Will Ford learn that software isn't manufactured?

Ford Motor Company has just convincingly demonstrated that being an excellent industrial manufacturer doesn’t automatically mean that you are an excellent maker of digital technology. Despite Ford’s improvements in manufacturing quality, their overall ratings fell precipitously this year due solely to the poor software interaction on their dashboards. A recent article in the New York Times discusses Ford’s plummeting fall in user rankings this year, focusing the blame on their new touch screen interface.

Ford Display MyFord Touch on new Ford Edge—heavily criticized in J.D. Power's research and "frustrating" according to Consumer Reports.

According to the article, J.D.Power, the auto industry arbiter, dropped Ford’s ranking from 5th to 23rd, and subsidiary Lincoln’s ranking from 8th to 17th place. J.D.Power acknowledges that both Ford and Lincoln’s fit and finish are excellent. It was the “annoying” behavior of their driver-facing interactive systems that caused their ratings to plummet. Other reviewers concur, as Consumer Reports yanked their “Recommended” rating from Ford’s new 2011 Edge model.The automotive industry calls in-car digital information systems “telematics” and these include navigation, telephone, Internet, climate control, and entertainment. Drivers use telematics to communicate and control almost everything in their cars these days other than the accelerator, brakes, and steering.

Ford’s troubles follow a familiar pattern of older, industrial companies struggling with digital age problems. The challenges of digital technology, particularly its human-facing aspects, can’t successfully be addressed with technical skills rooted in manufacturing. What's more, the organizational structures of the industrial era can become counterproductive when applied to the people who make digital systems.

Neither can those manufacturing companies dodge the problem. Digital solutions are so much cheaper and more flexible than mechanical ones that they will eventually come to dominate the entire company. Companies who can master the challenge of software’s unique nature, and particularly of how humans interact with it, will thrive. Ford is learning the opposite lesson.

In my 1998 book, The Inmates are Running the Asylum, I posed a riddle for the information age: “What do you get when you cross a car with a computer?” Common sense will tell you that you get a smart car but, as usual, common sense is wrong. You get a computer with a motor and wheels. The dominant behavior, the behavior that the user perceives, is no longer the behavior of an automobile. Now it is the behavior of a computer, and making computers behave in non-annoying, let alone enjoyable, ways is a unique and very difficult problem.

Automobile manufacturing companies like Ford need to acknowledge that they are no longer making automobiles with attached computer systems. In reality, they are making computer control systems with attached motion mechanisms. The digital computer is increasingly dominating the driver’s attention, even more so than the steering and brakes. If auto makers don’t give equivalent attention to the design and implementation of these digital systems, they will fail, regardless of the quality of the drive train, interior furnishings, and other manufactured systems.

A major obstacle in Ford’s way is their failure to understand that software design and development is neither an engineering problem nor a styling problem, and that all of their significant expertise in these disciplines won’t help them with telematics. An even bigger impediment is Ford’s failure to understand that the very structure of their product development teams, optimized for old-school engineering-and-styling, stands in the way of fielding effective digital development teams.

Teams that can effectively build successful software systems are very different from industrial product teams, and they require different kinds of support from the organization. Digital development teams are small, composed of interaction designers, developers, and testers. They are trained to work together and they need an unfamiliar degree of freedom to find their correct path. This freedom is problematic within older manufacturing organizations because it usually transcends long-standing political and organizational boundaries. The telematic dilemma is a good illustration of the problem.

As currently experienced by the person driving, the modern automobile cockpit is the control nexus for several complex information systems. Each of these systems is arguably more complex than simply maneuvering a car down a road. These telematic systems include GPS navigation, telephone and internet communications, environmental controls, entertainment controls, weather reports, traffic reports, performance evaluators, timers, proximity detectors, auto maintenance sensors, tire pressure sensors, event recorders, and rear view cameras.

While interacting with one of these telematic systems can’t harm you directly, it can be extremely dangerous by distracting you from the road. It is impossible to overemphasize the importance of designing telematic systems that are not distracting, yet that is exactly where Ford has fallen down.

Ford and other car makers currently conceive of the auto cockpit as a car-control station, rather than as a car-and-systems-control station. In the existing cockpit control paradigm, the steering wheel is the largest and most prominent control. It is located right in front of the driver, up close, and the fact that it partially obscures the dashboard isn’t a problem because the only attention-critical information displayed on it is the speedometer. All other telematic systems are relegated to the center console, out of the driver’s line of sight, and positioned far forward enough to demand that the driver shift his body position to reach it.

Manipulating controls on the center console means taking hands off the wheel and attention off of the act of driving. This fact alone should make clear how dangerously wrong this arrangement is, but the center console controls are those of a digital computer, so they are complex, hard to use, unresponsive, obtuse, and frustrating in the extreme. I believe the main reason why these systems don’t kill more travelers than they do is simply because most drivers turn them off in exasperation.

Once you recognize that a modern automobile is a collection of digital systems in a traveling package, you can instantly see that all telematic systems must be placed directly in front of the driver, up close where he can interact with them without having to grossly realign his hands and attention.

This means that the dashboard in front of the driver becomes an important interaction device. Not only must the driver be able to read the central display, but he must be able to manipulate controls on it. It simply cannot be obscured by a large steering wheel.

Side Sticks

There are many solutions to this problem, from aviation-style side-stick controllers to Formula 1 racing car style steering wheels studded with dozens of controls and many more solutions exist on the drawing board. Modern cars can be equipped with inexpensive screens and LED readouts that keep the driver informed without forcing his attention off of the road.

Formula One steering wheel

Circumventing the conventional steering wheel is mostly a cultural problem. Cars have had these wheels for a hundred years, and changing such a fundamental control mechanism upsets people’s ideas of what a car is, or should be. Then again, that’s exactly what digital technology is in the process of doing, so it is entirely appropriate that people’s notion of rightness is being challenged. When older, senior executives express their reluctance to alter the steering wheel paradigm, younger engineers get the message: “The steering wheel stays.” Thus the newer technology is unsuccessfully wedged in the interstices left by the persistence of the older forms.

There are other, even greater organizational obstacles that perpetuate those interstices, quite literally. Ford is loosely structured the same way that its automobiles are structured. There are separate groups responsible for the drivetrain, the body, the physical cockpit space, and for the cockpit instrumentation. Those instruments are typically supplied by outside vendors, so the cockpit team assigns each one some physical space their product must fit.

This method of specifying reserved space for an outside vendor was developed back in the day of dashboard radios, when Philco or Motorola would supply an appropriately sized AM or FM radio to fit space allocated by the company. A modern telematics system plays a much more important role than a simple radio, and this arm’s length model is simply no longer effective. For a telematics system to be effective, it has to be fully integrated with all parts of the cockpit to the point of displacing the steering wheel. This means that the design and implementation of telematics systems is integral to the design of the entire cockpit, if not of the entire vehicle.

Back in the 1960s and 70s, it was efficient for an automobile company, with core competencies in big manufacturing, to outsource dashboard electronics to specialized vendors. but now those little radios have become all-encompassing telematics, and Ford, whether it likes it or not, has to integrate the design of its electronic solutions with the design of its manufacturing business. It's the riddle for the information age again: Ford isn’t a car company with digital capabilities, but it is a computer company with big manufacturing capabilities.

Designing and building a better automobile cockpit is the tip of the iceberg. The biggest task facing Ford and other car companies is changing the way they think and the way they work.