Self-Driving Cars Must Meet 15 Benchmarks in U.S. Guidance • Hacker News Board

Re: Self-Driving Cars Must Meet 15 Benchmarks in U.S. Guidance

#111

Post by sharemywin » Tue, Sep 20, 2016, 5:22 PM UTC

wyager wrote: post #76

gumby wrote: post #6

Earlier quoted context omitted.

> I'm astounded that it seems like these regulations are going to be sensible... Was that hyperbole? I would say the majority of regulations (at least in OECD countries) are sensible, and many that are not are intended to be, are outdated, or are politicized.

A regulation that seems sensible to one cultural group might seem like ridiculous overreach to another cultural group. The US is one of the most polycultural countries in the world, so any regulation is bound to piss a lot of people off. It's better to be judicious with regulation in such situations. Gun control is a great example that seems to confuse a lot of non-Americans. To your average San Franciscan, who has n…

“Democracy must be something more than two wolves and a sheep voting on what to have for dinner.” ― James Bovard, Lost Rights: The Destruction of American Liberty

I think lately far to many people already have the answer before there is any discussion

I think democracy only functions when people are open minded and willing to put themselves in others shoes.

Re: Self-Driving Cars Must Meet 15 Benchmarks in U.S. Guidance

#112

Post by danharaj » Tue, Sep 20, 2016, 5:23 PM UTC

2bitencryption wrote: post #95

owyn wrote: post #79

It wasn't mentioned in the bloomberg article, but the 15 areas covered are: • Data Recording and Sharing • Privacy • System Safety • Vehicle Cybersecurity • Human Machine Interface • Crashworthiness • Consumer Education and Training • Registration and Certi cation • Post-Crash Behavior • Federal, State and Local Laws • Ethical Considerations • Operational Design Domain (operating in rain, etc) • Object and Event Dete…

This seems suspiciously like government getting something right... with regards to a quickly-evolving new technology market... has this ever happened before?

Yes. There's been a decades-long smear campaign against public management of common infrastructure.

Re: Self-Driving Cars Must Meet 15 Benchmarks in U.S. Guidance

#113

Post by bostonpete » Tue, Sep 20, 2016, 5:24 PM UTC

euroclydon wrote: post #41

I'm surprised that self-driving technology is focusing on replacing the driver as an autonomous actor, processing visual and radar/lidar signals in order to know about its surroundings. I've always thought we'd get further faster by having automobiles also talk to other vehicles nearby, and design roads to support the computer driven vehicles. Two examples are: 1) If the vehicle is talking to the cars in front of it,…

As others have mentioned, vehicle-vehicle communication only works if you trust the other vehicles. This sort of thing is almost certainly coming for fleets of trucks though (Google "truck platooning"), where a known set of vehicles can communication with each other.

Re: Self-Driving Cars Must Meet 15 Benchmarks in U.S. Guidance

#114

Post by ghaff » Tue, Sep 20, 2016, 5:28 PM UTC

nommm-nommm wrote: post #51

Filligree wrote: post #12

Earlier quoted context omitted.

Cars move. Splitting the laws by state would mean also splitting the market, and allowing cars on the road which would be illegal to e.g. drive into Texas. Really, it should be international.

I was under the impression that most of the requirements only applied to registering a car in that State not operating the vehicle?? So, for example, NY requires yearly safety inspections and you'll get a ticket if your inspection lapsed. But you don't have to get a safety inspection to drive in NY if your car is registered in a state that doesn't require safety inspections. I could be mostly off base on this one. Th…

>Though some laws are so local sometimes it's impossible for an out of towner to know the local laws like going right on red is, as far as I know, illegal in NYC but legal... Everywhere else? How does someone from Texas supposed to know that?

You've pretty much picked an outlier. And I might be inclined to argue that someone from Texas trying to drive in Manhattan for the first time has other problems :-)

There are a few other things like whether you can pass on the right on an interstate and the aforementioned when headlights need to be on (though I often see this last point signed). But these are usually getting into corner cases and don't really affect how the average person has to approach driving.

Re: Self-Driving Cars Must Meet 15 Benchmarks in U.S. Guidance

#115

Post by andrewpi » Tue, Sep 20, 2016, 5:28 PM UTC

sharemywin wrote: post #91

abakker wrote: post #71

Earlier quoted context omitted.

If it is on my property, then I think brick color is free speech.

Except that you signed a contract when you bought the property.

Search for the "unconstitutional conditions doctrine". Government can't get around constitutional limits via contract requirements.

Re: Self-Driving Cars Must Meet 15 Benchmarks in U.S. Guidance

#116

Post by sharemywin » Tue, Sep 20, 2016, 5:28 PM UTC

shanusmagnus wrote: post #99

angryasian wrote: post #80

Earlier quoted context omitted.

