Software vendors rarely acknowledge their debt to research, indeed often are unaware of it, and rarely even appreciate the importance of such acknowledgement. The long lead times, and tortuous adoption paths, for software engineering research contributions also cloud perception of the actual source of popularly adopted software engineering technologies.

The purpose of this panel is to introduce the Impact Project to the community, and to engage the community in a broad ranging discussion of the project's goals, approaches, and methods. Some of the project's early findings and directions will be presented.

An analysis is made of the contribution of the temporal operators alone to the complexity of deciding temporal logics by suppressing the role the usual Boolean connectives play in determining lower bounds on the complexity of their decision procedures. Several temporal logics are exhibited which can state many properties useful in describing temporal systems and which restrict combinations of temporal and Boolean operators so as to be decidable in low deterministic polynomial time. It is also shown that relaxing any of the constraints placed on the syntax of these logics results in intractability, thereby demonstrating that there is a fine line separating tractably decidable sets of temporal formulas from intractable ones.>

The authors consider the essential properties of a multiparty interaction construct which serves as a primitive for interprocess communication and synchronization in distributed programs. It is claimed that more general constructs, which violate the suggested properties, are appropriate for abstraction but should not be seen as a communication primitive, and that both facilities are needed. Several acceptability criteria are posed for multiparty interactions, and various possibilities for constructs satisfying these criteria are presented. These include introducing a novel kind of nondeterminism within the assignments of an interaction, weakening the synchronization among the participants in an interaction, and varying the number of participants in order to provide a high-level treatment of fault tolerance.>

In a recent paper(1)the author has presented evidence that, contrary to several studies in the literature, certain software complexity metrics are not consistently sensitive to the application of program style rules. In the present paper, these results are summarized and extended to two additional metrics. The new results indicate both that current complexity metrics are improper indices of program quality, as measured by style, and that many commonly used style rules do not address questions of minimizing interprocedural information flow complexity.