Comment by ckastner | Hacker News Reader

Comment by ckastner | Hacker News Reader

Parent Full thread

> His hands were not detected on the steering wheel for the final six seconds prior to the crash.

> Tesla has said that Huang received warnings to put his hands on the wheel, but according to the NTSB, these warnings came more than 15 minutes before the crash.

> Tesla has emphasized that a damaged crash attenuator had contributed to the severity of the crash.

These may or may not have been factors contributing to the death of the driver, and ultimately may or may not absolve Tesla from a legal liability.

However, the key point here is that without question, the autopilot failed.

It is understandable why Tesla is focusing on the liability issue. This is something that they can dispute. The fact that the autopilot failed is undisputable, and it is unsurprising that Tesla is trying to steer the conversation away from that.

The discussion shouldn't be either the driver is at fault or Tesla screwed up, but two separate discussions: whether the driver is at fault, and how Tesla screwed up.

The deficiency of Tesla AP has been abundantly clear to anyone with eyes, for a long time. Only fingers-in-ears Musk fans cannot see that.

Or perhaps us fingers-in-ears Musk fans don't expect the AP to be perfect. Especially considering how new the fields are.

Autopilot doesn't need to be perfect, just better than humans.

So because the field is new, it's okay that a human being is dead?

It's not just that the field is new, it's that 1) SO MANY people die constantly in cars, and this is the start of trying to change that permanently but it cannot be perfect out of the gate, 2) nor should we wait until it's perfect if waiting that long means more people die.

Ultimately, we cannot rely on mere anecdotes. We need statistics that show it is worse. If it's as bad as the media coverage claims it is, it should be easy to demonstrate such statistics.

If human beings dying weren't at least somewhat ok, we wouldn't drive cars.

Thomas Edison would agree with your stance. Alternating Current is not safe and therefore should never have been implemented. https://en.wikipedia.org/wiki/War_of_the_currents#Edison's_a...

The thing is that the problem isn't just the Autopilot driving over white lines. Even cheap VW Up's and other micro cars would have applied the brakes automatically when an object was in the way. It is a huge failure that a Tesla can do both without stopping. Two huge fails.

You're right, it was a failure, a terrible failure, but it will be improved.

I think the issue is, if you read https://news.ycombinator.com/item?id=17256900 is that Autopilot is not better than humans.

I did, unimpressive, he doesn't have any relevant data.

My reply to the post though:

Obviously someone shaping stats to their own biases. They want to compare Tesla and Mercedes Benz crash rates because they are in a similar price range?

Mind you he has no data on automated driving crashes.

Ridiculous and short-sighted, the goal (IMO) is for all cars to be automated. And he assumes that just because lidar is expensive now, that it will be in the future, which is just naive.

Really, no comments from the down voters?

Only fingers-in-ears Musk fans cannot see that.

Speak of the devil...

It’s amazing that given the slander Tesla has been passing off as PR in this case, that anyone could defend them in this, yet here we go.

It must also never be worse than humans.

In this case, it very clearly was.

So human drivers wouldn't crash here? Or a good, attentive, best-case human driver wouldn't crash here? Because those are not the same set, and it seems the former set has crashed into this particular barrier multiple times.

I don't think it's acceptible that a vehicle with adaptive cruise, lane keeping, and emergency braking would hit this barrier without any mitigating braking or action, but the actual state of drivers on the roads and human traffic fatalities is pretty grim as-is.

I think it's "rather unlikely" to the point of implausibility, that even the human drivers who have crashed into that barrier:

1) _aimed directly at it_ rather than glancing blows/swerving at the last moment, etc., and

2) _accelerated beyond the speed limit as they did_, rather than applying any form of braking, let alone AEB.

> I think it's "rather unlikely" to the point of implausibility, that even the human drivers who have crashed into that barrier.

So implausible that the barrier was destroyed just weeks earlier by a human driver?

I find it equally implausible that you willfully ignored the context of my quotes. To the extent that you literally changed the colon that lead to the context of those quotes into a period to make it sound like that was the extent of my statement.

Yes, human drivers crash into crash barriers, as noted.

Now, how many of those human drivers "actively aimed at the barrier", as the NTSB stated? And how many of those human drivers, having aimed at the barrier, "then accelerated in excess of the speed limit" as they did so?

And if you do reply, try not to take sentence fragments as an excuse to change quotes to misrepresent someone's statements.

I ignored it intentionally because that is a mischaracterization of what the NTSB report said. That is what clickbait headlines said (misleadingly). The autopilot was following the lines, which looked a lot like a lane. Human drivers make similar mistakes, and it is quite likely that’s what happened earlier and why the barrier was destroyed beforehand. The lines are about the same size as a lane and lead you straight into the barrier.

> So human drivers wouldn't crash here? Or a good, attentive, best-case human driver wouldn't crash here?

I'd argue that a Reasonable Person [1] wouldn't crash here.

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

I think that's actually a really good method to apply in these cases, and I agree.

However I also think that the roads are currently occupied by many people who fall below that standard.

Should a car with assistive technologies only be a success if it is as good as a reasonable person, or if it is better than a typical person?

Obviously the former is prefereable to the latter, but people seem much more content with the idea of driving themselves into walls (or another typical person on the road might crash into them and cause that outcome) than with the idea that a computer might do that same.

See also fear of flying vs driving even though flying is demonstrably safer, but feels much more out of the hands of the flyer.

Except a human driver had just recently crashed in the exact same spot - that's why the divider's crash bumper was damaged at the time of the Tesla accident.

I think there's no excuse for shipping an "autopilot" system that can't detect massive, stationary objects in broad daylight, but humans also do crash in those exact circumstances all the time.

It needs to be better than a sober, awake, human of the type who typically buys a Tesla. That would be the fair comparison, not just a random member of the public with a 20 year old car. Better than a drunk human driver, or a 90 year old is not a standard reasonable people care about when they say “better than a human.”

A human driver crashed in that same exact spot one week earlier, so you can't really say AP is necessarily worse because of this accident.

Teslas are a tiny minority of the cars driving on the road.

Even if a car crashed into that barrier once a week (it's way less than that), a single Tesla crash in all of Tesla's history so far is an order of magnitude worse.

https://news.ycombinator.com/item?id=16720848

They were also, as they have been on multiple occasions, more than happy to make sure the public narrative was "our system was working, warning him, not our fault!" when they absolutely, utterly, knew that those warnings were utterly irrelevant to the situation at hand.

Just as you or I as Tesla drivers need a subpoena to get access to the black box data for our own insurance or criminal claims against other drivers, but Tesla is more than happy to put out press releases on your driving telemetry if they feel it suits their narrative.

Tesla acknowledges that autopilot can fail.

Reply on news.ycombinator.com