Hacker News Reader

Hacker News

Comment by FireBeyond | Hacker News Reader

Parent Full thread

FireBeyond

Other than the "The car will be able to self-drive, coast to coast, this year" back in Jan/Feb, you mean?

View on HN

>back in Jan/Feb

May 2017

https://twitter.com/elonmusk/status/866482406160609280

It’s not unreasonable to heavily critique the fact that Tesla deadlines are blown repeatedly, and yet who among us has gone from 0 to 300k+ vehicles manufacturer and sold per year in 12 years and delivered an EV into an elliptical orbit with a privately developed heavy lift vehicle?

Don’t say something is impossible while someone else is delivering, slowly and missed deadlines or not.

@sbov (deleted reply while I was replying)

> So are you saying Elon's statement was right? Or are you just saying "its hard, give him some slack for making statements he knows are complete bullshit"?

A pessimist’s “bullshit” is an optimist’s missed deadline and inaccurate forecasting.

Aren't those solved problems though? Like, you're talking manufacturing and scaling up production, and there are other examples of heavy lift vehicles. While they are absolutely great accomplishments, it's tweaks on previous formulas.

There are no fully autonomous cars, he'd have to break entirely new ground for that.

> Aren't those solved problems though? Like, you're talking manufacturing and scaling up production, and there are other examples of heavy lift vehicles. While they are absolutely great accomplishments, it's tweaks on previous formulas.

Not at all. I'm unaware of any other heavy lift vehicles developed by a non-government entity that fly for $90 million (versus Delta Heavy's $350 million cost) and are completely reusable besides the second stage and the fairings. I'm unaware of any auto company created in the last century in the US, besides Tesla (and the ghost of GM shepherded through bankruptcy by the Obama administration), that is still in business [1].

So to answer your question, these are "solved" in the same way if you proposed to yourself, "Hey, I can build an iPhone because Apple can build and iPhone". Could you? Absolutely. Are you discounting the enormous amount of work and capital necessary to pull it off? Absolutely.

[1] https://en.wikipedia.org/wiki/List_of_defunct_automobile_man... (List of defunct automobile manufacturers of the United States)

There seems to be more than a few auto companies made in the last century[1]. I'm not saying he's not making these things better, but they existed previous to his invention - which bits of the lifter are completely novel? Complete automation would be entirely novel - it's not just a matter of finding a cheaper/better way to do it.

[1] https://en.wikipedia.org/wiki/List_of_automobile_manufacture...

Novelty doesn't define success. No car company has previously existed that is scaling towards 500k EVs delivered per year, with the capability to manufacture 135GW of energy storage per year. The Horndale Power Reserve in Australia has not only saved Australian energy consumers tens of millions of dollars, but proved that utility scale energy storage could be rapidly deployed and is cost effective.

If these problems are easy, why is Tesla the first to seek solutions to them? Why did Tesla have to struggle to prove the validity of EVs to the public while European car makers committed fraud with diesel emissions? If you're an established car maker, how bad at manufacturing, supply chain management, and product development do you have to be that you can't compete against Tesla with EVs? Questions are rhetorical.

Again, not saying the problems are easy, and that Tesla hasn't had success, I'm just saying in the case of the self driving car, novelty WILL define success. I'm saying this is a different category of problem for Tesla to solve compared to others.

Perhaps, although I think full self driving will be how my kids develop. They crawl, they walk, they observe, they mimic, they vocalize, they can associate faces with people. They've become inquisitive. They reason. They update their world model with input. Slowly, but eventually, become a full blown adult. But there is no switch that causes it to happen overnight, and I think full self driving will be the same. You will see a feature here (lane keeping), a feature there (red light detection), an improved feature (automatic emergency braking reliably using the camera and structure from motion to assist front facing radar in discriminating potential obstacles), and all of a sudden you find yourself in a full self driving car (just as one day I'll look at my kids and think, "you're people now!").

fairings will be reused from this FH launch, so the only part not being reused is the second stage.

