My first ride in Google's self-driving car was, all at the same time, thrilling, fascinating and a little disappointing.
The car was in Washington DC where Google representatives met with groups like the AARP and the National Council for the Blind, groups which might have an interest in cars that that could act as chauffeurs for those who, for one reason or another, can't drive themselves.
I got to ride along on a loop around several DC blocks with two Google engineers in the front seats. Google's "self-driving cars" must always have someone seated at the controls, whether in Nevada -- which recently licensed Google's cars -- or anywhere else.
The drive was thrilling and fascinating because, come on, the car drives itself. In traffic! Disappointing because it's clearly not going to be ready for public use for years and years.
For now, at least, the car only drives routes it's been trained to drive. My ride in Washington DC was along a route that Google engineers had driven with the car earlier. Google refused to allow the car to be driven anywhere beyond this well-studied environment, at least not with the media tagging along.
Still, that doesn't mean it was a cake walk.
No Google engineer taught the car that a bunch of kids on a field trip would march out in front of it at an intersection. It stopped and waited for them on its own. And no-one told it that, right after that, another car would run the four-way stop sign right in front of it. It handled that, too, avoiding a collision all on its own.
At first, those interactions seemed boringly normal to me until I remembered... no-one was driving! The car had done that all itself while the man in the driver's seat sat passively watching.
All of this is made possible by an array of sensors that would make a spy satellite jealous. The Google car has three GPS antennae, radar systems, cameras to read street signs and traffic lights and it's topped with fast-spinning laser eye that looks like something out of a cheap '50s sci-fi movie. That all-seeing eye scans for cars, pedestrians and obstacles.
Google doesn't make cars or sensors, though, so the part the California tech company is really interested in is the software that ties all this together to let it make -- we hope -- safe and rational driving decisions.
Since the Google car only just got its learner's permit, it drives accordingly. During our test loop, it stopped a few times for phantom threats, like a parked truck that was just a little wider than the cars around it. Then there was the jerking halt on a side street caused by a car that stopped a little abruptly almost two car lengths ahead.
When it wasn't sure what to do, the car would hand control back to the driver, announcing it was doing so in a friendly female voice. (The driver can always take control at any time by just by moving the steering wheel or touching the pedals, even slightly.) "Self driving" was resumed by pushing a big green button on the Prius's center console near the even bigger red "kill switch."
Surprisingly, one thing the car can't do all on its own is use the turn signals. The driver still has to do that.
"That's been on our to-do list for a long time now," said the engineer riding shotgun.
Back-seat "driver": I had to ride in the back. A second Google engineer rode in the shotgun seat with a laptop computer. On his screen was a triangle -- representing us -- surrounded by a vast army of colored boxes, representing cars, people and stationary objects, all sliding across a black screen. It was reminiscent of the old arcade game "Tank Commander," minus the explosions.
Times when the car lost its nerve and let the engineer take over, such as when it encountered an on-coming car on a narrow street and wasn't sure there was room to get around it, weren't just useless glitches, I was assured. The data from each situation would be ingested and analyzed so the car could learn what to do in the future. Those lessons could, hopefully, be applied to a broad range of driving conundrums.
Before Google realizes the dream of a truly "driverless car" there are many steps yet to be taken and some of those steps remain far off. The first will be allowing the car to stray from routes that it has been specifically trained to drive. Until then, this is all just baby steps.
But the biggest step will be to create a car that will let me just sit in the back seat with no-one at all in the driver's seat. That step still seems -- to me -- many years off. If Google can get there before a major automaker beats them to it, I'll be really impressed.