Abstract. Software must be constantly adapted to changing requirements. The time scale, abstraction level and granularity of adaptations may vary from short-term, fine-grained adaptation to long-term, coarsegrained evolution. Fine-grained, dynamic and context-dependent adaptations can be particularly difficult to realize in long-lived, large-scale software systems. We argue that, in order to effectively and efficiently deploy such changes, adaptive applications must be built on an infrastructure that is not just model-driven, but is both model-centric and contextaware. Specifically, this means that high-level, causally-connected models of the application and the software infrastructure itself should be available at run-time, and that changes may need to be scoped to the run-time execution context. We first review the dimensions of software adaptation and evolution, and then we show how model-centric design can address the adaptation needs of a variety of applications that span these dimensions. We demonstrate through concrete examples how model-centric and context-aware designs work at the level of application interface, programming language and runtime. We then propose a research agenda for a model-centric development environment that supports dynamic software adaptation and evolution. 1

Reasoning about object-oriented programs is difficult since such programs usually involve aliasing, and it is not easy to identify the ways objects can relate to each other and thus to confine a program’s heap. In this paper, we address this problem in the context of a relationship-based programming language. In relationship-based programming languages, relationships are first-class citizens and allow a precise description of inter-object relationships. Relationships enforce a modularization discipline that is closer to the natural modularity inherent to many problem domains and that yields, as a result, program heaps that are DAGs. We further describe a mechanism, member interposition, that leverages the new modularization discipline and supports encapsulation of fields of shared objects. We have implemented the described modularization discipline and the mechanism of member interposition in the context of Rumer, a relationship-based programming language with support for contract specifications. We discuss the implications of member interposition for the modular verification of object invariants with an example. Relationships and interposed members provide an alternative to ownership type systems.

Abstract. Relationships capture the interplay between classes in object-oriented programs, and various extensions of object-oriented programming languages allow the programmer to explicitly express relationships. This paper discusses how relationships facilitate the verification of multi-object invariants. We develop a visible states verification technique for Rumer, a relationship-based programming language, and demonstrate our technique on the Composite pattern. The verification technique leverages the “Matryoshka Principle ” embodied in the Rumer language: relationships impose a stratification of classes and relationships (with corresponding restrictions on writes to fields, the expression of invariants, and method invocations). The Matryoshka Principle guarantees the absence of transitive call-backs and restores a visible states semantics for multi-object invariants. As a consequence, the modular verification of multi-object invariants is possible. 1