A Lesson from Toyota's Travails

The problems now being experienced by Toyota - braking problems on hybrids and sudden acceleration on many other models - may be mechanical, electrical or software related. Purely mechanical problems are often relatively simple to solve. Electrical problems, however, are often intermittent and difficult to diagnose while software problems are an order of magnitude harder to find.

These thoughts, written by James Arnheim, are applicable to system design:

"Engineering: the art and science of molding materials we do not fully understand; into shapes we cannot precisely analyze; to resist forces we cannot accurately predict..."

Although he was discussing mechanical and structural engineering, Mr. Arnheim's remarks are also applicable to the design and writing of software. As an example of the difficulty of creating complex systems, defects in the various iterations of Microsoft's Windows(R) operating systems demonstrate the size of the task. It is close to impossible to create massive software structures that do not contain unpredictable and largely impossible to find errors - at least when the requirement of economic viability is applied to the testing process.

The aggravations and pain, imposed by bugs in Windows, are trivial compared to the defects that are will certainly be present in other, yet to be written, applications that will control critical processes such as, among other things, air traffic control, the electrical grid and the safe operation of our personal vehicles.

Automated systems have provided much convenience. They have also made major improvements in the quality of our lives: some appear to have even become essential. The potential dangers from unanticipated defects, however, may threaten our entire standard of living even if not our existence.

There is not much an average citizen can do except to become as little dependent as possible on any technology. The Amish - a uniquely American religious group that shuns technology - epitomize this lack of dependence. Most of us, without much thought, simply consider the risks and rewards of most technologies to be acceptable and the Amish way of life to be far too primitive for our tastes and our comfort.

Since technology is often a Faustian bargain, we should, but probably will not, evaluate the risk - reward proposition very carefully. But no matter how careful we are, no one should be surprised to find that the fundamental principle by which the Universe operates (the Law of Unintended Consequences) takes precedence over all others.

There just ain't no such thing as a free lunch. No matter how much we may wish it were so.