Cecilia Aragon is on a mission to help scientists organize and manage the tottering piles of data they gather for their research. "This is a fundamental problem that we're facing across all fields of science," says Aragon, a computer scientist in the Lawrence Berkeley National Laboratory's Computational Research Division in California. She uses the number-crunching power of computers to represent abstract information visually, creating tools to help scientists recognize the essential patterns.
Outside the lab, Aragon is on a different kind of recovery mission, using her own story to help students and colleagues uncover hidden potential. Growing up the child of Latino immigrants in a nearly all-white Indiana community, she experienced the kind of discrimination, subtle and overt, that she says can keep people down. Despite an extraordinary record, and over the objections of her adviser, she abandoned her doctoral studies. "I was convinced … I just wasn't smart enough to write my thesis and complete my Ph.D.," she says.
Aragon turned things around, and she did it in a very interesting way. She learned to fly--literally, not metaphorically, and not as a mere recreational flyer but as an award-winning stunt pilot. Flying gave her the confidence she needed to return to science--14 years after abandoning the profession. Today, Aragon is one of a very small number of Latino women working in computer science. She mentors women and minority students from middle school through college, and she helped establish a support network for Latinas working in the computing field. And she just won a PECASE award, the Presidential Early Career Award for Scientists and Engineers.
"As a child, I was terribly shy and afraid of almost everything," Aragon says. A relentless stream of negative messages reinforced those fears. Her father, a native of Chile, was a professor of theoretical physics at the local university. Her mother was a guidance counselor. She remembers local merchants refusing to serve her family, a realtor taking a home off the market when her parents made an offer, and teachers overlooking her accomplishments and favoring inferior students.
Aragon was a smart, well-behaved, and conscientious child. She excelled in math, carrying a straight-A average in that subject throughout her schooling. Yet she recalls being assigned to the "slow" groups at the beginning of each school year. "I remember getting 'the look' when these teachers handed back the first exam," she says. "The teachers looked completely surprised that I had done so well."
Once, a science project--a rock collection--was sabotaged; all the rocks were torn off the display. In middle school, she was beaten by a group of boys who taunted her with racial epithets.
She worked hard in other subjects, too, but says, "My teachers were always finding something wrong with my writing or art." In elementary school, she received average marks in writing and art.
And yet she was, in fact, a gifted writer. Her English teacher during her freshman year of high school suggested she consider writing as a profession. "It was the first time a teacher, somebody outside my family, had ever told me I was good at something," she says. She entered a national essay contest and won first prize. She later won a poetry contest, too. "Part of the reason I went into math and science is because when you get the answers right in math, the teachers had to give you an A," she says.
Yet not all of her teachers were supportive, even in math. During high school, Aragon had the highest math grades in the school. Her SAT score was the highest in the state that year, yet a teacher chose other students to compete in the International Mathematical Olympiad. "I didn't know about the Math Olympiad, and no one had ever mentioned it to me during high school," she says.
At the time, Aragon "didn't have a clue" what was going on. Only years later, when she read a book called Life on the Color Line: The True Story of a White Boy Who Discovered He Was Black, did she recognized the pattern of discrimination and discouragement that she had experienced as a child.
"My parents were overwhelmingly supportive, but they just were not aware of the pervasive nature of discrimination that I and my sister were exposed to in school--by teachers, the administration, by the other kids," she says. "We grew up basically being told that we were inferior and we weren't good enough."
Going to California
Despite these challenges, Aragon graduated third in her class of 200 and went on to study mathematics and literature at the California Institute of Technology. At Caltech, she made friends quickly and enjoyed a rich social life. She did very well, graduating with a degree in math--and with straight A's in all her courses. Yet she thought, "the fact that these problems are hard means that I'm not really very smart," she says today. "I simply never had a chance to develop confidence in myself."
Aragon started her doctoral studies in computer science at the University of California, Berkeley, the following fall. After a few years, she switched advisers to focus on theoretical aspects of computer science for a Ph.D., working with Raimund Seidel, who had recently joined the Berkeley faculty from Cornell University. She finished the coursework and passed the preliminary exams--but in 1987, she decided to take a master's degree and leave the program. "I remember my major adviser telling me I shouldn't leave because my thesis work was very good," she says. "But I had so little self-confidence that I thought he was just being nice."
Learning to fly
While Aragon was in graduate school at Berkeley, she worked summers at the Digital Equipment Corp. in Palo Alto, California. She was working there one summer when a co-worker offered to take her for a ride in a Piper Archer, a four-seat airplane. Aragon almost turned the offer down because she considered small planes dangerous. Once airborne, her co-worker handed her control of the airplane. At first she was terrified, but soon "we flew out over the Golden Gate Bridge and over the coast, and I got to manipulate the plane and steer it," she recalls. "The experience evoked a real passion in me."
