I'm a little biased but I don't think Waymo is shortcutting this. If you look at the data they're looking at and how it's being processed, it's several orders of magnitude more reliable than what Uber was using. I have yet to see a sensor data video from Waymo that hasn't targetted all obstacles successfully, even if they're not categorized right. If anything, I think Waymo is, in some cases, being too conservative with its driving data (although, admittedly, I don't think I really mean that since the alternative is the loss of human life). If you've ever driven behind a Waymo vehicle, they're annoyingly strict when it comes to following posted speed limits, stopping for obstacles, and reacting/erring on the side of caution. It's an infuriating exercise in patience but I hope that it will pay off in the long term.

> If you've ever driven behind a Waymo vehicle, they're annoyingly strict when it comes to following posted speed limits, stopping for obstacles, and reacting/erring on the side of caution. It's an infuriating exercise in patience but I hope that it will pay off in the long term.

I've noticed when I am riding with a good (human) driver who obeys pretty much all rules of the road and drives slower than the posted speed limit (at a speed they are comfortable driving) that there are always a few drivers behind us who will honk their horns, flash their lights, or even pass into the next lane just to pass back a little too close for comfort. I don't know if they are always horrible humans but that's besides the point.

How would a self-driving car react in such a situation? Would human drivers be better behaved around a Waymo vehicle because the Waymo car has a lot of cameras and sensors and can pretty much show a clear cut case of road rage?

This is exactly where I think that Waymo wouldn't get into that situation. A Waymo vehicle would let the car in as its cameras will actually react to the turn signal. Technically, a line of cars could stop a Waymo vehicle in its tracks just by getting next to it and turning their blinkers on. The car would try to yield and zipper merge but will almost always defer to letting cars in if they get too close. Again, it's super annoying as a human but I think I prefer how conservative they are when it comes to safety.

Speed limits are generally set too low. Someone driving even slower than the posted limit can often be assumed to be intoxicated or impaired.

Sorry, any speed limit will be "too low"; people will consistently drive at appx the speed limit +15%, whatever the number. Shifting the speed limit to old limit1.15 will result in speeds of new limit1.15.

That's not how it works, because it turns out that drivers are generally not suicidal.

The studies have been done, they've been replicated, and the statistics are very clear. Google the '85th percentile rule'.

I haven’t had the experience of driving in proximity to a Waymo vehicle like you have. Your statement could be viewed as taking a conservative approach to that automation is in itself a shortcut. That of course is speculation; I feel confident that when it comes to automated driving by virtue of the infancy of the technology, there are things we don’t know that we don’t know yet.

>Your statement could be viewed as taking a conservative approach to that automation is in itself a shortcut.

Can you expand on that? I'm not sure I follow what you're saying here.

I’m thinking the conservative approach in this context is a shortcut to solving all of the problems. For example, hitting the brakes at any unrecognized object keeps the passengers safe, and is a shortcut in that the recognition models for that particular object don’t have to be ironed out. Or, selecting a test city like Chandler, approx 60 years old, having the signal pole and crosswalks positioned at the same offset from every intersection is in itself a shortcut. That tech will never work in The Hollywood Hills, for example.