Done well, separation of concerns can provide many soft-ware engineering benefits, including reduced complexity, im-proved reusability, and simpler evolution. The choice of boundaries for separate concerns depends on both require-ments on the system and on the kind(s) of decompositionand composition a given formalism supports. The predominant methodologies and formalisms available, however, support only orthogonal separations of concerns, along sdngle dimen-sions of composition and decomposition. These characteris-tics lead to a number of well-known and difficult problems. This paper describes a new paradigm for modeling and im-plementing software artifacts, one that permits separation of overlapping concerns along multiple dimensions of composi-tion and decomposition. This approach addresses numerous problems throughout the software lifecycle in achieving well-engineered, evolvable, flexible software artifacts and trace-ability across artifacts.

: Separation of concerns is at the core of software engineering, and has been for decades. This has led to the invention of many interesting, and effective, modularization approaches. Yet many of the problems it is supposed to alleviate are still with us, including dangerous and expensive invasive change, and obstacles to reuse and component integration. A key reason is that one needs different decompositions according to different concerns at different times, but most languages and modularization approaches support only one "dominant" kind of modularization (e.g., by class in object-oriented languages). Once a system has been decomposed, extensive refactoring and reengineering are needed to remodularize it. Multi-dimensional separation of concerns allows simultaneous separation according to multiple, arbitrary kinds (dimensions) of concerns, with on-demand remodularization. Concerns can overlap and interact. This paper discusses multi-dimensional separation of concerns in general, our particular approach to providing it, called hyperspaces , and our support for hyperspaces in Java^TM , called Hyper/J^TM. 2 Chapter # 1.

Requirements such as distribution or tracing have an impact on multiple classes in a system. They are cross-cutting requirements, or aspects. Their support is, by necessity, scattered across those multiple classes. A look at an individual class may also show support for cross-cutting requirements tangled up with the core responsibilities of that class. Scattering and tangling make object-oriented software difficult to understand, extend and reuse. Though design is an important activity within the software lifecycle with well-documented benefits, those benefits are reduced when cross-cutting requirements are present. This paper presents a means to mitigate these problems by separating the design of cross-cutting requirements into composition patterns. Composition patterns require extensions to the UML, and are based on a combination of the subjectoriented model for composing separate, overlapping designs, and UML templates. This paper also demonstrates how composition patterns map to one programming model that provides a solution for separation of cross-cutting requirements in code---aspect-oriented programming. This mapping serves to illustrate that separation of aspects may be maintained throughout the software lifecycle.

Aspect-oriented programming (AOP) controls tangling of concerns by isolating aspects that cross-cut each other into building blocks. Component-based programming (CBP) supports software development by isolating reusable building blocks that can be assembled and connected in many different ways. We show how AOP and CBP can be integrated by introducing a new component construct for programming class collaborations, called aspectual component. Aspectual components extend adaptive plug-and-play components (AP&P) with a modification interface that turns them into an effective tool for AOP. A key ingredient of aspectual components is that they are written in terms of a generic data model, called a participant graph, which is later mapped into a data model. We introduce a new property of this map, called instance-refinement, to ensure the proper deployment of components. We show how aspectual components can be implemented in Java, and demonstrate that aspectual components improve the AspectJ language for AOP from Xerox PARC.

In this paper, we investigate whether module concepts for capturing multi-object collaborations can be effectively used to implement crosscutting concerns in reusable, independently developed modules for a-posteriori integration into existing systems. A new kind of collaboration module, called Object Teams, is proposed which combines the best features of existing approaches, further enhances them with concepts for expressing crosscutting relations between independent collaborations, and facilitates a-posteriori integration of such collaborations into existing systems.

There is a well documented problem in the software engineering field relating to a structural mismatch between the specification of requirements for software systems and the specification of object-oriented software systems. The structural mismatch happens because the units of interest during the requirements phase (for example, feature, service, capability, function etc.) are different to the units of interest during object-oriented design and implementation (for example, object, class, method, etc.). The structural mismatch results in support for a single requirement being scattered across the design units and a single design unit supporting multiple requirements -- this in turn results in reduced comprehensibility, traceability and reuse of design models.

A program's object model captures the essence of its design. For some programs, no object model was developed during design; for others, an object model exists but may be out-of-sync with the code. This paper describes a tool that automatically extracts an object model from the classfiles of a Java program. Unlike existing tools, it handles container classes, by inferring the types of elements stored in a container, and eliding the container itself. This feature is crucial for obtaining models that show the structure of the abstract state, and bear some relation to conceptual models. Although the tool performs only a simple, heuristic analysis that is almost entirely local, the resulting object model is surprisingly accurate. The paper explains what object models are and why they are useful; describes the analysis, its assumptions and limitations; evaluates the tool for accuracy, and illustrates its use, on a suite of sample programs. 1 Introduction Womble is a tool that extracts obj...

There has been substantial recent interest in captured design expertise expressed as design patterns. Prevalent descriptions of these design patterns suffer from two demerits. Firstly, they capture specific instances of pattern deployment, rather than the essential pattern itself, thus the spirit of the pattern is often lost in the superfluous details of the specific instances described. Secondly, existing pattern descriptions rely upon relatively informal diagrammatic notations supplemented with natural language annotations. This can result in imprecision and ambiguity. This paper addresses these problems by separating the specification of patterns into three models (role, type, and class). The most abstract (role-centric) model presents patterns in their purest form, capturing their essential spirit without deleterious detail. A role-model is refined by a type-model (adding usuallydomain-specific constraints), which is further refined by a class-model (forming a concrete deployment). We utilise recent advances in visual modelling notation to achieve greater precision without resorting to obtuse mathematical symbols. A set-oriented view of state, operations, and instances is adopted, permitting their abstract presentation in models via this visual notation. This paper utilises these ideas in the unambiguous specification of a selection of prominent design patterns. The expectation is that precise visual pattern specification will firstly enable clear communication between domain experts and pattern writers (and ultimately pattern users), and secondly enable CASE tool support for design patterns, permitting the designer (pattern user) to operate at a higher level of abstraction without ambiguity.

Testing is an important part of any software development project, and can typically surpass more than half of the development cost. For safety-critical computer based systems, testing is even more important due to stringent reliability and safety requirements. However, most safety-critical computer based systems are real-time systems, and the majority of current testing and debugging techniques have been developed for sequential (non real-time) programs. These techniques are not directly applicable to real-time systems, since they disregard issues of timing and concurrency. This means that existing techniques for reproducible testing and debugging cannot be used. Reproducibility is essential for regression testing and cyclic debugging, where the same test cases are run repeatedly with the intention of verifying modified program code or to track down errors. The current trend of consumer and industrial applications goes from single microcontrollers to sets of distributed micro-controllers, which are even more challenging than handling real-time per-see, since multiple loci of observation and control additionally must be considered. In this thesis we try to remedy these problems by presenting an integrated approach to monitoring, testing, and debugging of distributed real-time systems. For monitoring

Abstract. Metamodelling is becoming a standard way of defining languages such as the UML. A language definition distinguishes between concrete syntax, abstract syntax and semantics domain. It is possible to define all three using a metamodelling approach, but it is less clear how to define the transformations between them. This paper proposes an approach which uses metamodelling patterns that capture the essence of mathematical relations. It shows how these patterns can be used to define both the relationship between concrete syntax and abstract syntax, and between abstract syntax and semantics domain, for a fragment of UML. A goal of the approach is to provide a complete specification of a language from which intelligent tools can be generated. The extent to which the approach meets this goal is discussed in the paper. 1