Monday, June 2, 2014

When it comes to technology journalism, reports of progress don't have to be directionally accurate

First the disclaimer: I have a number of friends and relatives who, for reasons of age or vision, do not or cannot drive. For these people, a self-driving car would be a tremendous, life-changing technology. That reason alone more than justifies the time and money going into this research.

Still, as eager as I am to see this realized, I am completely sick of the Pavlovian coverage it invariably generates. Every year or so, Google calls a press conference and the tech reporters start to salivate. This latest example is particularly embarrassing:
Whether we like it or not, self-driving cars are coming. Soon. Perhaps sooner than we may think. Research and development by Google has been going on for some time, but now the technology giant has revealed its latest project: its own self-driving car prototype. In the past, Google used the Toyota Prius and Audi TT as the test bed for this technology. Its latest prototype is just a two-seater and has a top speed of only 25 mph.

There is no steering wheel, gas or brake pedals, just a display screen showing the planned route and buttons to start and stop. The interior is limited to two seats with seatbelts and a space for passenger’s belongings.
I apologize for another ddulite tirade but in commercial and public policy discussions, technology needs to be evaluated in terms of functionality and costs. Being more advanced means doing more and, if possible, doing it for less. At least based on the short version you get in most news stories, Google's new approach represents a far less technologically ambitious project. Compared with their earlier prototypes, their current cars have more limited functionality, less compatibility and are constrained to a very low speed.

That's not to say that Google isn't making progress (I'm sure they are) nor an I saying that focusing on this less ambitious approach won't, in the long run, actually lead to faster progress (we often end up getting more done when we don't try to do too much too soon). It is, however, difficult to see how, in and of itself, switching over to prototypes that require customized cars and which simply avoid the thorny problem of going from automated to manual and back again suggests an advance.. .

Just to be clear, the point of this post isn't to criticize Google or its X lab. First because, barring strong evidence to the contrary, I'm going to assume these are smart people who know what they're doing and second because, even if I doubted their competence, I wouldn't be qualified to discuss what they were doing wrong.

All I know, as the saying goes, is what I read in the papers, but that's the problem. The story I read doesn't seem to support the conclusions the reporters drew.

No comments:

Post a Comment