"As a child, I had this dream that I could fly, and I used to imagine myself flying through the air. It never occurred to me to do it using an airplane. "
She signed up for flying lessons as soon as they landed. But learning to fly was neither smooth nor easy. Aragon remained terrified of heights. She went through three flight instructors before finishing her training. In contrast to her schooling, this time, she says, her teachers' lack of confidence in her was justified. "I think my instructors kind of gave up on me because I was so hopeless," she says, laughing. "But I loved flying so much that I had to keep doing it." She worked two jobs to support her habit.
She earned her pilot's license and went on to become a flight instructor and an airline transport pilot. Then she became an aerobatic pilot--think spins, stalls, etc.--winning a coveted spot on the U.S. Aerobatic Team and twice representing the United States at the World Aerobatic Championships, the Olympics of flying. In 1994, her team won a bronze medal, and she won an award given to the top individual performers. She established an aerobatic school specializing in students who were afraid to fly.
Landing in the lab
After 10 years of flying, Aragon began to miss the intellectual challenges of science. She put the word out among her friends that she was looking for research work, and someone introduced her to a group at NASA that was working to develop a data-visualization system for wind tunnels. "They were studying how air flows over aircraft, and I thought, 'This would be a great opportunity to combine my talents and interest,' " Aragon says.
The work made her yearn even more for a return to real research. "Flying taught me that fear is not an intrinsic part of you; it can be overcome," she says. "I now knew that it had just been my lack of confidence and my fear that had stopped me from achieving what I really wanted to do."
She returned to Berkeley and completed her Ph.D., working with Marti Hearst. She finished her thesis in 18 months.
Diversity and computer science careers
Despite an apparent decline in opportunities for information technology professionals in recent years, the job market for computer science Ph.D.s remains strong, with an unemployment rate below 1%, according to the latest Taulbee Survey from the Computing Research Association (CRA), which was released in May. Available faculty posts, including tenure-track positions, increased year over year at a 5% clip. In the last application cycle covered by the survey, one in four faculty openings went unfilled. Meanwhile, a dramatic mid-decade drop-off in jobs in industry for computer science Ph.D.s has completely reversed itself; despite the strong growth in academia, today a higher percentage of Ph.D.s are employed in industry than was the case during the industry's previous peak during 1996–7.
Diversity in the computer science professoriate, meanwhile, remains poor. Last year, just two out of 192 new tenure-track faculty hires were Hispanic. That sounds awful--until you realize that those two new hires were then the ONLY Hispanic computer science assistant professors in the country and that there are NO associate professors and only four full professors, according to the survey.
So, what about the future supply? New computer science Ph.D.s reported in the latest Taulbee Survey totaled 1597 students, of which 1454 reported their ethnicity. Of that group, 21--or 1.4%--were Hispanic.
CRA should be commended, by the way, for having the best work force statistics in all of science.
When she finished her doctorate in 2004, Aragon had seven offers, most from research labs in industry and government. She accepted an offer from a group of astronomers at Berkeley, part of a collaboration studying a rare type of supernova that some believe holds the key to measuring the expansion of the universe.
Greg Aldering, a member of the group, says the team was looking for someone to help manage the massive amounts of data taken in each night. Their images included artifacts from airplanes, stars, even their own instruments. The group wanted to flag those artifacts in the 30,000 nightly images. Aragon's experience handling different kinds of data, and her contacts in industry, allowed her to "quickly build a good team of people to help with the implementation," Aldering says.
Aragon and her team worked with the scientists to develop novel machine-learning algorithms to sort the supernovae from the background noise. A task that used to take six people 12 hours a day was whittled down to 30 minutes of one person's time. "That cut down our amount of work by something like a factor of 100," Aldering says.
Aragon and her team went on to develop organizational tools for the group, making it easier to pull up, sort, and share the information gathered at various sites. Aragon's efforts, Aldering says, took the scientists from a "state of paralysis" to a point where they were able to analyze the data as it came in.
Teaching others to fly
Today, Aragon continues to teach flying lessons through her aerobatics school. Three years ago, she helped forge a group called Latinas in Computing, which serves as a support network for Hispanic women working in all levels of computing. She frequently volunteers to teach and mentor women and minority science students from middle school through college. She reaches out to others who feel isolated because of their cultural background, race, or gender, striving to instill in them a sense of mastery and confidence.
"Discrimination still exists, and girls going into science are likely to experience some of the similar kinds of discouragement that I experienced," she says. "Whether it's overt or … subtle, it still destroys your confidence. And confidence is critical when you're going to be a scientist."
Aragon's efforts were formally recognized in July when she received a PECASE. The awards committee praised not just her scientific work but also her dedication to community service and her efforts to advance diversity in the computing community.
"If we want to have the next generation of mathematicians and scientists, and if we want to train them here in the United States, we need to encourage everybody," Aragon says. "If my parents hadn't been supportive, I would have been lost forever, and whatever contributions I could have made would not have been made."
Susan Gaidos writes from Portland, Maine.