The User/Volvo/whatever approach may deliver earlier value, but the Google approach is what's needed to have safe cars in uncontrolled urban environments. I ge

The User/Volvo/whatever approach may deliver earlier value, but the Google approach is what's needed to have safe cars in uncontrolled urban environments. I get quite frustrated with articles and headlines and even analysts who don't understand the quite fundamental difference. Uber, who lives almost entirely in that space, will be in for a rude awakening IMO. Unless they already know they are ten years behind Google, and plan to use instrumented human semi-operated vehicles for learning, and cynically market it as "self driving" when it isn't. That'd work, too.