A recent article in the New York Times highlighting the drawbacks of onboard electronics should be mandatory reading for anyone designing advanced technology cars, such as hybrids--or anyone considering buying one. It raises two basic issues: More sensors and actuators (tiny electronic motors) result in more things to break down, and in addition, "Some complaints turn out to be not failures, but features that are difficult to use." Both of these problems can damage customer satisfaction and erode market share and profitability.
Too many of us have experienced the first problem category. I once owned a Toyota Camry with an intermittent electrical fault that periodically shorted out the power to all the accessories, including the air conditioning. It took half a dozen dealer visits to isolate the fault to a single defective chip. Now, this sort of thing is annoying in a car with an otherwise bulletproof reputation for reliability, but it could be a mortal blow to an entirely new model, especially one with a novel powertrain.
The second concern stems from engineering hubris; some carmakers mistakenly believe that if they can do something, they should. Despite a few obvious refinements, the basic "user interface" of steering wheel, pedals, and dashboard knobs and buttons hasn't changed much in fifty years, for good reasons. At 70 miles per hour or on a congested city street, any control that isn't immediately intuitive is a dangerous distraction. Contrast the Toyota hybrid drive screen, which has gotten high marks from critics and owners for the way it displays power status, versus those all-in-one displays that require "joystick" navigation through multiple sub-menus just to change the climate setting.
The introduction of entirely new powertrains, such as hybrids and fuel cells, will create a new set of reliability hurdles, since neither system enjoys the billions of vehicle-years of experience that the standard internal combustion engine (ICE) has behind it. Although the reliability of these new car types should improve quickly--and in the case of fuel cells may ultimately exceed that of the ICE--much of the consumer's experience of owning these cars will be governed by the user interface and accessories, in addition to the new drivetrain, so they present a dual reliability challenge to designers.
While I'm skeptical about the quote from IBM suggesting that within a decade all cars will have essentially the same mechanical systems and differ only in software, it underscores the need for automobile software to be more like that of an Apple computer and less like a Windows PC. After all, on a crowded Interstate highway, the dreaded "Blue Screen of Death" could be just that.