Steffen Lehnert

A taxonomy for software change impact analysis

Most software is accompanied by frequent changes, whereas the implementation of a single change can affect many different parts of the system. Approaches for Impact Analysis have been developed to assist developers with changing software. However, there is no solid framework for classifying and comparing such approaches, and it is therefore hard to find a suitable technique with minimal effort. The contribution of this paper is a taxonomy for Impact Analysis, based on a literature review conducted on related studies, to overcome this limitation. The presented classification criteria are more detailed and precise than those proposed in previous work, and possible candidates for all criteria are derived from studied literature. We classify several approaches according to our taxonomy to illustrate its applicability and the usefulness of our criteria. The research presented in this paper prepares the ground for a comprehensive survey of Software Change Impact Analysis.

Rule-Based Impact Analysis for Heterogeneous Software Artifacts

Typical software engineering activities, such as program maintenance or reengineering, result in frequent changes of software which are often accompanied by unintended side effects. Consequently, research on impact analysis put forth approaches to assess the adverse effects of changes. However, understanding and implementing these changes is often aggravated by inconsistencies and dependencies between different types of software artifacts. Likewise, most impact analysis approaches are not able to detect the possible side effects of changes when different types of software artifacts are involved. We present an approach that combines impact analysis and multi-perspective modeling for analyzing the change propagation between heterogeneous software artifacts. Our approach assists developers with understanding the consequences of changes by identifying impacted artifacts and determining how they are affected. We utilize a model repository for combining UML models, Java source code, and JUnit tests by mapping them on a unifying meta-model. We introduce a novel impact propagation approach that analyzes dependencies between software artifacts according to the type of change which is applied upon them. Our approach is implemented by a set of impact propagation rules which are evaluated by a case study.

A Taxonomy of Change Types and Its Application in Software Evolution

Changes play a central role in software evolution, where the types of changes are as multifarious as their consequences. When changing software, impact analysis and regression testing are required to preserve the quality of the system. However, without a consistent classification of types of change operations, a well-founded impact analysis methodology cannot be developed. Existing works which analyze and apply change types are typically limited to a certain aspect of software, e.g. source code or architecture. They also lack a thorough investigation of change types, which lead to duplicated proposals and the absence of a consistent taxonomy. In this paper, we review the usage of change types for impact analysis and regression testing, and illustrate how both activities are affected by different types of changes. Therefore, we outline how existing work deals with different types and granularities of changes. Our main contribution is a generic, graph-based description of changes which distinguishes between atomic and composite change types. We show how existing change types and classifications can be mapped onto our proposed approach and change taxonomy. Finally, we illustrate how our proposed change types can support real developer activities, such as refactorings, impact analysis, and regression testing.

Combining Architectural Design Decisions and Legacy System Evolution

Software development is characterized by ongoing design decisions that must take into account numerous requirements, goals, and constraints. When changing long-living and legacy systems, former decisions have to be considered. In order to minimize the risk of taking wrong or misleading decisions an explicit representation of the relevant aspects is crucial. Architectural decision modeling has proven to be an effective means to represent these aspects, the required knowledge, and properties of a potential solution. However, existing approaches do not sufficiently cover the ongoing evolution of decisions and artifacts. They fail in particular to represent relations to existing systems on a fine-grained level to allow for impact analysis and a later comprehension of decisions. Furthermore, the effort for capturing and modeling of design decisions has to be reduced. In our paper we integrate existing approaches for software architectural design decision making. We extend them by fine-grained traceability to elements of existing systems and explicit means for modeling the evolution of decisions. We show how relevant decisions can easily be identified and developers are supported in decision making.

Towards Comprehensive Modelling by Inter-model Links Using an Integrating Repository

Model-based development techniques enable a high efficiency and the mastering of complexity. For many domains more than one model has to be used to express the relevant information. Many methods use different models without a tight coupling, with a high risk of inconsistencies. Other approaches are based on metamodel extension or unified metamodels, with a limited tool support as consequence. We present an approach for the interconnection between several models in a joint repository by means of dependency relationships. The interconnection is shown between UML models, BPMN models and feature models by examples for variable workflows in mobile systems. The presented approach is implemented by the EMF-based repository EMF Trace, with XML for model representation. Dependencies are determined either automatically by a rule set or manually by explicit references.

An IHE-conform telecooperation platform supporting the treatment of dementia patients

Abstract Ensuring medical support of patients of advanced age in rural areas is a major challenge. Moreover, the number of registered doctors—medical specialists in particular—will decrease in such areas over the next years. These unmet medical needs in combination with communication deficiencies among different types of health-care professionals pose threats to the quality of patient treatment. This work presents a novel solution combining telemedicine, telecooperation, and IHE profiles to tackle these challenges. We present a telecooperation platform that supports longitudinal electronic patient records and allows for intersectoral cooperation based on shared electronic medication charts and other documents. Furthermore, the conceived platform allows for an integration into the planned German telematics infrastructure.

A Framework for Traceability Tool Comparison

The need for automated traceability support has been advocated by various researchers and practitioners alike, as manual traceability is cumbersome and error-prone. Thus, many tools and approaches have been developed to overcome the limitations of manual traceability detection and usage. However, the large amount of available traceability tools is lacking suitable means for reasonable comparison based on welldefined criteria, which in turn complicates the process of choosing the appropriate tool. We propose and discuss a framework for classifying traceability tools, demonstrate its feasibility by organizing three tools according to its criteria, and illustrate how its criteria can be used when searching for a suitable tool.

Tackling the Challenges of Evolution in Multiperspective Software Design and Implementation

The design and implementation of software require the usage of different perspectives and views to cope with its static structure, dynamic behavior, and requirements. Artifacts of different views are dependent on each other and subject of frequent changes. Anticipating those changes becomes difficult, as most impact analysis approaches are not designed to work in multiperspective environments. They treat artifacts of different perspectives in isolation, which tends to introduce further inconsistencies and faults. In the related research area of consistency checking, the problem of multiple views has been addressed for a long time and solutions have been developed. We aim on combining the predictive capabilities of impact analysis with the approach of multiperspective consistency checking, to bridge the gap between artifacts of different views. We propose an approach which facilitates impact analysis of different UML models and Java source code. Our approach is based on impact propagation rules, which analyze traceability relations between software artifacts and the type of change applied by a developer.