Position Papers
Home Call for Papers Position Papers Resources Results

 

RESAFE - International Workshop on Software Reuse and Safety

Thursday, 15 June 2006, Torino, Italy

 Held in conjunction with the Ninth International Conference on Software Reuse (ICSR 9)

 

Selected position papers from the RESAFE 2006 Workshop

A selection of position papers from among those presented during the workshop  is provided in this section.

Reuse, Reliability, and Safety – M. Tortorella

Software reuse has been advanced as a powerful tool that, in the best of circumstances, transfers known good qualities of a proven software product to a new product still in development. Presumably this would also apply to reliability and safety properties. A careful examination of the interactions between reuse, reliability, and safety is required to build confidence in this approach. (Paper)

Reuse and Safety  –  B. Frakes

The problems with reuse in terms of safety apparently arise from inserting an asset into a new system environment without sufficient understanding of the context required for its use, as was the case in the Therac accident. This raises several questions. 1. How might context be represented? 2. Is it possible to assure that the contextual specification is correct and complete?
(Position Paper)

How FMEA improves hardware and software safety and design reuse  –  N. Bidokhti

Once software is determined to be reusable for a desired application, it is important to perform the appropriate analysis to identify all possible failure modes associated with the design with respect to the new environment and its association with the components of the current architecture. One technique proven useful is Failure Modes and Effects Analysis (FMEA).
(Position Paper)

Issues in Object Orientation and Software Safety  –  J. Favaro

The object oriented approach to software development has become the most popular paradigm for software development today. Particular claims are made about its contribution to software reusability, through characteristics such as encapsulation, inheritance, and polymorphism. But it is claimed by many that those very characteristics make it infeasible to create verifiable safety critical systems.
(Position Paper)