Overcoming Obstacles to DoD Software Technology Transition

June 1, 2008

Air Force Research Laboratory

The term "Software Wind Tunnel" (SWiT) denotes an institution proposed to be established as a means of overcoming obstacles that, heretofore, have impeded technology transition in connection with research on, and development and utilization of, software needed for software- intensive systems of the Department of Defense (DoD). The term "technology transition" should not be confused with the term "technology transfer," which denotes a process in which an item of technology developed by or for the government is transferred to industry (usually) or vice versa. Instead, "technology transition" denotes a process in which an item of technology is made to evolve from its developmental form into a mature form and in which that item is adopted by its intended end users.

Experts in Diverse Specialties involved in technology transition would work together within the SWIT.

The SWiT would serve as both a collaboration environment and a shared experimental testbed that would bring together participants from different communities — researchers, developers, and domain experts — to defragment the knowledge necessary for achievement of technology transition. Some of the participants would provide challenge problems; some would provide candidate solutions to challenge problems; some would perform experiments; some would collaborate on challenge problems, candidate solutions, and/or experiments; and some would administer the SWiT infrastructure (see figure). Some participants might perform more than one of the functions described above. Overall, the SWiT would enable program engineers from DoD acquisition programs to interact with industrial and university researchers involved in the DoD's Software Intensive Systems Producibility Initiative (SISPI) to define, discover, and evaluate technology-transition opportunities.

The SWiT would provide hardware, software, and other infrastructure of an open environment for collaborative research and development to demonstrate, evaluate, and document the ability of novel software tools and methods to enable affordable and more predictable production of software- intensive systems. The SWiT hardware may be an aggregation of centralized resources, distributed resources connected via the Internet, and local (standalone) resources. The unifying element of SWiT would be its software suite. The software would enable distributed, interactive collaboration among users; integration of software tools; and measurement of, and collection of data on, performance.

There could be multiple instances of the SWiT, each dedicated to enabling collaboration within a specific community, which would typically be dedicated to a technological specialty (e.g., distributed real-time embedded weapons software systems). Given the distributed nature of users in such a community, the SWiT would be made remotely accessible via textual and other visual interfaces. The SWiT would be hosted and operated by the DoD or a designated contractor, but would be publicly accessible via the Internet, though only unclassified data would be available to the general public. Inasmuch as classified evaluations would be parts of technology-transition processes, users of SWiT could download its software tools to perform these evaluations within their classified program environments. Likewise, users could set up classified SWiT instances within their classified program environments.

This work was done by Richard W. Buskens, Patrick J. Lardieri, Bennett C. Watson, Jennifer Lautenschlager, and Douglas C. Schmidt of Lockheed Martin Corp. for the Air Force Research Laboratory.

AFRL-0071