Functioning cities often shutdown for a day here or there for weather. I live in a northern city where we laugh at southern cities for shutting down for 1 inch of snow - but it is the right thing for them because it doesn't happen enough to be worth dealing with. If my city shutdown for 6 inches of snow we would be shutdown unacceptably often so we instead have higher taxes to pay for all the infrastructure needed to deal with snow (though honestly this isn't much $ in the total budget).

Which is to say cities need to figure out what is the best use of their efforts/money. It is wrong to fault Atlanta for not dealing with this. If you live there you as a voter should learn all the pros and cons (I suspect there are some unexpected environmental ones) and consider if you should vote for a change or just deal with it. The rest of us won't don't live there though should keep our fingers out of their local issues.

And when it does happen: A Waymo should not fucking drive through it.

I remember once when the mall in my hometown flooded. It was at the top of a hill.

IIRC: The top of that hill received something like 6" of rain in less than 15 minutes, in a very "Fuck you in particular" sort of way.

The vaguely-greater surrounding area was fine. It was a very localized event.

They were not prepared for this. It was a mess.

And gosh: The streets near there flooded, too. The drainage systems were simply not up to the task.

It had never happened before, and it has never happened since, but: Quite clearly, it happens.

(I don't understand your deflection here, at all. If your main point is that "If cities were designed better, then the deficiencies of autonomous cars wouldn't be a big deal for those autonomous cars at all" then I might reasonably conclude that you're just not particularly observant of the world.)

---

edit: People also screw things up. We (people) drive through flooded roads sometimes -- we even do it on purpose from time to time, even though the guidance is to avoid it.

Some other times, we get surprised by flooded roads. Especially at night, they can be hard to detect. We screw things up. We take risks. Sometimes, those risks even work out OK.

But back in context: Waymo. Waymo is an autonomous taxi cab. It works on regular public streets, and on a long-enough timeline: Some of those streets will be flooded.

I probably never want my taxi driver to try to ferry me through a flooded roadway, whether it has a human brain or a computer brain calling the shots.

(I did get to spend a week getting ferried ~daily through flooded roads in a Jeep once in an unrelated flood, but by a high-ranking deputy Sheriff was (who would not become confused by a power outage[1]), and this Jeep was a proper cop car with the lights and the logos. We had some mutual problems that needed solved that involved public safety, and both of us were being paid to solve those problems. That worked fine, I knew what I was getting into before we set forth, and we'd have had extraordinary support if anything went very wrong.)

[1]: https://news.ycombinator.com/item?id=46342412

There are problems.

There is money you can throw at those problems.

And there are some problems that are rare & low impact enough that it's not worth throwing money at them.

See also: keeping snowplows in Atlanta.

In some cities, certain streets are designed to flood during heavy rain, and are an integral part of routing the water away from neighborhoods and businesses, and into the drainage systems.

Also it's not like we never have flooded roads here in the valley.

Whatever is going on, I'm confident it's not a result of straightforward parochialism in the way that you say you're comfortable assuming.

The engineers whose expertise you assume away are actually debating corner cases like the one we saw of someone carrying a bicycle on their skateboard.

In fact the companies run test campaigns in shitty weather all over the country on purpose, at great expense.

At a previous employer, we've seen anything from cars getting mooned, a SUV slowly driving past the AV, the rear window roll down, and someone poke their head out and start throwing dollar bills at the AV, a convention of people dressed up in animal costumes, the "Miami left," and so on.

So it's much less of "maybe we should test that" and more of "we don't know what we don't know, so let's gather some data." In practice, the cars have lidar so they won't crash into solid objects that aren't recognized, they just end up getting stuck in embarrassing situations like these.