<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by The Blog of Systems and Software on Medium]]></title>
        <description><![CDATA[Stories by The Blog of Systems and Software on Medium]]></description>
        <link>https://medium.com/@JSSoftware?source=rss-4f58beca6fe3------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*vm3TbniprXdbd72A.jpg</url>
            <title>Stories by The Blog of Systems and Software on Medium</title>
            <link>https://medium.com/@JSSoftware?source=rss-4f58beca6fe3------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 10:22:26 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@JSSoftware/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[JSS Happy Hour — TikTok Edition]]></title>
            <link>https://medium.com/@JSSoftware/jss-happy-hour-tiktok-edition-905f4b74290b?source=rss-4f58beca6fe3------2</link>
            <guid isPermaLink="false">https://medium.com/p/905f4b74290b</guid>
            <category><![CDATA[happy-hour]]></category>
            <category><![CDATA[computer-science]]></category>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[research]]></category>
            <category><![CDATA[tik-tok]]></category>
            <dc:creator><![CDATA[The Blog of Systems and Software]]></dc:creator>
            <pubDate>Tue, 07 Sep 2021 16:31:01 GMT</pubDate>
            <atom:updated>2021-09-09T12:16:45.294Z</atom:updated>
            <content:encoded><![CDATA[<h3>JSS Happy Hour — TikTok Edition</h3><p>The Journal of Systems &amp; Software will be hosting another happy hour, usually attended by 100s of software engineering researchers and enthusiasts, on <strong><em>October 15th </em></strong><em>at </em><strong><em>10am EST</em></strong>. However, this time we are extending an invitation to the community, including authors that have recently published papers at JSS, TSE, EMSE, IST, TOSEM, ICSE, ASE, IEEE Software, and FSE. All you have to do is to submit a vertical video detailing your most recent paper that is 30–90 seconds long. We will feature as many videos as possible (depending on video length and quality) at our live happy hour presentation on YouTube Live.</p><p><strong><em>Requirements:</em></strong></p><p>Must be based on a recent (2020–2021) paper published in JSS, TSE, EMSE, IST, TOSEM, ICSE, ASE, IEEE Software, or FSE.<br>Must be in .mp4 format.<br>Must be in portrait orientation (NOT landscape).<br>Must be filmed at 1080p resolution.<br>Must be submitted by October 6, EOD, AOE.<br>Must be factually accurate.<br>Must be informative.</p><p><strong><em>Suggestions:</em></strong></p><p>Can include music.<br>Can be funny.<br>Can be filmed and uploaded using TikTok’s video editing software.<br>Can include researchers “dancing”.</p><p><a href="https://calendar.google.com/event?action=TEMPLATE&amp;tmeid=N3FxYWMwcW5oMTFjcG1vNmtzbXFobjllYmogZGF2aWRzaGVwaGVyZEBt&amp;tmsrc=davidshepherd%40gmail.com">Add this event to your calendar.</a></p><p><a href="https://forms.gle/teh4FQPSWWQkuitQ7">Fill out this form to submit your video.</a></p><p>All the best,<br>David C. Shepherd<br>Paris Avgeriou</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=905f4b74290b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Architectural Knowledge Management Tools]]></title>
            <link>https://medium.com/jss-papers-of-the-year/architectural-knowledge-management-tools-7c8fdb3a2c88?source=rss-4f58beca6fe3------2</link>
            <guid isPermaLink="false">https://medium.com/p/7c8fdb3a2c88</guid>
            <dc:creator><![CDATA[The Blog of Systems and Software]]></dc:creator>
            <pubDate>Sat, 27 Feb 2021 11:18:36 GMT</pubDate>
            <atom:updated>2021-02-28T10:15:54.788Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*edcQhDjm02622qSZ35WkDA.png" /></figure><h3><strong>Research on Architectural Knowledge Management Tools — 11 Years On</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/56/0*mDfh3TduICxFB2t-.jpeg" /></figure><p>Contemporary software architecting practices consider Architectural Design Decisions (ADDs) as first-class elements; as such, they should be captured and shared. While this has been established since 2005, there are no commercial tools to support the capture and use of the relevant Architectural Knowledge (AK). However, a number of research tools supporting different AK processes have been developed and used over the years.</p><h3>Architectural Knowledge Management (AKM) in Practice: The “Second brain”</h3><p>Software architects currently consider capturing design decisions and their underlying rationale as of paramount importance. Architects can share and explain how architectures are built and what design alternatives were considered.