I focused too much on your wildly inaccurate statements after this instead of the meat of the subject, which is this. Google thought they could launch a service in Phoenix last fall but it didn't really pan out. But they were confident they could do it as recently as last summer. They hit some bumps and postponed things (because Google is playing the super long card here) and, even though they lead the world in this, thought some more work was needed (largely because of Uber and Tesla being reckless). We're not 20 years from this ****. We're maybe two, maybe ten, maybe fifteen years from it at most. Personally, I think we're about ten years from it. Tesla has its problems but it's collecting *mad* driver data despite its hardware issues. Cruise (the GM company) is going hard at San Francisco streets to show how its software can take the most challenging arenas. Google is going much wider; it's deploying its cars in a variety of streets and in massive numbers, but not as aggressively focused as Cruise. This is going to play out quickly. I don't think people realize that once an algorithm and machine learning is implemented on this kind of scale, how much things will change... about everything. And how much everything that comes before it will change it. Teaching a machine to put a bolt in a hole is easy. We've done that for years. But the collaborative effort of teaching machines to understand and predict behavior in an open environment... Well, that's something else. And we're on the cusp of that happening. And once it happens, all hell breaks loose economically.