Personal Statement Greg J. Badros Draft 2.00 October 27, 1994

My name is Greg, and I have a problem. It all started when I was seven. The local Radio Shack store was offering a night course on computer programming in BASIC. I was fascinated, and simply had to take the class. After learning all I could from the course, I wanted more. Through incessant begging, I finally persuaded mom and dad to get a home computer: the state-of-the-art Commodore Vic-20, complete with 3.5K RAM and tape drive. For the next six years, I learned almost everything there was to know about my new toy, using it to write games, draw graphics, and keep my personal baseball statistics. By the age of eight, I was addicted to computing (and had a .372 batting average in the pee wee league).

When I entered junior high school, I convinced my parents it was time for a new computer: a Commodore 128. Again, I was in awe of what I could do; there seemed to be no limits! I began to apply my computer skills to various other projects: I was editor of my junior high's newspaper, and helped my mom do the complex word-processing needed to publish her doctoral dissertation. My programming experience grew too. I quickly learned C and 8502 assembly language programming, and started taking computer science courses at Salisbury State University. In the summer after eighth grade, I had my first article and program accepted for publication by Compute's Gazette, a favorite magazine that I spent many rainy afternoons reading.

Despite a busy schedule going to high school, taking night classes at the University, tutoring friends in math and science courses, playing tennis as the top seed on my varsity team, and serving my peers in various student officer positions, I always made time for learning more about computers. As junior class president, I applied my programming abilities to my interest in helping my class make money. I founded Data Date, a computer personality matchmaking service for high school organizations. A year later, building on Data Date's success in my county, dozens of schools in Maryland were making happy couples with my fundraiser.

Now, at Duke University, my addiction still grows. I spend my free time playing ultimate frisbee, tennis, learning the piano, and of course, continuing to explore the world of computing. I've continued to refine my programming skills and placed third (first in the US) in the ACM International Collegiate Programming Contest last spring. To gain experience working in a group on a large project, I took a job at Microsoft this past summer, designing and implementing features for the Excel spreadsheet, while taking classes there to improve my knowledge of C++ and Windows coding.

To me though, computer science is more than just programming. Sophomore year I was the teaching assistant for an introductory programming class where I enjoyed helping my fellow Duke students appreciate Pascal. At the same time, their feedback gave me a much better grasp of which ideas are difficult for novices. To further develop my love of teaching and gain experience designing a course, I spent that summer working at the Johns Hopkins University Center for Talented Youth program as a TA for both the Digital Logic and Data Structures & Algorithms classes. The best way to understand a concept is to explain it to someone else preparing and giving lectures at CTY was a challenging and exciting way to push my abilities, and improve my teaching style.

My research interests have provided yet another means for expanding my horizons. After a fascinating discussion with a professor friend over Christmas break, I began researching Genetic Algorithms in my first independent study. I plan to continue that research this spring by completing the implementation of a package to help me test the effectiveness of an advanced variant of domainindependent meta-level genetic operators. Last year, I began another project with a professor and graduate students developing compression techniques to analyze and predict international stock markets. My background in economics proved invaluable to our team, and the project has given me some interesting ideas which I continue to investigate on my own.

Graduate school will give me the opportunity to attack two other problems which pervade computing. First, the majority of society is completely computer-illiterate. Most people who use computers are very limited in what they know how to do. More tragically, though, many don't have a clue about even what tasks can be simplified. Modern applications have tremendous capabilities, most of which go unused. Powerful software needs to not only step the user through complex tasks, but also alert her when she could use an undiscovered feature. A computer should act as an intelligent friend, assisting in daily work, and adapting to its users, instead of forcing them to learn its semantics and idiosyncracies. Second, although computing power continues to double every two years, software advances lag far behind. Modern programming tools, environments, and APIs are stagnated by incompatibility and steep learning curves. Re-use of code is still the exception, and mundane programming tasks are all too often performed by expensive software developers. In short, to accompany Moore's law, I believe in Badros's law: software should get twice as good every two years. I look forward to tackling these and other issues at XXXX university while earning my Ph.D., so I can further advance the computer industry through my students, and their contributions.