</p><p>AKM approaches to capture relevant AK and design rationale for reuse are understood by most software architects. Architects can use such approaches to avoid knowledge vaporization, but also to solve new design problems. Integrating AKM practices into current software architecting tasks require activities such as those shown in Figure 1. Different types of decisions (e.g. strategic, technical, operational) use different reasoning that are based on general and context knowledge. An architect distills, synthesizes and applies knowledge for creating her solutions. She captures the knowledge for future references. Consumer of that knowledge can benefit from the searches, traces and knowledge evaluation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KY-HRQwEilwM_F37hXOZdw.png" /><figcaption><strong>Figure 1: Architectural Knowledge activities</strong></figcaption></figure><p>In order to support AKM, there is a strong need to perform research into usable AKM tools to capture and manage AK, and use these tools to assist software designers in design-thinking and software modeling activities. AK can act as a “second brain” a metaphor where senior and novice architects can store and leverage the knowledge, experiences, and reasons that can be leveraged for solving new design problems, compared to other forms of static knowledge like simply reusing design patterns or architectural styles. AKM tools should help software architects to retain their expertise and be used by novices to learn from past decisions. Furthermore, these tools help with retaining knowledge and fight knowledge vaporization within an organization.</p><h3>AKM: Work over the past 10 years</h3><p>Our research from 2010 revealed the maturity of the first two generations of AKM tools. We provided a framework to compare the tools. We mapped the AK concepts used by the tools, and related them to the IEEE 1471–2000 Std. as Recommended Practice for Architecture Description of Software-Intensive Systems. Nevertheless, there is still a lack of incentives using these tools in industrial settings apart from some research experiences. We continued to study AKM in 2016 and followed through three generations of AKM tools and practices (as described in [2]). AKM faces new challenges such as global software development and Agile practices. AKM longevity and sustainability are key considerations to counter knowledge vaporization. Training and education are the other AKM considerations. We also identified six barriers in the adoption of AKM technologies, all of which relate to training developers to consider AK as a long term investment. The development of the majority of modern AK tools stopped around 2015, being ADMentor one of the latest ones. Figure 2 shows a screenshot of the ADMentor tool which illustrates the duality problem space and solution space for capturing the design decisions. Recently, instead of having dedicated AKM tools, people are starting to use architecture decision logs as bare bones to reuse knowledge and experience. Hence, although the concepts underlying AKM tools are finding their way into industry, the associated tooling support is still lacking.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WUQNuhjsnp9tLv6wiNjS8Q.jpeg" /><figcaption><strong>Figure 2: ADMentor tool capturing the design decisions</strong></figcaption></figure><p><strong>AK Trends: What’s next</strong></p><p>It has been over 10 years since we published our first analysis on AKM tools in 2010. Despite some convincing features, most of these AKM tools remain as research tools and are now unavailable. Whilst the concept of capturing design rationale is well understood by architects, the lack of commercial tools supporting the AKM processes that we identified remains the same. New techniques such as capturing design decisions by voice have appeared recently but they require validation and adoption. The current revision of the ISO/IEC 42010:2011 Std. on Systems and software engineering — Architecture Description suggests a refinement of the AKM concepts that will need to be supported by a new generation of AKM tools or commercial ones. Having an AKM tool as a “second brain” to support software architects is still challenging. Also, as AKM concepts remain intact but technologies may change, productivity supports apps can be used for supporting AKM processes. Whilst some current Integrated Development Environments (IDEs) and UML tools have limited AK capture in text form, we remain hopeful that new and more sophisticated commercial tools will be available to capture AK easily and efficiently what’s in architects’ minds in a non-intrusive way.