Final Report: An Object-Oriented Integrating Framework for Multi-discipline Ecosystem ModelingEPA Grant Number: R827959
Title: An Object-Oriented Integrating Framework for Multi-discipline Ecosystem Modeling
Investigators: Stotts, David
Institution: University of North Carolina at Chapel Hill
EPA Project Officer: Packard, Benjamin H
Project Period: October 1, 1999 through September 30, 2002 (Extended to July 20, 2005)
Project Amount: $863,049
RFA: Computing Technology for Ecosystem Modeling (1999) RFA Text | Recipients Lists
Research Category: Environmental Statistics
The original proposal was to construct an object-oriented (OO) framework in which scientific models, such as the ones built and studied by U.S. Environmental Protection Agency (EPA) researchers, could be combined easily into larger models of more complex systems, to solve problems that could not be solved with individual models.
When the grant was awarded, we discovered that a local EPA group was beginning a project within EPA to accomplish the same thing. This system is now known as MIMS and the effort was led by Steve Fine in the Research Triangle Park facility. Rather than duplicate those efforts, I decided to take our research in complementary directions and to participate in the MIMS development and planning meetings.
The objectives of our research project were to:
- Develop analysis methods for determining if OO program architectures are “good” or adhere to standards and best practices. This emphasis led to a unique approach based on use of a formal theorem prover, a toolset called SPQR that embodies these methods, and a patent application.
- Develop a framework for model interconnection but from a different technical perspective than the style taken by MIMS. Our early approach was functional, in the programming language Haskell. Recent approaches have focused on Aspect Oriented Programming and support for grid-based execution of models.
- Investigate technology that will assist the considerable intellectual communication and information exchange that must go on for collaborating scientists to construct these “group” models. This has led to prototype systems called Facetop that lend unique video support to paired remote collaborations, as well as video-based assistive technology based on Facetop that support hearing-impaired computer users and collaborators. This work also produced a second patent application and a best paper award at the 2004 Agile Software Development conference.
Our original intent was to collaborate with scientists at the North Carolina Supercomputing Center, specifically those who had developed a hand-written model federation for weather forecasting. Early in the grant period, the North Carolina Supercomputing Center changed its mission and reorganized significantly, resulting in the disbanding of the group with which we were seeking to work. We also had difficulty getting permission to work on the model codes they had developed. Given these early problems, we altered our original intentions as stated in the proposal and worked with environmental scientists at the University of North Carolina (UNC) at Chapel Hill and UNC-Charlotte, using their scientific models as a basis for our infrastructure research.
Prototype Software Developed
The following software systems were developed with this grant support:
- SPQR – A toolset for design pattern detection in OO programs.
- DeCo – A functional framework for interconnecting scientific models, using the language Haskell.
- JDeCo – A Java form of DeCo that is grid enabled.
- Facetop – A video-based support system for paired remote collaborations (Mac version and Windows version)
- FaceNote – A variation of the Facetop technology that runs on a Tablet PC and allows a deaf user to view an American Sign Language interpreter semitransparently onscreen while taking hand written notes.
SPQR is being reviewed by IBM for possible commercialization. Facetop and FaceNote are undergoing patent protection at UNC but are being released under specific license agreements to individual researchers in academia for user trials. DeCo and JDeCo are no longer supported directly, as we currently are investigating better approaches to the problems those systems were designed to solve.
Students Supported by the Grant
These students were supported in part over the lifetime of the grant to work on the research projects.
- Dean Herrington (M.S. student).
- Swaha Das (M.S. student).
- Dennis Jen (M.S. student).
- Jason Smith (Ph.D. student).
- Karl Gyllstrom (Ph.D. student).
- Dorian Miller (Ph.D. student).
- Uma Devi (M.S. student).
- Angus Antley (M.S. student).