Logo des Repositoriums

Auflistung nach Autor:in "Steinbeck, Marcel"

1 - 5 von 5
  • Konferenzbeitrag
    Code Cities in Virtual and Augmented Reality
    (Softwaretechnik-Trends Band 40, Heft 2, 2020) Koschke, Rainer; Steinbeck, Marcel
    In this paper, we summarize existing research on visualizing software as code cities with a focus on virtual and augmented reality technology (VR/AR). The review of existing knowledge shows that several of the hopes in VR/AR have not been fulfilled yet. A possible reason for that may be that existing means of visualization and interaction cannot simply be transferred one-to-one from desktop to VR/AR environments as our own research suggests. Moreover, we view the true potential of VR/AR in supporting the communication among multiple developers. All studies we are aware of have had a single developer trying to solve a given task. VR/AR can bridge spatial gaps and more research should be targeted at that particular advantage for distributed teams.
  • Konferenzbeitrag
    How EvoStreets Are Observed in Three-Dimensional and Virtual Reality Environments
    (Software Engineering 2022, 2022) Koschke, Rainer; Steinbeck, Marcel
    We present our paper published in the proceedings of the 27th IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER). In an experiment in which 34 participants had to analyze cloning in existing software systems using the EvoStreets visualization, we found indications that the movement patterns of the participants differ depending on whether they are in a 2.5D (pseudo-3D; monitor with keyboard and mouse) or virtual reality (VR) environment. In a follow-up study, which we present in this work, we analyzed the results of this experiment in more details, to examine whether not only movement is affected by these environments, but also the way how EvoStreets are observed. Beyond that, the paper proposes six visualization and user interaction concepts that are specific to the kind of environment.
  • Zeitschriftenartikel
    LibVCS4j: A Java Library for Repository Mining
    (Softwaretechnik-Trends Band 38, Heft 2, 2018) Steinbeck, Marcel
    Analyzing the evolution of software systems has become an emerging field of research. In the last couple of years, different studies investigated the evolution of various systems by utilizing version control systems and issue trackers. Due to the absence of suitable libraries, however, most of these studies implemented their own data extraction tools. This paper presents LibVCS4j, a Java programming library that tries to accommodate this shortcoming. The library integrates existing software to access repository routines, adds additional features for data analysis, and defines a common API to make subsequent analyses independent from particular repository systems.
  • Zeitschriftenartikel
    A New Approach of Visualizing Code Smells
    (Softwaretechnik-Trends Band 36, Heft 3, 2016) Steinbeck, Marcel
    Code smells are indicators of design flaws that may have negative effects on the comprehensibility and maintainability of a software system. Previous studies on software clones have shown that visualization of findings are useful to identify clones suitable for refactorings. However, techniques to visualize code smells in general are rare and, thus, an interesting field of research to bridge the gap between code smell detection and code smell refactoring. This paper presents an approach that is supposed to help in assessing the extent of code smells and in ranking the relevance of refactoring opportunities.
  • Textdokument
    Partielle Transformation von C++-Programmcode in einen Java-AST zur Erkennung von Code Smells: Erste Ergebnisse
    (Softwaretechnik-Trends: Vol. 37, No. 2, 2017) Steinbeck, Marcel; Nolte, Tobias
    Code Smells sind Strukturschwächen innerhalb von Programmcode, die einen negativen Einfluss auf die Verständlichkeit und Wartbarkeit eines Softwaresystems haben können. Basierend auf den Ergebnissen vergangener Studien, wurden in den letzten Jahren verschiedene Werkzeuge zur automatischen Erkennung von Code Smells entwickelt. Eine beliebte Technik ist dabei die Verwendung abstrakter Syntaxbäume (ASTs) zur Erhebung verschiedener Metriken. Insbesondere für die Programmiersprache Java wurde bereits eine Vielzahl solcher Werkzeuge entwickelt. Tools zur Erkennung von Code Smells in C++- Projekten sind jedoch rar, was nicht zuletzt daran liegt, dass nur wenige geeignete AST-Generatoren für C++ verfu ̈gbar sind. Im Folgenden pra ̈sentieren wir CPP2Spoon, ein Werkzeug zur partiellen Transformation von C++-Programmcode in einen Java-AST.