</p><blockquote><strong>Few software architects doubt the importance for documenting key design decisions. They just don’t want to spend the efforts to document them. The research efforts in these years have not solved this human issue. Perhaps when we manage to convince architects to capture AKM, similar to convincing programmers to writing comments for software codes, and when we support them with easy-to-use tools and apps, we can democratize AKM as another common software development practice.</strong></blockquote><h3>References</h3><p>[1] <a href="https://dblp.org/pid/97/2801.html">Antony Tang</a><em>, </em><a href="https://dblp.org/pid/30/3651.html">Paris Avgeriou</a><em>, </em><a href="https://dblp.org/pid/77/857.html">Anton Jansen</a><em>, </em>Rafael Capilla<em>, </em><a href="https://dblp.org/pid/03/1909.html">Muhammad Ali Babar</a><em>: </em>A comparative study of architecture knowledge management tools. <a href="https://dblp.org/db/journals/jss/jss83.html#TangAJCB10">J. Syst. Softw. 83(3)</a><em>: 352–370 (2010)</em></p><p>[2] Rafael Capilla<em>, </em><a href="https://dblp.org/pid/77/857.html">Anton Jansen</a><em>, </em><a href="https://dblp.org/pid/97/2801.html">Antony Tang</a><em>, </em><a href="https://dblp.org/pid/30/3651.html">Paris Avgeriou</a><em>, </em><a href="https://dblp.org/pid/03/1909.html">Muhammad Ali Babar</a><em>: </em>10 years of software architecture knowledge management: Practice and future. <a href="https://dblp.org/db/journals/jss/jss116.html#CapillaJTAB16">J. Syst. Softw. 116</a><em>: 191–205 (2016)</em></p><p>[3]Tiago Forte: Building a Second Brain, “https://www.youtube.com/watch?v=SjZSy8s2VEE&amp;t=1050s&quot;</p><p>[4] Olaf Zimmermann, Lukas Wegmann, Heiko Koziolek, Thomas Goldschmidt: Architectural Decision Guidance Across Projects — Problem Space Modeling, Decision Backlog Management and Cloud Computing Knowledge. WICSA 2015: 85–94</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/328/1*hUISIkA4YG6RedmmFJVnBw.jpeg" /><figcaption><a href="https://rafacap1201.medium.com/"><strong>Rafael Capilla</strong></a></figcaption></figure><p>Written by <a href="https://rafacap1201.medium.com/">Rafael Capilla</a>. <em>Originally published at </em><a href="https://rafacap1201.medium.com/architectural-knowledge-management-tools-3114c92db48e"><em>https://rafacap1201.medium.com</em></a><em> on February 12, 2021.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7c8fdb3a2c88" width="1" height="1" alt=""><hr><p><a href="https://medium.com/jss-papers-of-the-year/architectural-knowledge-management-tools-7c8fdb3a2c88">Architectural Knowledge Management Tools</a> was originally published in <a href="https://medium.com/jss-papers-of-the-year">JSS Papers</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Diversity, Diversity, Diversity:]]></title>
            <link>https://medium.com/jss-papers-of-the-year/diversity-diversity-diversity-e972cec64acd?source=rss-4f58beca6fe3------2</link>
            <guid isPermaLink="false">https://medium.com/p/e972cec64acd</guid>
            <dc:creator><![CDATA[The Blog of Systems and Software]]></dc:creator>
            <pubDate>Fri, 26 Feb 2021 11:10:40 GMT</pubDate>
            <atom:updated>2021-02-27T12:52:10.487Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OLh-JEay0mEsefsYvdlshQ.png" /></figure><h3>Diversity, Diversity, Diversity: Three Words that Summarize the Impacts of Adaptive Random Testing</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/96/0*aPa3yfZkaH1rVWwQ.jpeg" /></figure><p>When we consider real estate, we know that “There are three things that matter in property: location, location, location” — a statement attributed to Lord Harold Samuel in 1944.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*emKMkLDuNq877QiPVmbGMg.jpeg" /></figure><p>The article titled “Adaptive random testing: The ART of test case diversity” by T.Y. Chen, F.-C. Kuo, R.G. Merkel, and T.H. Tse was published in <em>Journal of Systems and Software</em> in 2010. We would like to mimic Lord Samuel and summarize the impacts and influences of our paper in three words: “diversity, diversity, diversity”. Intuitively speaking, the most effective way to reveal software failures is through the design of diversified test cases, because programmers may make all kinds of mistakes.</p><p>On the one hand, the impacts of the ART 2010 paper are at the fundamental level. It has enhanced the foundations of software testing, and opened up new research directions. On the other hand, the 2010 paper has also strongly influenced researchers to work on new ART algorithms, and motivated practitioners to apply ART to real-life software testing. These impacts have become apparent in the decade following the publication of the ART 2010 paper, as illustrated in this blog entry.