Updating the Recipe Book

A chef can spend years finding the right ingredients and steps to create a perfect dish. We would think she was crazy if she never wrote down the recipe to repeat her masterpiece after all that work. Yet, we don't bat an eye when scientists publish their results without explaining the data and calculations they used in enough detail to replicate and validate the results. My mentors, Barbara Lerner and Emery Boose, are working on a program that creates a kind of recipe from the code that scientists write to manipulate data. The program creates a visual representation of the code in the form of a DDG, or Data Derivation Graph. My job is to make our graphs easier and more helpful for scientists who want to analyze the way their code works with their data.

I spent the first three weeks of this summer reading through Java code (there is a lot of it) and playing with our Data Derivation Graphs to see what features to add. Then, with the help of Barbara, Emery, and my partner Luis, I created a wish list of almost 30 capabilities we want to add to our graphs. It’s impossible for me to complete everything on the list before I have to leave, but I am proud of the work that I have completed so far. I added the beginnings of a toolbar and smoothed out the look of the visualization through anti-aliasing. Then I spent almost three weeks creating the Overview and its Viewfinder. The Overview sits on the side of the DDG and shows a zoomed-out view of the graph. The Viewfinder rests inside the Overview and frames what part of the DDG you are looking at in comparison to the entire graph, kind of like a You Are Here sticker on a map.

We are rounding up week eight out of eleven, and I want to make the most of the time I have left, so my last project is a big one. I am going to streamline the entire user interface of the program into one window. First, I have to plan out how to best integrate all of the parts into a window while keeping everything accessible and intuitive for a user. Then I’ll get into the code and start changing the way our separated windows depend upon each other.

There are tons of new things I have learned here at Harvard Forest; some are expected and some aren’t. I’ve learned to trust myself when faced with more than 6,000 lines of foreign code. I’ve also enjoyed keeping a detailed log of my work each day, so that I can see all of my hang-ups and successes as the days progress. I also learned to appreciate the boundaries of work and home. In computer science, sometimes it is easy and fun to take work home with you, but at Harvard Forest, I realized that I am most efficient when I close my laptop at 5:00 each day. Even when I am disappointed in my progress, I return the next morning with more vigor and clarity when I let everything go for a few hours. Besides, it is so easy to fill my nights with soccer, telenovelas, and night walks in the rain with my beautiful friends.