Classroom

Professor Patrick Henry Winston ’65, SM ’67, PhD ’70

Usually, when I fly into Boston, I head straight to my car. This had been a particularly rough trip and I was hungry, so for the only time in my life, I decided to stop at Au Bon Pain in Terminal A for a cup of soup.

I was about half way through the soup when I heard a voice.

“Are you Professor Winston?”

“I think so,” I said, trying to be funny.

“I’m a postdoc at MIT. I recognized you because I’ve watched your `How to Speak’ talk several times.”

“Oh.”

“The version you recorded at Harvard.”

“Ah.”

“I’ve watched it because, well, times are tough and it is hard to get job offers.”

“Yes.”

“I’m on my way to give my first job talk now, in Europe.”

“Hmm.”

“Actually, I wonder if you would have a minute to look at my presentation?”

“Sure.”

As he was booting up his computer, I said, “Too many words; too many slides.”

“How do you know? You haven’t seen them,” he said.

“Will the sun set?”

The talk had problems, all readily fixed, but I was having trouble finding the job-wining slide, even after looking through the slides a few times. Then, there it was, in small type, buried in a wordy, cluttered slide.

One hundred fifty research labs were using software based on his research. “Aha,” I said, “Every research lab wants to be famous, and that happens if their people are famous. If 150 research labs are using your stuff, you’re famous, but they won’t know that unless you tell them.”

I told him to make a slide titled Used in 150 Research Laboratories and put it up front, making it the second or third slide, so that everyone would see it before fogging out in the technical detail.

He dropped by my office a few weeks later to report he got several offers. He took the one from the place he was headed when, unlikely as can be, he ran into me in Terminal A.

{ 0 comments }

Professor Patrick Henry Winston ’65, SM ’67, PhD ’70

My spring class centers on discussion, and because the class is generally oversubscribed, I have to hold a lottery on the first day. For fun, I always include a puzzle on the lottery form. Here is the one I used this year:

Consider an old-fashioned milk bottle, filled with old-fashioned milk and cream. Left alone, the cream rises into the neck of the bottle. Inside the bottle, at the bottom, is a pressure sensor. If you shake the bottle, mixing up the cream and the milk, does the measured pressure increase, decrease, or stay the same? Explain.

When the students were well into that part of the questionnaire, I announced, “Oh, by the way, you can work together on the puzzle.”

Soon, as I expected, many students started talking and gesturing with their hands. These students tended to have better answers and explanations. I wasn’t surprised, because we humans think with our hands and we think with our mouths.

Still, of 80 students, none got it completely right, which didn’t surprise me either, because we think with our eyes, too, and few of the students drew pictures. I don’t know why; I could never solve physics problems without drawing a picture, and I don’t see how you can solve the milk-bottle problem without drawing a picture. Maybe, when tablet computers get a lot better, students will start to draw again.

What’s the answer? Here is a hint: draw a picture, then start taking parameters to extreme values. You’ll see in the solution that the pressure changes. My unchecked quantitative answer for ordinary sizes and densities is about ½%.

So, one reason computers make us stupid is that they discourage drawing. Another reason is that they encourage looking up answers, rather than practicing the art of heuristic problem solving. And of course, you don’t learn much if you are surfing the web or answering email rather than listening, writing, drawing, looking, and gesturing.

Computers, including IBM’s new Watson system, won’t be really smart until they think, like we do at our best, with full engagement of language, motor, and visual faculties. So understanding the contributions of those faculties is what I work on.

{ 12 comments }