I just read a pretty useful article on the Vox site about self-driving cars. It could turn out to be wrong in many ways, but a few things clicked for me after reading it, like a better understanding of why Apple or Google might want to get involved in the automobile space, and why their expertise might make sense there too.
It’s a move from car as a widget that a car company sells you to car as a service to which a service provider sells you access. Repairs? That’s their problem. Gas or electricity? It’s baked into the price. Parking? Don’t need as much of it, because like a taxi, these cars don’t sit still most of the time. They move on to a new user who needs a ride. So those are some potentially cool features. On the other hand, if you want to jump in the car on two minutes notice and go somewhere, well, that may not always work so well.
At the end of the day, a self-driving car service like that would be very similar to a service like Spotify in the music space (or Netflix in the movie space). Instead of buying the hard product CD or DVD widget, people buy the right to access the same music or video from the service provider.
We’ll see how it all plays out, and I’m not sure if I’m totally sanguine about that outcome, particularly if Uber is a big player. But it is interesting to see how the success of services like Spotify, Netflix, and Office 365 have started to expand the adjacent possible of this sort of business model outside of the realm of intangibles, like digital music files, and into the realm of large tangible things, like cars.
Obviously, technological changes are driving some of this shift. But to a certain extent, it’s also about changing people’s cultural expectations around the value of “ownership” of a thing vs. access to the service that the thing has historically provided to us.
In the final analysis, the project of 20th-century, consumer capitalism was about extending the possibility of widget ownership as broadly and as deeply as possible. But as the Internet continues to link more and more people and information together, it’s those intangible linkages that provide much of the innovation and the power, rather than the widgets themselves, which are much more conduits for those linkages (i.e., more commoditized).
So more of the value is in the service itself rather than in the widget that delivers the service. And particularly if the service is delivered via centralized cloud servers, it’s much easier for a service provider to control access to the underlying service. As a result, digital piracy is much harder for the average user to accomplish. It also makes it easier for the service provider to roll out updates and security fixes. On the other hand, individual privacy rights may suffer when everybody’s information lives on a central server (self-driving cars would be no different–it would be much easier to track people’s movements if most travel happened in self-driving cars).
That doesn’t mean there is no money to be made making and selling certain classes of widgets (e.g. smartphones), but as more and more classes of activities are being mediated through the same widget, the markets for a lot special-purpose widgets have started going by the wayside.
We’ve already seen that process starting to play out in the context of things like music, books, and movies. But if it continues into things like cars, that’s going to mean some disruptive long-term changes that dwarf anything we’ve experienced thus far. It’s not quite the Jetsons. But it’s getting closer.
Anyway, nothing really profound in this post. Just some quick thoughts that came clearer for me. So apologies in advance if this is some Captain Obvious stuff.
Update:
A friend asked me the following question today after reading this post:
What about taxis? Don’t they already provide the service-based model for auto travel I’m talking about above?
Of course, the answer is “yes,” up to a point. With a taxi, you are buying the service instead of the vehicle, but the proliferation of the taxi is far more limited than the proliferation I imagine if the driver-less car were to really take-off. At best, at least in most American cities and towns, Taxi service is currently an adjunct to private cars and transit.
The networked nature of the driver-less car also makes it more easy to imagine the viability of a subscription plan for driver-less car service rather than the metered fee-for-service approach of taxis. This is what would make such a service the Spotify of auto travel, either a capped or all-you-can eat subscription within a given geographic area.
In this regard, the driver-less car might be more like public transit, where you buy a monthly pass. Indeed, it might work as a sort of compliment to public transit, ferrying people to and from trunk lines where it’s not efficient for them to provide service. This would mean less space needed for parking and less traffic on the roads, particularly if there was a way for multiple people to easily arrange to share a driver-less car to a shared destination (or where it’s efficient to hit contiguous destinations in the same trip).