Who cares that the "promised" deadline for Falcon Heavy and rapid reusability was also blown several times. This result is something completely unimaginable couple of years ago and nobody is currently able to match it. It will be similar with Tesla, only much more people feel entitled to talk about cars than rockets, so they forget the bigger picture.

Your defense against Tesla’s missed deadlines is that its CEO is busy doing things with his other companies? The Obama administration also “shepherded” the massive EV subsidies that Tesla baked into its pricing.

All automakers receive the EV subsidies you mention, not just Tesla. If you want the subsidy, make EVs! If you don't, buy ZEV credits from those who do (many thanks California and the EU) when you don't meet your fleet emissions targets. Will the US government support Tesla if they fail in the same way GM did? Hardly, because the GM bailout was a jobs program.

My defense is hard problems have deadlines that will be blown passed, and it's not uncommon unless the problem is incredibly well scoped or you're just lazy and set the bar low by design (ie most automakers).

This is not to say Tesla and Musk don't have problems; they do, but credit should be provided where due.

Musk doesn’t run SpaceX. Gwynne Shotwell does.

Musk funded its startup, and was the driving force in building their own engines after getting laughed out of Russia [1]. Shotwell does great managing the day to day, and deserves credit for Falcon Heavy, as she was the one who told Musk they should still test fly the first one and offer it as a service (instead of going straight to BFR).

[1] https://www.inverse.com/article/34976-spacex-ceo-elon-musk-t... (When SpaceX Tried to Buy Missiles From Russia: Vodka and a Run-Around)

> The third and final meeting happened back in Russia. Musk flew there with Cantrell, prepared to purchase three ICBMs for $21 million. But to Musk’s disappointment, the Russians now claimed that they wanted $21 million for each rocket, and then taunted the future SpaceX founder. As Cantrell recounted to Esquire:

> “They said, ‘Oh, little boy, you don’t have the money?”

> This insulting event, however, played a part in inspiring Musk to found SpaceX, which in 2017 alone has successfully launched nine rockets into space and has twelve more launches on the docket this year. On the flight back, Musk turned to Cantrell and said:

> “I think we can build a rocket ourselves.”

Yeah and when Musk tweets, Gwynne needs to emergency hop on an intercontinental flight to calm down the customer.

Disclaimer: fan of spacex

I think it's part of the strategy. These deadlines are eventually blown, but even then what we get in the end is pretty remarkable. I guess the hyper ambitious attitude actually attracts people who then make it work. Plus the obvious media buzz.

If you get enough other-people's-money, is it really that hard to do? How much have the cumulatively burned now?

It’s not unreasonable to heavily critique the fact that Tesla deadlines are blown repeatedly, and yet who among us has gone from 0 to 300k+ vehicles manufacturer and sold per year in 12 years?

Well, for starters the Korean automakers managed to go from 0 to several hundred thousand vehicles in under 5 years. Kia went from 26 cars to 95,000 in a year...

A pessimist’s “bullshit” is an optimist’s missed deadline and inaccurate forecasting

An occasional missed deadline is one thing. But missing every self-proclaimed deadline? That's not just inaccurate forecasting--at some point that crosses the line into fraud.

that's not the same thing as "fully autonomous"

What's the nuance then?

You can get from one coast to the other entirely on a single interstate highway (e.g. I-80), and do it in dry weather and clear conditions. It's the easiest possible thing for an autonomous car to do and is basically just adaptive cruise control.

The hard part is making it work with crazy cyclists on city streets and blizzard conditions and defending against malicious adversaries who purposely try to confuse the car into crashing into something.

https://www.theverge.com/2016/10/19/13341100/tesla-self-driv...

> Musk also promised a demonstration of a fully autonomous drive from Los Angeles to New York by the end of 2017

As in "without the need for a single touch from the driver". Doesn't really matter, AFAIK they are nowhere close.

Reply on news.ycombinator.com