I think the biggest benefit for self-driving cars is that there is really no big corporation or lobbying group that is against this technology. Car manufacturers probably would have put up the biggest battle but they are all pushing for the adoption of the technology. Had car manufacturers felt this technology was a thread to their business model, you'd definitely see a lot of push back and crazy stipulations.

Interesting point -- presumably this means that the existing manufacturers think they can compete in this market, or steer it someplace where they can compete. I wonder if they're right.

There's no "Car neutrality" law so far.

Re: Self-Driving Cars Must Meet 15 Benchmarks in U.S. Guidance

#117

Post by hx87 » Tue, Sep 20, 2016, 5:34 PM UTC

sharemywin wrote: post #98

sharemywin wrote: post #82

Earlier quoted context omitted.

Regulatory Capture will by definition always result in Regulations working in the interest of the major companies(or special interest) in an industry, rather than the public interest. Definition: Regulatory capture is a form of government failure that occurs when a regulatory agency, created to act in the public interest, instead advances the commercial or political concerns of special interest groups that dominate t…

The question is does regulatory capture always happen?

It tends to happen when the following are true:

a) Previous government experience is highly valuable in private sector employees

b) Government pay is less than this value

b) affects regulatory capture in two ways: it allows civil servants the opportunity to get massive raises by going private (and incentivises them to be nice to future employers) and cripples the recruitment of highly talented individuals who are less dependent on industry advice. I don't think attacking a) is feasible in a modern regulatory state, but b) is readily doable if a government is willing to significantly deviate from standard salary scales for high-value industries. For example, SEC salaries would have to be much, much higher than Department of the Interior salaries. AFAIK, Singapore already does this and has very high talent retention rates. Even within the US government, it isn't entirely unprecedented, since an E-3 Navy special forces operator probably makes 8x the salary of an E-3 Army public relations specialist.

Re: Self-Driving Cars Must Meet 15 Benchmarks in U.S. Guidance

#118

Post by wwkeyboard » Tue, Sep 20, 2016, 5:34 PM UTC

2bitencryption wrote: post #95

owyn wrote: post #79

It wasn't mentioned in the bloomberg article, but the 15 areas covered are: • Data Recording and Sharing • Privacy • System Safety • Vehicle Cybersecurity • Human Machine Interface • Crashworthiness • Consumer Education and Training • Registration and Certi cation • Post-Crash Behavior • Federal, State and Local Laws • Ethical Considerations • Operational Design Domain (operating in rain, etc) • Object and Event Dete…

This seems suspiciously like government getting something right... with regards to a quickly-evolving new technology market... has this ever happened before?

https://en.wikipedia.org/wiki/Saturn_V

Re: Self-Driving Cars Must Meet 15 Benchmarks in U.S. Guidance

#119

Post by raverbashing » Tue, Sep 20, 2016, 5:35 PM UTC

jcranmer wrote: post #53

peteretep wrote: post #34

Earlier quoted context omitted.

> I would say the majority of regulations (at least in > OECD countries) are sensible I think it can be shocking to non-Americans just how much the Americans distrust and think their lawmakers and -- especially shockingly, their civil servants -- are both incompetent and have malicious intent. American friends have found it incredible -- for example -- that something like NICE[0] can exist and people don't assume it'…

> I think it can be shocking to non-Americans just how much the Americans distrust and think their lawmakers and -- especially shockingly, their civil servants -- are both incompetent and have malicious intent. The EU is often criticized (e.g., Brexit) as being something that promulgates useless regulations (e.g., curvature of a banana).

The "curvature of a banana" thing is a myth, the regulation just says it should have a good appearance

Re: Self-Driving Cars Must Meet 15 Benchmarks in U.S. Guidance

#120

Post by ghaff » Tue, Sep 20, 2016, 5:42 PM UTC

krisoft wrote: post #66

ams6110 wrote: post #29

Earlier quoted context omitted.

Disagree with almost all of this. Liability should be with the owner of the car, like it mostly is today. They are the ones buying it and operating it on the public roads. They should have insurance to cover this liability.

I'm worried about the following: Hapless Joe buys a nice, well reviewed, good looking and performant self-driving car. He enjoys it very much for a year, goes for all the required maintenance checks. Then suddenly as he is taking a nap on his usual daily commute the car swerves onto the sidewalk and runs over my wife. Turns out there is a very rare race condition in the path finding algorithm, which makes it do that…

In your hypothetical, I have a hard time giving an example from current consumer products where it isn't the manufacturer's fault. Approved device, used as directed, properly maintained causes a fatality. With what consumer device today wouldn't it be the fault of the device maker/seller.

At the same time, I'm hard put to think of products/services used by consumers--outside of medicine perhaps--where it's the norm that sometimes "shit happens" because statistics and it's not considered reasonable to sue the manufacturer (though we often still do).

Interesting times. Especially given that the facts (and the degree of automation) won't typically be nearly as clear cut especially over the next few decades.