8th International Workshop on Quantitative Approaches to Software Quality (QuASoQ 2020) Report on the 8th International Workshop on Quantitative Approaches to Software Quality (QuASoQ 2020) Horst Lichtera , Selin Aydina , Thanwadee Sunetnantab and Toni Anwarc a Research Group Software Construction, RWTH Aachen University, Aachen, Germany b Computer Science Academic Group, Faculty of Information And Communication Technology, Mahidol University, Bangkok, Thailand c Faculty of Science and Information Technology, Chair Computer & Information Systems, Universiti Teknologi Petronas: Bandar Seri Iskandar, Perak, MY 1. Introduction quality specification and quality assurance are crucial. Although there are lots of approaches to deal with quan- After a successful 7th QuASoQ workshop we slightly titative quality aspects, it is still challenging to choose adjusted the list of topics for the workshop. The topics a suitable set of techniques that best fit to the specific of interest included project and organizational constraints. Even though approaches, methods, and techniques are • New approaches to measurement, evaluation, known for quite some time now, little effort has been comparison and improvement of software quality spent on the exchange on the real-world problems with • Application of metrics and quantitative ap- quantitative approaches. For example, only limited re- proaches in agile projects search has been devoted to empirically evaluate risks, • Case studies and industrial experience reports efficiency or limitations of different testing techniques on successful or failed application of quantitative in industrial settings. approach-es to software quality Hence, one main goal of the workshop was to exchange • Tools, infrastructure and environments support- experience, present new promising approaches and to ing quantitative approaches discuss how to set up, organize, and maintain quantitative • Empirical studies, evaluation and comparison of approaches to software quality. measurement techniques and models • Quantitative approaches to test process improve- ment, test strategies or testability 2. Workshop History • Empirical evaluations or comparisons of testing The QuASoQ workshop series has been started in 2013. techniques in industrial settings Since then, the workshop is always organized as a col- located event of the Asia-Pacific Software Engineering Overall, the workshop aimed at gathering together Conference (APSEC). researchers and practitioners to discuss experiences in These are the past workshop editions: the application of state of the art approaches to measure, assess and evaluate the quality of both software systems • 7th QuASoQ 2019 as well as software development processes in general and Putrayaya, Malaysia | CEUR Vol-2511 software test processes in particular. • 6th QuASoQ 2018 As software development organizations are always Nara, Japan | CEUR Vol-2273 forced to develop software in the ”right” quality, the • 5th QuASoQ 2017 Nanjing, China | CEUR Vol-2017 QuASoQ 2020: 8th International Workshop on Quantitative • 4th QuASoQ 2016 Approaches to Software Quality, December 01, 2020, Singapore Hamilton, New Zealand | CEUR Vol-1771 Envelope-Open lichter@swc.rwth-aachen.de (H. Lichter); aydin@swc.rwth-aachen.de (S. Aydin); • 3rd QuASoQ 2015 thanwadee.sun@mahidol.ac.th (T. Sunetnanta); New Delhi, India |CEUR Vol-1519 toni.anwar@utp.edu.my (T. Anwar) • 2nd QuASoQ 2014 GLOBE https://www.swc.rwth-aachen.de (H. Lichter); Jeju, Korea | IEEE Xplore https://www.swc.rwth-aachen.de (S. Aydin); https://www.ict.mahidol.ac.th (T. Sunetnanta); • 1st QuASoQ 2013 https://www.utp.edu.my (T. Anwar) Bangkok, Thailand | IEEE Xplore Orcid 0000-0002-3440-1238 (H. Lichter); 0000-0002-0390-8749 (T. Anwar) Since the first edition 58 papers have been presented; © 2020 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). the average acceptance rate is 76 %. The following chart CEUR Workshop Proceedings http://ceur-ws.org ISSN 1613-0073 CEUR Workshop Proceedings (CEUR-WS.org) depicts where the authors of accepted papers come from. 1 8th International Workshop on Quantitative Approaches to Software Quality (QuASoQ 2020) 4. Workshop Contributions Altogether 12 papers were submitted. Finally, the follow- ing 10 papers were accepted by the program committee for presentation and publication covering very different topics. • Sousuke Amasaki Augmenting Window Contents with Transfer Learning for Effort Estimation • Syed Fatiul Huq, Md. Aquib Azmain, Nadia Na- har and Md. Nurul Ahad Tawhid On the Evolutionary Properties of Fix Inducing Changes • Alejandra Duque-Torres, Dietmar Pfahl, Anas- tasiia Shalygina and Rudolf Ramler Using Rule Mining for Automatic Test Oracle Generation • Konrad Fögen and Horst Lichter An Industrial Case Study on Fault Detection Effec- Figure 1: Origin of QuASoQ authors tiveness of Combinatorial Robustness Testing • Azeem Ahmad, Ola Liefler and Kristian Sandhal An Evaluation of Machine Learning Methods for 3. Workshop Format Predicting Flaky Tests • Barry-Detlef Lehmann, Peter Alexander, Horst Because of the covid-19 pandemic, the workshop was ex- Lichter and Simon Hacks ecuted digitally using the video conferencing tool Zoom. Towards the Identification of Process Anti-Patterns Based on our former experience we wanted the work- in Enterprise Architecture Models shop to be highly interactive. In order to have an inter- • Benyamin Shafabakhsh, Robert Lagerström and esting and interactive event sharing lots of experience, Simon Hacks we organized the workshop presentations applying the