GALOIS, INC. — Department of Defense SBIR Phase I: HR001121S0007-30

Description

The overall vision of the CAMDEN project is to explore the use of mechanism design as a solution to incentivize the creation of collaborative API designs for the DoD.  APIs are not merely software interfaces: they enable a decentralized marketplace of services. It would be highly desirable to apply this decentralized API-driven approach to DoD missions. We believe the barriers to this vision are largely not technological: this world already exists in industry. Instead, the barriers involve the proper selection of incentives. The commercial world has had great success in applying mechanism design as a problem-solving technique: High-profile successes include the design of frequency spectrum auctions, which are constructed so that players have the incentive to reveal their true valuations of the items up for auction. For the auction designers, this accomplishes their objective of maximizing revenue.  Our proposed effort consists of four interlocking research themes that comprise the technical objectives for the project.     1.    Get it right in theory: understand how mechanism design works in theory for DoD API-relevant use cases.     2.    Get it right in implementation: develop a software-based simulation environment for the evaluation of candidate API design mechanisms.     3.    Get it right in practice: Develop simple “wargaming” scenarios with human participants who are encouraged to ‘game’ the system in order to test the ability of the mechanisms to induce the right player behaviors even when the system is perturbed and under stress.     4.    Get it right in evaluation: Develop appropriate metrics for the evaluation of the efficacy of incentive mechanisms This 8-month effort aims to provide clarity on the fundamental questions and issues involved in mechanism design, evidence of the feasibility of our approach, and a roadmap for further research leading to deployable API.