Matthias Saft

The EMISQ Method - Expert Based Evaluation of Internal Software Quality

Internal software quality, e.g. the quality of code, has great impact on the overall quality of software. Besides well known manual inspection and review techniques more recent approaches utilize tool-based static code for the evaluation of internal software quality. Despite the high potential of static code analyzers the application of tools alone cannot replace well founded expert opinion. Knowledge, experience and fair judgement is indispensable for a valid, reliable quality assessment, which is accepted by software developers and managers. The EMISQ method (evaluation method for internal software quality), guides the assessment process for all stakeholders of an evaluation project. The method is supported by a tool that assists evaluators with their analysis and rating tasks and provides support for generating a code quality report. The application of the method in a pilot project has shown its applicability.

The EMISQ method and its tool support-expert-based evaluation of internal software quality

There is empirical evidence that internal software quality, e.g., the quality of source code, has great impact on the overall quality of software. Besides well-known manual inspection and review techniques for source code, more recent approaches utilize tool-based static code analysis for the evaluation of internal software quality. Despite the high potential of code analyzers the application of tools alone cannot replace well-founded expert opinion. Knowledge, experience and fair judgment are indispensable for a valid, reliable quality assessment, which is accepted by software developers and managers. The EMISQ method (Evaluation Method for Internal Software Quality), guides the assessment process for all stakeholders of an evaluation project. The method is supported by the Software Product Quality Reporter (SPQR), a tool which assists evaluators with their analysis and rating tasks and provides support for generating code quality reports. The application of SPQR has already proved its usefulness in various code assessment projects around the world. This paper introduces the EMISQ method and describes the tool support needed for an efficient and effective evaluation of internal software quality.

A Method for Continuous Code Quality Management Using Static Analysis

The quality of source code is a key factor for any software product and its continuous monitoring is an indispensable task for a software development project. We have developed a method for systematic assessing and improving the code quality of ongoing projects by using the results of various static code analysis tools. With different approaches for monitoring the quality (a trend-based one and a benchmarking-based one) and an according tool support we are able to manage the large amount of data that is generated by these static analyses. First experiences when applying the method with software projects in practice have shown the feasibility of our method.

Automatic Checking of Quality Best Practices in Software Development Documents

Quality assurance tasks of software products typically take place throughout the entire software development life-cycle. Frequently, paper-based inspections methods are used to review software development documents (e.g. requirements specifications, design documents, test plans). Nevertheless, even though the quality of these documents has a major impact on the quality of the developed software product, they are often not as rigorously reviewed as source code. This paper presents a tool-based approach that facilitates the software inspection process in order to determine defects of generally accepted documentation best practices in software development documents. Moreover, we present results of a conducted empirical study and show how this tool-based approach helps to facilitate inspection tasks and to support gathering information on the quality of the inspected documents.

The Value of Software Documentation Quality

This paper presents the results of a study on software documentation quality in practice. Goal of this study is identifying the current state of software documentation quality and used analysis techniques for determining software documentation quality. Moreover, we aim at finding out, whether there is a demand for a tool-based software documentation quality analysis approach. This approach consists of a documentation quality model and a document checking tool, as proposed in previous work. We developed an online survey and asked about 300 experts to answer it. The survey was completed by 88 experts and the overall results confirm the importance of software documentation quality as well as the need for better tool support. The survey shows that the most important quality attributes with regard to documentation quality are accuracy, clarity, consistency, readability, structuredness, and understand ability. Most of these quality attributes are currently covered by our software documentation quality analysis approach, some of them (e.g., accuracy, structuredness) still need more attention, i.e. better support in our quality model and tool.

A Comprehensive Code-Based Quality Model for Embedded Systems: Systematic Development and Validation by Industrial Projects

Existing software quality models typically focus on common quality characteristics such as the ISO 25010 software quality characteristics. However, most of them provide insufficient operationalization for quality assessments of source code. Moreover, they usually focus on software in general or on information systems and do not sufficiently cover the particularities of embedded systems. We have developed a quality model that covers quality requirements for source code that are specific for embedded systems software. It provides comprehensive operationalization (with 336 measures) for C and C++ systems, which allows for largely automated quality assessments. The empirical evaluations performed acknowledge moderate completeness of the requirements and the associated measures. Therefore, we still see room for improvements to allow covering even more aspects of embedded systems software quality. Nevertheless, the empirical validation (based on three industrial products) shows good concordance between the results gained by the automatic model-based assessment and independent expert judgment on code quality.

