In [14], Voas outlined 13 major software engineering issues needing further research: (1) what is software quality? (2) what are the economic bene is behind existing software engineering techniques?, (3) does process improvement matter?, (4) can you trust software metrics and measurement?, (5) why are software engineering standards confusing and hard to comply with, (6) are standards interoperable, (7) how to decommission software?, (8) where are reasonable testing and debugging stoppage criteria?, (9) why are COTS components so difficult to compose?, (10) why are reliability measurement and operational profile elicitation viewed suspiciously, (11) can we design in the "ilities" both technically and economically, (12) how do we handle the liability issues surrounding certification, and (13) is intelligence and autonomic computing feasible? This paper focuses on a simple and easy to understand metric that addresses the eighth issue, a testing and debugging testing stoppage criteria based on expected probability of failure graphs.
