Alyssa Goodman

As it turns out, a software tool called 3D Slicer developed at Brigham and Women's Surgical Planning Lab-normally used for guiding biopsies and craniotomies in the operating room-can help astronomers visualize in three dimensions the inner cores of clouds that spew forth stars. The visualization is better than what's available through current astronomical methods.

That kind of shared need that crosses scientific borders and requires massive data crunching is at the heart of the new Harvard-wide Initiative in Innovative Computing (IIC), explained Alyssa Goodman, professor of astronomy at Harvard University. She delivered a Dean's Distinguished Lecture on "The Future of Scientific Computing at Harvard" on May 16 in Snyder Auditorium.

Goodman directs the IIC and is a member of the University's Task Force on Science and Technology. She is also the thesis advisor to Borkin, whose work serves as an early proof-of-concept technology for IIC. Goodman learned of 3D Slicer last fall from Michael Halle of the Surgical Planning Lab, when they both attended an NIH-NSF conference.

IIC grew from a report last year of the Task Force on Science and Technology, which includes HSPH Professor Dyann Wirth. The Task Force identified key trends in the sciences, including the decreasing importance of disciplinary boundaries; the need for large collaborative efforts; and the importance of large databases and powerful computing tools that can analyze that information.

"We looked at a lot of projects for what the future of Harvard might be about in terms of science," said Goodman. "And this one, having to do with the common, shared challenges in research scientific computing across the University, came out as one of the ones that was going to be pursued."

The purpose of IIC is to undertake specific, well-defined projects, proposed by groups of Harvard researchers, that require extraordinary computing-related capabilities and that produce results to be shared across disciplines.

"More and more research goes on like this-where you have an automated, high-throughput data-acquisition system, and then you have to deal with the gigantic volume of data that is created," said Goodman.

The current plan is to organize projects according to six branches: visualization technologies; distributed computing, involving large collaborations over the Internet; databases and data provenance; analysis and simulations; instrumentation; and education and outreach. (Goodman hopes that, as part of IIC outreach and education efforts, a new museum can be built at Harvard showcasing the kind of scientific work enabled by computing made possible through IIC).

There will be a call for project proposals this summer, said Goodman. The proposals will be vetted in a two-step process. First, a program advisory committee will select candidates, and then a team comprised of IIC leaders will judge whether the projects are feasible and whether they are interesting from a computer-science point of view. The proposals most likely to be selected early on are the ones that go beyond one particular traditional scientific discipline and that involve close collaborations between computer scientists and discipline scientists, noted Goodman.

A major goal of IIC is to attract top-notch computational scientists to Harvard. Far more prevalent in industry, these professionals bridge the needs of traditional academic researchers and pure computer science theorists.

There is also a large educational component to IIC, which will offer courses focused on data-intensive science, as well as project opportunities for students and postdoctoral fellows at all levels. In addition, IIC has proposed a certificate program for graduate students who are enrolled in other Schools. Seminar series and colloquia will also be in the mix.

"As data sets in all fields get bigger and bigger, it's not that we expect everybody to become a programmer," said Goodman. "It's that we need more of these generally applicable tools and techniques, and we need to value the scholarship that's involved in building those tools."

For more information about IIC, visit http://cfa-www.harvard.edu/~agoodman/IIC/index.html. A webcast of Goodman's talk is available at http://www.hsph.harvard.edu/ddl/.

Call for Proposals

WHEN: This Summer

BY WHOM: Harvard-wide Initiative in Innovative Computing

WHAT KIND OF PROPOSALS:

IIC is looking to fund projects that:

• involve groups of researchers at Harvard

• require extraordinary computing-related capabilities

• require close collaborations between traditional scientists and computer scientists

• address an unsolved need that applies to more than one discipline

• is of interest to computer science

For more information, visit the web site available at http://cfa-www.harvard.edu/~agoodman/IIC/index.html