Evaluating the Impact of Inter Process Communi- author-discussant model. cation in Microservice Architectures Based on this workshop model, papers are presented • Toukir Ahammed, Moumita Asad and Kazi Sakib by one of the authors. After the presentation, a discussant Understanding the Involvement of Developers in starts the discussion based on his or her pre-formulated Missing Link Community Smell: An exploratory questions. Therefore, the discussant had to prepare a set Study on Apache Projects of questions and had to know the details of the presented • Hina Anwar, Iffat Fatima, Dietmar Pfahl and Us- paper. The general structure of each talk was as follows: man Qamar Detection and Correction of Android-specific Code • The author of a paper presented the paper (15 Smells and Energy Bugs: An Android Lint Exten- minutes). sion • After that, the discussant of the paper opened • Kristiina Rahkema and Dietmar Pfahl the discussion using his or her questions. Finally, Comparison of Code Smells in iOS and Android we moderated the discussion among the whole Applications audience (5 minutes). The presentations were divided into four sessions with 5. Summary of the Discussions a ten minute break inbetween. Each session was accom- panied by a moderator who tried to ensure that the sched- About 20 researchers attended the workshop and partici- ule was kept to. A particular challenge were the different pated in the discussions. The author-discussant model time zones of the participants. We decided to hold the was well received by the participants and led to inten- workshop in the afternoon of the timezone in Singapore, sive discussions among them. Hereby, other participants, so that presenters don’t have to attend at nighttime. The apart from the discussant, also joined the resulting dis- order of presenters were also determined by their respec- cussions. tive timezone. Some papers conducted machine learning experiments, which lead to discussions about possible biases of the ap- plied machine learning model. An example for this was 2 8th International Workshop on Quantitative Approaches to Software Quality (QuASoQ 2020) the discussion of the paper by Ahmad et al. where predic- program committee (some of them for many years) and tions of test-flakiness were made based on the number of supported the workshop by soliciting papers and by writ- times a keyword appears in a test-case. A participant sug- ing peer reviews: gested, that other kinds of frequencies, like tf-idf could be considered. Since this was an experimental study, • Ana Nicolaescu, Daimler AG, Germany the authors plan to investigate more into the features • Wan M.N. Wan Kadir, UTM Johor Bahru, and consider weighting them. Besides that, a participant Malaysia pointed out that the predictions may be biased due to • Maria Spichkova, RMIT University, Melbourne, imbalanced classes, since the number of flaky tests used Australia in training is less than the number of non-flaky tests. • Tachanun Kangwantrakool, ISEM, Thailand Furthermore, many proposed methods or approaches • Jinhua Li, Qingdao University, China offer potential for further research. The paper by Huq et • Apinporn Methawachananont, NECTEC, Thai- al., for example, only considers commits in GitHub repos- land itories in order to analyze the evolution of fix-inducing • Nasir Mehmood Minhas, BTH Karlskrona, Swe- changes. Different participants were also interested in den the effect of including bug repositories and differenti- • Chayakorn Piyabunditkul, NSTDA, Thailand ating between major and minor releases. The authors assume that this would lead to different results. • Sansiri Tanachutiwat, Thai German Graduate The paper by Shafabakhsh et. al. lead to a similar School of Engineering, TGGS, Thailand discussion. In their evaluation of the impact of inter • Hironori Washizaki, Waseda University, Japan process communication in microservices, they focus on • Hongyu Zhang, University of Newcastle, Aus- maintainability and availability. Other participants men- tralia tioned to also consider other attributes, like security or • Minxue Pan, Nanjing University, China the development effort that comes with implementing a certain IPC method. Especially the second factor may have an impact on the maintainability of the application. However, this impact has not been quantified yet. Finally, some presented methods were discussed by participants regarding their usefulness in practice. As a result, the approaches presented by Duque-Torres et. al. and Ahmad et. al. could be used to improve fault de- tection and localization. On the other hand, the android smell and bug detector and corrector by Anwar et. al. has potential to become an extension of the existing and widely used tool Android Lint. The discussions show, that empirical studies and the results of experiments are of high value and lead to a deeper understanding of the subject that has been inves- tigated. To conclude, in the course of this workshop the par- ticipants proposed and discussed different approaches to quantify relevant aspects of software development. Es- pecially the discussions led to new ideas, insights, and take-aways for all participants. 6. Acknowledgments Many people contributed to the success of this workshop. First, we want to give thanks to the authors and presen- ters of the accepted papers. Furthermore, we want to express our gratitude to the APSEC 2020 organizers; they did a perfect job and gave us the freedom to conduct the workshop virtually based on our experience. Finally, we are glad that these people served on the 3