INDUSTRY

For the second year, NCSA participated in the Google Summer of Code, a program that offers student developers stipends to write code for various open-source projects. Four students from around the world worked with NCSA mentors this summer.

"Google Summer of Code has been a great way for us to attract highly motivated students from around the world," said Jim Myers, leader of NCSA's Cyberenvironments and Technologies Directorate. "We're one of a relatively few groups that participate in the program with projects geared directly toward scientific software. These projects can be challenging, given the additional knowledge of science and mathematics required, but they expose students to the excitement of being part of research that is working to understand fundamental laws of nature, improve our lives, or both."

Students who worked with NCSA included:

• Anthony Chansavang, a recent graduate of the University of Technology of Belfort-Montbéliard in France. Anthony's project, supervised by Alex Yahja, Robert McGrath, and Guy Garnett, aimed to create a standards-based common interface for a large-scale agent-based social-network simulation applied to epidemiology with the virtual world environment mWorlds. This combination provides researchers with a visualization of their data and an intuitive, easy way to interact with their simulation. "I have learned a lot this summer," Anthony says. "Not only have I sharpened my technical skills, but I also learned a lot about working on big projects with other people's code and about managing time and effort."
• Daniel Long, University of Tennessee at Chattanooga. Mentored by Kenton McHenry, Daniel worked to provide 3D file format support within Java using the JOGL OpenGL bindings and a simple mesh data structure. "This is beneficial for people who are working with a variety of these formats," he explains, "as it allows them to load a model regardless of its format into a simple mesh data structure, granting easy access to the vertex coordinates, facial information, material data, and texture coordinates, without having to hunt down and try to interpret various complex specifications."
• Néstor Morales Hernández, Universidad de La Laguna on the Spanish island of Tenerife. Néstor and mentors Rob Kooper and Peter Bajscy worked to create an algorithm to automatically stitch all the images in a dataset into a single larger image based on geo-referencing. The work was based on a corpus of aerial images of Costa Rica and creates images as large as 30,000 x 50,000 pixels.
• Yigang Zhou, Wuhan University, China. With Myers as a mentor, Yigang worked to integrate two NCSA software projects: Defuddle, which pulls information locked in binary formats into standardized, well documented models for preservation, and Tupelo, a data and metadata management system based on semantic web technologies. The combination allows information within files to be combined with information about the files to create rich research records that can last beyond the lifetime of individual applications and file formats. Yigang's work automates and simplifies the invocation of Defuddle as part of overall data management.