Tool Support for Expert-Centred Code Assessments

There is empirical evidence that the code quality of software has an important impact on the external, i.e., user perceptible, software quality. Currently a large number of source code metrics exist that seem to ease the evaluation of code quality. Nevertheless, studies show that the theoretical foundations are weak and promising approaches for the automatic assessment of code quality are to be considered with great caution. We therefore came to the conclusion that the metric values and other findings provided by various static code analysis tools can only be used in the context of an expert-centred assessment of internal software quality. In order to be able to carry out code quality assessments in a timely and efficient manner it is inevitable to have additional tool support. For that purpose we developed the eclipsed based tool software product quality reporter (SPQR) that supports expert-centred evaluation of source code - from the formulation of project-specific quality models up to the generation of preliminary code quality reports. The application of SPQR already proved its usefulness in various code assessment projects around the world.

Measuring, Assessing and Improving SoftwareQuality based on Object-Oriented DesignPrinciples.

Abstract Good object-oriented design is crucial for a successful software product. Metric-based approaches and the identification of design smells are established concepts for identifying design flaws and deriving design improvements thereof. Nevertheless, metrics are difficult to use for improvements as they provide only weak guidance and are difficult to interpret. Thus, this paper proposes a novel design quality model (DQM) based on fundamental object-oriented design principles and best practices. In course of discussing DQM, the paper provides a contribution in three directions: (1) it shows how to measure design principles automatically, (2) then the measuring result is used to assess the degree of fulfilling object-oriented design principles, (3) and finally design improvements of identified design flaws in object-oriented software are derived. Additionally, the paper provides an overview of the research area by explaining terms used to describe designrelated aspects and by depicting the result of a survey on the importance of object-oriented design principles. The underlying concepts of the DQM are explained before it is applied on two open-source projects in the format of a case study. The qualitative discussion of its application shows the advantages of the automated design assessment that can be used for guiding design improvements.

MUSE: A Framework for Measuring Object-Oriented Design Quality.

Good object-oriented design facilitates the maintainability of a software product. While metric-based approaches and the idea of identifying design smells have been established, there still remains the gap of verifying the compliance of design best practices in source code. Thus, there is no comprehensive set of metrics or design best practices that does not only support design measurement and evaluation but can also guide the improvement process. This paper proposes a novel approach based on measuring design best practices that closes the gap between the identification of design flaws and the support for improvements. An expert group six researchers captured a set of 67 design best practices that are implemented by the framework MUSE (Muse Understand Script Engine). For a first validation of MUSE in this paper, its measuring result is compared with QMOOD, which is an established metric-based approach for measuring the quality of object-oriented design. The qualitative assessment based on data from six versions of the Java tool jEdit shows that MUSE is better suited to guide improvements than QMOOD, e.g., for the design property encapsulation QMOOD indicates no substantial changes in the design quality while the data provided by MUSE highlights that the encapsulation property of jEdit became worse over time. These first promising results of the application of MUSE have to be further validated and future work will concentrate on measuring object-oriented design principles.

A Survey on the Importance of Object-Oriented Design Best Practices

To measure object-oriented design quality, metric-based approaches have been established. These have then been enhanced by identifying design smells in code. While these approaches are useful for identifying hot spots that should be refactored, they are still too vague to sufficiently guide software developers to implement improvements. This is why our previous work focuses on measuring the compliance of source code with object-oriented design best practices. These design best practices were systematically derived from the literature and can be mapped to design principles, which can help reveal fundamental object-oriented design issues in a software product. Despite the successful applications of this approach in industrial and open source projects, there is little accepted knowledge about the importance of various design best practices. Consequently, this paper shows the result of an online survey aimed at identifying the importance of 49 design best practices on design quality. In total, 214 people participated in the survey, resulting in an average of 138 opinions for each practice. Based on these opinions, five very important, 21 important, 12 moderately important and 11 unimportant design best practices could be derived. This information about importance helps managing design improvements in a focused way.