Teen's left, right brains both work in overdrive

Polytechnic senior is finalist for prestigious science award



When Myers Abraham Davis decided he wanted to spend two years buried in a lab, laboring over a sophisticated computer science project, he had never so much as taken a computer programming class or written one line of a program. The teenager couldn't type without looking at his hands.

In other words, Abe was just being Abe.

People who know the Polytechnic Institute senior say he loves challenges that carry a whiff of impossibility - and he's very determined.

If this were a movie, there would be a montage at this point set to a pulsating soundtrack. Starting with a book on the basics, Abe taught himself computer programming, ambled into the field of computer graphics, connected with an academic mentor at the Johns Hopkins University, raised funds for his own high-powered computer, overcame obstacles and spent untold hours toiling in the basement of the university's New Engineering Building.

Crescendo here: His hard work paid off Hollywood-style, when he was named one of 40 finalists in the Intel Science Talent Search in January. The prestigious award, a kind of mini-Nobel Prize for the AP set, comes with a top prize of a $100,000 college scholarship. The winner is to be announced Tuesday.

This is only the second time in 50 years - last year was the first - that a Baltimore student has been a contest finalist. In both cases, the students came out of the Ingenuity Project, a scholastic achievement program for public school students gifted in mathematics and science. (Another local student, Jeffrey Chunlong Xing of Clarksville, is also a finalist this year, and a third Ingenuity student, Owen Forgione Hill, was a semifinalist.)

"I just thought it was fascinating," Abe said of his journey into the thorny wilds of computer science. "Computer graphics in general seems like a nice combination of science, math, technology and art. ... I like the idea of having to do something technological but that has an artistic side."

A week before the start of the competition, the thoughtful and straightforward 17-year-old was tucked in a corner of the lab, putting the finishing touches on a contest poster. With thousands of dollars at stake, he said, it's not exactly like completing a school project.

Abe, the third and youngest son of two lawyers, has a sweet grin and the complexion of someone who spent his summer in the sun-deprived lab. He was dressed neatly enough in wire rim glasses, black jeans and a white, short-sleeve shirt, but those who know him say it's not unusual to find Abe in a whirl of papers, with his glasses askew and hair in a scramble.

"He's such a teenage boy. He's disorganized and whatever. But he's also amazingly brilliant," said Charlotte Saylor, a former science teacher. "In that outside chaos is this understated genius."

At 10:30 a.m., claiming to still be half asleep, Abe sipped a Pepsi and patiently explained "collision detection for physical simulation." For mere mortals, the standard English version of what Abe has managed to do goes something like this:

"If I have a tennis ball and throw it as hard as I can at a wall, what happens?" Abe started.

To figure out the answer, one has to picture the objects, recall their properties and employ basic physics. In other words, he has do a physical simulation in his head.

That's what Abe is teaching a computer to do, except in his example it's 64 or 125 Ping-Pong-like spheres dropping from the sky into what looks like a box with a doughnut in it. His program is "guaranteed" to detect every single possible collision, he said. Plus, it looks cool.

For this very complex computation, Abe is relying on relatively inexpensive, commonly available hardware known as a graphics processing unit or a graphics card. Such cards can perform parallel rather than sequential processing, which Abe describes as the difference between one person and 20 people painting a fence at once.

Usually associated with video games and movie special effects, computer graphics can be applied to any number of scientific and research fields. Physical simulations like Abe's could be used, for instance, to develop medical tools or to make cars safer by studying crashes. His project demonstrates a potentially cheaper, faster way to conduct these sorts of simulations using a personal computer and graphics card in a novel way.

His father, Michael H. Davis, described himself as unimaginably proud and equally confused about Abe's work. "It's far, far beyond my understanding. ... I tried to read his report, and frankly I can't digest it," he said.

"I certainly wasn't doing anything like that when I was in high school," said Abe's mentor, Johns Hopkins computer science professor Jonathan Cohen. "Abe definitely keeps you on your toes."

Baltimore Sun Articles
Please note the green-lined linked article text has been applied commercially without any involvement from our newsroom editors, reporters or any other editorial staff.