The purpose of the COSMIC ( Complementary Open Software for Manipulation of Imagery for Communities )project is to explore the interface between commercial software packages and open source software tools . The project is motivated by the expectation that the synergies between open source and commercial software will allow researchers to more efficiently move algorithms described theoretically in journal articles into working software. Strategically the three main goals of the project are to: (1) identify, develop, test, refine, and demonstrate these synergies using real world examples from remote sensing research, (2) explore ways how the open source community can support users that depend on commercial software packages and (3) show ways how it can be in the self interest of commercial vendors to support open source development. At the ‘tactical’ level particular emphasis is being placed on using Python and NumPy as “glue” to tie together commercial and open source software. Initial results of using Python with the Leica Imagine commercial software package are encouraging. For instance, since Python is a high level language it is possible to quickly build on researchers past programming experience to program new algorithms. At the same time the powerful Imagine tools allow the researcher to quickly display, further process, and analyze the results. Furthermore, once an abstract journal description of an algorithm is translated into Python, the algorithm: (1) can be easily translated into a lower level language (e.g. C or FORTRAN) for speed, (2) can be explained to a wider audience,(3) moved between operating systems and/or (4) more easily incorporated into a commercial system.
While initial major emphasis is on algorithms dealing with image processing, GIS methodology and physics associated with remote sensing research it is expected that the COSMIC project experience will provide important insights for other disciplines that do, or can, make extensive use of commercial software