Electrical Engineering: A Diminishing Role?

Before this year's superbowl, there were television ads featuring the famous Green Bay quarterback, Brett Favre. Carelessly spinning the football with a look of supreme confidence, he told the camera, "I can be anything I want". Then after a moment's pause, a shadow seemed to cross his face, and he added as a confessional afterthought, "… except an engineer." A slight smile indicated that he thought that this wasn't such a bad failing to admit to.

Ah, well, I wouldn't do so well as a professional athlete, either. But throughout my career I've always been proud of being an electrical engineer. Of course, there were momentary lapses at parties when I would say that I was a scientist or mumble something unintelligible. For the most part, however, electrical engineering has been an honorable and rewarding career for me. But now, for the first time, I am worried about the future of our profession.

I see a trend in the downward enrollment in electrical engineering and in the changing focus of the work place around me. The problem is that maybe electricity matters less than it did. Attention in technology has moved increasingly to applications, irrespective of the mechanisms of the physical platform that supports them. Maybe, I think, an ice age is coming for classical electrical engineering.

Remember when computers were called "electronic computers"? That was a long time ago. Now no one seems to care that they run on electricity. They could contain gears and pulleys, as far as most people are concerned. No one bothers about what's inside today's computers, because for most of the world it's irrelevant.

I blame all this on Heathkit. By going out of business, they took the fun out of electrical stuff. For a while after their demise it was possible to play with the hardware in home computers, but that soon ended. I thought something else would come along, but it didn't. A lot of the fun things disappeared into the Lilliputian mazes of faceless black chips, though, electrical engineers were arguably among the most important people of this century.

Now, though, the huge boom is in what is called information technology. People in the most nontechnical of occupations talk knowingly about their problems with the registry in Windows 95. I can't get on an airplane without hearing conversations all around me about modem connect speeds and office 97 operability. The world is filled with users. In contrast, far fewer designers are needed. Even most of us engineers, trained in design, go out into the world as users.

It's not that electrical isn't important. It is terribly important. The world rests on it's shoulders. Someone has to know what's inside those chips. Progress depends on someone making them better along the lines planned out by Moore's Law. Still, unless you are that someone, in one of the few privileged design positions, the world will take that ordained process for granted.

The sacred legacy of Ohm's Law is sinking into an abyss of complexity. Even circuit designers can seldom be concerned with the details of electronics. Most will work at simulators, and even then at higher and higher levels of aggregation and abstraction. Only at these higher levels will designers be sufficiently empowered. When a billion transistors sit on a chip, how many people will care about a single circuit? Moreover, the higher levels of abstraction will become progressively isolated from the principles of classical; electrical engineering.

There is, of course, a middle ground. Electrical engineers also learn algorithm design. This is fertile ground, because of the burgeoning power of the electronics underneath. Complex algorithms unspeakably costly a few years ago have become nearly free to implement. So we inhabit the middle and lower layers. But I look jealously on all those applications above. That territory is being occupied by a growing number of computer science graduates, and is not considered the natural province of electrical engineers.

I see the world as an inverted pyramid. It balances precariously on the narrow point at the bottom, which is occupied by the physical layers of the real world and peopled with engineers and scientists who build devices. This point is being impressed into the ground by the heavy weight at the wide top of the inverted pyramid, where all the applications reside. Like Atlas, those relatively few physical designers at the bottom have the weight of the world on their shoulders. The teeming hordes above have little appreciation for their travails.

So there is a dilemma. Classical electrical engineers are trained in design. A few designers will be critical, eagerly sought after and handsomely paid. Educators will pride themselves on the demand for their shrinking number of graduates that they produce, while other disciplines will produce growing numbers of informed users who will work at the application levels. Electrical engineering will be in danger of shrinking into a neutron star if infinite weight and importance, but invisible to the known universe.

I tell myself that the downsizing of any profession is a natural thing. Perfectly good occupations come and go. In the Middle Ages it was undoubtedly prestigious to be a troubadour. Troubadours probably held learned gatherings where they told themselves how important they were to the world. Then the world changed, and the schools were producing more troubadours than were needed. Incoming students began to select other fields of study, like jousting. The old troubadours went on singing their songs, but no one was listening anymore.

If current trends endure, future computers will consist of a single chip. No one will have the foggiest idea what is on it. Somewhere in the basement of Intel or its successor will be a huge computer file with the chip's listing. The last electrical engineer will sit nearby, handcuffed to the disk drive in a scene out of Ben Hur. That engineer will be extremely well paid, and his or her every demand will be immediately satisfied. That engineer will be the last keeper of the secret of the universe: E=IR.