Be a student

Remarks by Jeong Kim President of Bell Laboratories at Alcatel-Lucent2008 G.W.C. Whiting School of Engineering Graduate Ceremony
The Johns Hopkins UniversityWednesday, May 21, 2008 Homewood Field

[Note: Prepared text. Not checked against delivery.]
Good evening. It's a privilege for me to address you today as you prepare for the next leg in your personal journey.
Some of you are preparing to depart from this sanctuary of knowledge. Some of you will be transitioning to a world where the college sweatshirt — I have to warn you — is not considered dress-up. Some of you will be returning to full-time careers, with new insights and new ways of responding to the challenges that define your professional lives.

But as we come together to mark this moment in your journey, as we acknowledge your successful completion of the program here at the Whiting School of Engineering, and as we celebrate your individual achievements, I'd like to make one plea: don't stop being a student.
In fact, never stop being a student. Because if you give up being a student, if you give up being a seeker of knowledge, you give up your capacity to grow. And you give up your capacity for wonder and revelation.
And wonder and revelation, ladies and gentlemen, are two of the essential ingredients of a great engineer.

Engineering is a remarkable profession, and I consider myself honored to be counted among you and the others who pursue it. I sometimes struggle when asked "how does one define an engineer?" I've taken to answering this by summoning up the image of a glass of wine.
You know the one. It's the glass where the surface of the wine is only half-way to the top. The optimist, of course, declares the glass half full. To the pessimist, the glass is half empty. But to an engineer, the glass is perhaps twice as big as it needs to be.
But what about the truly great engineer? The great engineer will actually taste the wine. Then, he or she will ask not only if the glass is the right size, but if it is the right shape for that particular wine. For as you may know, the shape and volume of air in the glass determine how the glass stores and then directs the bouquet to the nose — and the optimal combination of air and shape will vary according to the type of wine in the glass.
You see how your great engineer must be a connoisseur of life.

Forgive me if I appear preoccupied with wine. I just returned this weekend from visiting the great wineries in Bordeaux, France celebrating my 21th wedding anniversary. And while there, I learned that one of the worlds' greatest engineers — Leonardo da Vinci — lived out his final years of life north of Bordeaux, in a place called Loire Valley.
Da Vinci was indeed a great engineer. Think about his innovations: a calculator, the double hull, the concentration of solar power, a military tank, the design of a single-span 720- foot bridge, the concept for a helicopter... there's a man who never gave up on revelation.
The degrees from this fine university that you will soon hold in your hands bear witness to your qualifications as good engineers. But to be great — that is something no university can confer upon you. Yet to be great is my challenge to each of you.
To be a great engineer is to embrace wonder, to follow where your curiosity leads, and to act with passion and conviction on the insights that follow. Let me frame this attribute of great engineers by telling you about Sharad Ramanathan.
Sharad is one of our researchers at Bell Labs who loves to hike. Because he normally hikes early in the morning, he often had to clear the criss-crossing spider webs on the trail. It was really annoying to him that spiders can rebuild their webs so fast overnight — perfectly balanced and so well functioning.
Sharad was curious (note that word). How do spiders build their webs? He learned that spiders are nearly blind, and since webs have no odor, spiders can't use smell to build them, the size of webs are so much bigger than their bodies, and most importantly the spiders are in the plane of web so they cannot see the structure from a distance. This is like you building a 6 story tall web, blind folded, and by the way you cannot count your steps.
Sharad saw that a spider somehow manages to build this whole structure, one step at a time, sitting where it is, measuring something locally to produce what we might call a globally organized structure. Sharad hypothesized that what the spider measures might be elasticity ... perhaps measuring the tension at each node of the web, then using this measurement to improve overall web performance.
Sharad then wondered (note that word): what if, just like the spider, each node of today's wireless network was able to measure some property of the network locally and then communicate data to each of its neighbors? Then each neighbor in turn would take this new information, combine it with information it receives from its neighbors, and send a new value to all of its neighbors.
To cut a long story short, Sharad's insights became the foundation for developing spider algorithm, a revolutionary real time optimization algorithm for complex wireless networks or self-configuring network of the future. All thanks to curious and wonder-full Sharad Ramanathan — a great engineer.
The case of Sharad Ramanathan reminds me that to be a great engineer is to travel outside your own comfort zone, to explore alternative paradigms, to open yourself to alternative fields of knowledge. When he was becoming Bell Labs "spiderman," Sharad had the breadth of vision to engage biologists, neuroscientists, mathematicians, and other experts in disciplines quite distinct from his own.

