business would fill a warehouse. If it costs, say, $3B to earn a total of $1.5B in profits before you have to invest another $3B, you have a product that does not self-fund. JLR's products do not self-fund.

Well, the UAW's getting in on it. "We're going to fire hundreds of people all at once 'with cause', so we don't have to offer severance, and so they don't qualify for unemployment, but trust us, we're a wonderful wonderful wonderful company doing wonderful wonderful things for everyone and never ever trying to screw people over because Elon Musk is a genius and a fantastic human being who would never ever do anything shady and who always tells the truth and if he says that these hundreds of people all had to be let go all at once because they all failed their performance reviews all at once, then he is absolutely telling the truth and not disguising layoffs as terminations with cause in order to save himself a lot of money and keep his rickety ponzi scheme disguised as a car company in business."

Just popping in to observe that we are unlikely to remember Hackett's tenure with any fondness. And that's pretty much why I've been AWOL lately; I don't like what Hackett's doing, and rather than fill the forum with endless complaints about it--like some people I could think of--I'm taking a leave of absence until Hackett is replaced by someone who has a clue.

This picture pretty much sums up what I think of the stats that come out of China: