Car accidents (and unmanned vehicles). But mostly car accidents.

However, the “engineer?” was an employee of Uber. Companies are responsible for what employees do while in their employ. Well, there is a chance the driver/rider was hired as a subcontractor, which is exactly what Uber would totally do. However, the new leadership will not want to have to defend this case.

I grew up in Massachusetts, and went to college in Boston. My entire life, I have never once assumed that a car was going to stop for me, whether they were required to or not. Because honestly, in MA/Boston, they often don’t. As a result, I have always done exactly as you describe – go only when I’m sure I can make it across without any chance a car would hit me if it didn’t slow down. Most people I knew from Mass. did the same.

When we moved to Seattle ten years ago, everything I knew changed. I was constantly mystified by how every single pedestrian in the entire area just strolled out into the street, often without looking, sometimes looking but then going anyway. I mean, I always yielded right of way, though I’d grumble under my breath when they’d do it outside of a crosswalk. But yeah… it’s incredible how many people just walk into traffic and assume cars are going to stop for them.

I’ve done some factory work… and this is worst. You actually do something in the factory, the actual work. Sure you can get into auto mode and there is some risk, but they can’t possibly expect them to just focus straight ahead doing nothing for… hours. I mean they can expect that, but it’s not going to work

I worked at a printer right across from the USS Enterprise on 11th ave. We had all of the safety guards removed a long time ago. We would use our fingernails to scrape ink blots, called hickeys, from running rollers. We were told not to wear rubber gloves around the running machines. Yet twice in my time there we had new guys with gloves sucked into the rollers.

Have you ever seen an arm stripped of the skin? And rollers bent out of shape because of it? A hand becomes an arm really fast. And no matter how fast you hit the big red STOP button, it’s way too late.

One guy lost his hand. The other guy was lucky. He only got gloved. Degloved? Whatever.

Spent 11 years working in printers. No amputations, but definitely more than a few smashed fingers. One guys thumb got mangled so badly it was the size of a golf ball.

4 months later.

That said ours used blankets, so when there was a hickey, there was no scraping it off. Usually they were a chunk of ink or UV coating that dislodged and smashed a blanket

I agree with that. I mean, I’m from a small town in the south but you learned very quickly to respect the mass and momentum that even a small vehicle carries. It just seems like people out here … never learned that, somehow.

Dude, you have definitely seen (and done) some shit. Judging by the stories you’ve told here of your younger days, the “you need to write a book” cliche absolutely applies.

Pretend there is an animated gif from The Machinist here, but don’t actually look for it.

Very true. When I watched the video my first thought was that I would have almost certainly hit her even at 100% attention.

I find the hand-wringing about this somewhat puzzling. Even computer systems are going to make mistakes but even a semi-competent one is going to make far fewer mistakes than the average driver and be able to navigate congested roads far better. Fewer deaths + less drive time = win/win in my book.

I’ll try to make it clearer to you. Watch the other videos of that road. I think Uber might have released a misleading dashcam. I’e never seen a road that dark that had lights where the only thing you can see is that narrow of a ray of headlights.

I would have seen her in the left lane under the circumstances shown in the other dashcams. I see people corssing hwys here in the middle of the night wearing dark clothes all the time.

Calling that hand-wringing is just disrespectful.

Having said that, I don’t think Uber or a driver is at fault but their dashcam video showing actually line of sight is questionable.Also, if you’ve read anything, like any other article about these systems, you would also know LIBAR can see in the dark.

I don’t know if the other videos controlled for all the variables including time, phase of the moon, etc but it’s pretty obvious the Uber footage is not even close to indicative of what human vision would see. I don’t understand why the various sensor packages wouldn’t have picked her up, but they obviously should have.

Last night, i read Uber dropped their sensors down to one on the roof while everyone else has several more…

https://www.reuters.com/article/us-uber-selfdriving-sensors-insight/ubers-use-of-fewer-safety-sensors-prompts-questions-after-arizona-crash-idUSKBN1H337Q

I haven’t seen any info from Uber that states or even implies that the dashcam is reflective of what a human would have seen. I assume that it is the best optical camera footage they have of the incident. It’s not uncommon, at all, for dashcams to be pretty low quality. Now, if they have better optical camera footage and held that back to release this crappy one, instead, that would be scandalous.

So far, it seems that any assumptions of users that the dashcam footage is reflective of what they would have seen is just people’s assumptions.

Except there are several dashcam videos out now of that very road at night. Unless they know some roads lost lights or something, roads don’t suddenly become dark one night and not the other. That dashcam shows the surrounding lanes like they’re pitchblack and videos of that road says it is not based on other dashcam videos.

I’m not saying Uber equips their cars with the best dashcams on the market. I’m saying it’s probably (hopefully) the best camera footage that Uber had.

That could be true but…why wouldn’t they have some of the best cams on the market? These other videos are from individuals who are just locals driving in a car with dirty windshields and some dashcam they picked up at a store. Uber is sending out autonomous vehicles doing test runs in areas where there are a lot of people where they can and did kill someone… why wouldn’t that vehicle have some of the best dashcams available?

Actually, its more likely that the lighter images are from a cheaper cam because that is how they’ll compensate for a cheaper sensor in the camera. There are endless reasons for the differences in footage really. However Uber is evil so lets ignore all those and claim they doctored the footage instead.

Except I drive roads at night. I know how they normally look. That’s a lighted road. Did you know that. That the road that car is traveling on is actually very well lit?

Because the dashcam isn’t part of the driving sensor suite. It’s there for their internal review/processes. Not everything is going to be the best of whatever. Certainly them being a little cheap isn’t basis for assuming that they doctored the footage.

Well them cheaping out and going down to one rooftop option when their competitors and they themselves had more isn’t great optics either. I can certainly question any attempts to save costs in these areas just like you’re free to give them a benefit of the doubt and assume they went all out in all the right place.