DoD Software Engineering Science and Technology Summit
August 7-9, 2001, Los Angeles, CA

FYI: The Presentations are all available linked through the Agenda Page

University of Southern California
Los Angeles, CA

Context: To sustain its military pre-eminence, DoD needs strong and well-matched software engineering technology, practices, and management. Besides the general software engineering technology shortfalls identified in the 1998-99 PITAC Report and analyzed in two 1998-99 NSF workshops and the recent National Coordinating Office Workshop, DoD has additional software engineering challenges which need to be better addressed by software engineering science and technology. These include:

· More unconstrained and unrestrained adversaries, putting high premiums on software engineering techniques for superior security and survivability;

· A major culture change implicit in the transition from monolithic acquisition to evolutionary acquisition in the new 5000 series of acquisition regulations, surfacing many open questions on how best to reinvent DoD software engineering and acquisition practices, and how best to create the associated acquisition technology;

· The need to acquire and evolve very large, complex software-intensive systems with chronic shortage of expert in-house software engineering personnel, presenting challenges and opportunities for advanced training techniques, expert/non-expert collaboration techniques, and automated aids for non-expert software acquirers;

· Continuing constraints on DoD program and portfolio funding, acquisition practices, and management continuity, making many commercial best practices inapplicable and surfacing many open questions on how best to deal with such constraints.

· Major paradigm shifts in computational models toward massive complexes of autonomous mobile embedded devices and distributed processors, data, and decision makers.

For this Summit, "software engineering S&T" is broadly defined to include software product S&T, software process S&T and software management S&T. It also includes the full needs-driven S&T provisioning chain from basic research through technology transition.

Objective: In this context, the Summit will develop findings and recommendations for future DoD software engineering S&T strategies. Example questions to address are:

· What are the most critical DoD software engineering S&T needs (places where people in the DoD community reach on the shelf for needed S&T and find that it either isn't there or doesn't fit)? How are these needs likely to evolve in the future? Are the most critical needs in software product S&T, software process S&T, or software management S&T?

· How relevant are the findings and recommendations of the NSF and NCO software engineering S&T workshops to DoD needs (see the attached summary of the NSF S&T workshops)?

· Which of the needs are most likely and least likely to be satisfied by evolving commercial technology?

· What are appropriate near-term, mid-term, and long-term DoD strategies for addressing the S&T needs (including technology transition)?

The results will be used by OSD to help redefine DoD's software engineering S&T strategy and plans.

Approach: A 2½ day workshop, to be held at the USC Davidson Conference Center Los Angeles, CA, Tuesday-Thursday August 7-9, 2001. Attendance will include DoD, industry, related government, FFRDC, and academic personnel, and will be restricted to about 40 people.

Organizing committee: Dr. Delores Etter, DDR&E (acting); Dr. Charles Holland, DUSD/S&T (acting); Dr. Jack Ferguson, Director, OSD Software Intensive Systems; Richard Turner, OSD Software Intensive Systems; Dr. Stephen Cross, Director, SEI; Dr. Christine Davis, Chair, SEI Board of Visitors; LG(ret) Peter Kind, IDA; Dr. Barry Boehm, Director, USC-CSE (chair)