Think back to Da Vinci: it's no coincidence, in my view, that Leonardo — the consummate engineer — was also the archetypal Renaissance man: artist, scientist, inventor, journalist, anatomist. No stovepiped thinking there.
Exposure to new frameworks of thought is fundamental. It helps us examine challenges from different angles. It helps us ask the right questions. At Bell Labs, as at many of the great research institutions, we like to provide a critical mass of enabling disciplines and tools. Because as the adage goes, if your only tool is a hammer, then every problem becomes a nail. It's also a way to develop resistance to a disease that afflicts many good engineers — something called "the curse of knowledge." Let me elaborate.
The knowledge you have accumulated here in your graduate work could easily become a curse. It becomes a curse if you are so wedded to what is "known" and thought certain that you can't consider alternatives. Don't forget that earth-centric views of the universe were state-of-the-art science — for 2,000 years — until Copernicus thought otherwise. And don't forget that for 2,000 years, scientists went through mathematical gymnastics to reconcile blatant contradictions from their own observation with this earth-centric "certainty."
A more current example springs to mind: a recent Slashdot posting about a software engineer who spent 2 days effort working to save three bytes of memory in his program — in an era when memory is abundant and cheap. This is a trivial case (unless you happen to be his employer) but it speaks to a general tendency to cling to the rules of yesterday. As engineers we are trained to build on experience, and we expect our experiences to add value over time. Yet the landfills of commercial failure are stacked with products that had too much complexity, too many features, too high a cost, or too confusing an interface — all because a design engineer was locked into the experience of the past.

The idea that past experience could be a detriment — that what we know to be true and certain may in fact be false — is a difficult pill to swallow. But it is the pill swallowed by great engineers. To be a great engineer is to know when to close the door on one endeavor in order to fully apply one's efforts to another. Let me provide an historical example.
Xiang Yu was a Chinese general in the third century B.C. In one famous battle, Xiang Yu led his troops across the Yangtze River into enemy territory. To the horror of his troops, upon reaching the far river bank, he had his troops crush their cooking pots and burn their boats. He wanted his soldiers to focus on moving forward, he said: retreat was quite literally not an option. (And yes, they won.)

Good engineers are loath to fully retreat from the safety of what has worked in the past. They're understandably reluctant to commit all their resources and energy to an alternative. I've seen this hesitancy many times. I recently encountered it in one of our research planning meetings. One of our engineers was describing the challenge of meeting 10-year-out bandwidth capacity forecasts. He outlined a set of proposed projects similar to those that have been successful in the past. But he also worried that this effort might fall short.
Our response was to shut the door on that approach and devote all resources to new territory. By not fully committing to the new approach — by spreading ourselves too thinly across old and new efforts — we would assure at best delay, and more likely, failure.
One of the factors that contributes to such hesitancy is data- dependence. It appears that all engineers have a gene that won't let them select an outcome without considering all available data and optimizing the solution. But there is so much data nowadays. Consider that in the early part of the 20th century, the amount of information created was doubling over the average person's lifetime. Today, it doubles every 3 to 5 years.
In this context, our decision process could be infinitely protracted. And there is a huge cost to delay. To quote Frank Ogden, a futurist philosopher, "Better Late than Never? — NO! Better Never than Late."

In a world preoccupied with risk management, we often find it difficult to forego options — to burn our boats, as Xiang Yu did. We hesitate before abandoning a cherished project, even when the market signals a hostile reception. Wishful thinkers, most of us, we find ourselves distorting signals to reinforce our original direction, or overlooking those that should steer us right.
And that is a tragedy. Because the insidious cost of not closing the door on a doomed project or an outdated product is that it saps our creative energy — it substitutes empty hope for tangible discovery, and creeping stagnation for the bracing tailwinds of progress.
Let me conclude my remarks by speaking to you from the heart.

I am humbled to stand with you as members of this noble and ennobling engineering community, a profession that has achieved so much for our society. Never before have our learning institutions produced so many professionals so well qualified to meet the needs of humanity. All of you have proven yourselves in that regard, and you are to be congratulated.
But I urge you to reach out beyond good, and strive to be great. I urge you to hold on to your identity as students. Cherish your capacity for wonder and revelation. Give full rein to your curiosity and let it take you where it may. Have the bravery to move outside your comfort zone and the open-mindedness to test your perspective against the ideas of other thinkers and the learnings of other fields. Be alert to the curse of knowledge and alive to the moment when we must cut our ties to the reassurance of the past.

If you do these things, you will be a great engineer. And I, personally, can think of nothing that I'd rather be.
So, here is a toast to you — with imaginary glass of wine in hand — and my most sincere congratulations.