</p><h3>1. Adaptive Random Testing: The ART of Diversity in Test Cases</h3><p>Random testing (RT) has been recognized as important in software quality assurance. It attempts to exercise a diversified choice of test cases through random selection. However, it has been criticized for insufficiently effective as it does not make use of any available information to guide the selection of test cases. Adaptive random testing (ART) was proposed by T.Y. Chen in the late 1990s to enhance random diversification of test cases by algorithmically spreading them more evenly.</p><p>The ART 2010 paper published in <em>Journal of Systems and Software</em> was a reflection on adaptive random testing after a decade of its inception. The article discussed the importance of test case diversity, justified by research results demonstrating that the simple notion of even spreading across the input domain can significantly enhance failure detection capability in software testing.</p><p>The concept of “even spreading across the input domain” is easily understood by any person, but is difficult to describe precisely. The difficulty lies in the fact that there are many perspectives to achieve such a phenomenon. Hence, numerous ART algorithms have been developed from various approaches or intuitions. For instance, the first and most popular ART algorithm, FSCS-ART, is based on the approach to choosing the farthest among a set of potential candidates as the next test case in order to achieve this phenomenon.</p><p>Thus, adaptive random testing has evolved into a large family of enhanced random testing methods built on the very simple notion of diversity and even spreading. This notion leverages on the phenomenon that neighboring inputs are more likely to behave similarly, because most programs mimic our daily lives, which are rarely stochastic.</p><h3>2. Adaptive Random Testing: The ART of Diversity in Techniques</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-Oh5_moYk7qgCTryi7Ng5w.jpeg" /><figcaption>Cumulative No. of Papers on ART Techniques and Applications since 2010</figcaption></figure><p>Recognizing the limitations in efficiency of the initial techniques, numerous papers have been published since 2010, proposing improved ART algorithms. Some notable examples are ant colony optimization [4], ART by exclusion through test profile [15], ART4SQLi [25], ARTsum [3], category selection-based ART [27], clustering [23], code coverage [7], divide and conquer [9], distance-aware forgetting strategy [17], dynamic partitioning [10], KD-ART [22], KDFC-ART [18], OMISS-ART [5], orthogonal recursive bisection [1], randomized quasi-random testing [14], and static mirror adaptive random testing [21]. For instance, ART by exclusion [15] is based on the concept of predefining a criterion and taking the first candidate that satisfies this criterion as the next test case. The ART algorithms by partitioning [10] are sparked by the intuition that the next test case should be selected from a partition without test cases.</p><p>In particular, ART by bisection [1] and RBCVT-Fast using clustering by centroidal Voronoi tessellations [23] are linear-time algorithms for numeric programs. ARTsum [3] and category selection-based ART [26] are linear-time algorithms for nonnumeric programs. Two independent projects are currently working on linear-time algorithms for object-oriented programs.</p><p>Different ART methodologies have different favorable and unfavorable testing scenarios. Given the large family of techniques, the good news is that you can choose a cost-effective ART algorithm for the particular program under test. The bad news is that there is no free lunch — you need to understand ART techniques as well as the basic characteristics of the program under test before you can select the most appropriate one.</p><p>In addition to motivating the development of more effective and efficient algorithms, the ART 2010 paper has also suggested new insights, interpretations, and research directions. Most notably, the use of failure patterns, such as the distribution of failure-causing inputs, has given rise to the novel research direction of failure-based software testing [20]. Given such knowledge, we may successfully construct adversarial attacks of machine learning systems to test their robustness. Another example of failure patterns is the conceived geometry of the failure areas. This allows us to utilize any minimal information on hand to compare the effectiveness of software testing approaches, without the need to know the detailed algorithms or procedures.</p><p>Random ordering and random sequences have been commonly used as benchmarks for comparing test sequences generated by various algorithms in different problem contexts. Following ART, the new concept of adaptive random sequences [6][26] has evolved as a more effective alternative than random sequences.</p><h3>3. Adaptive Random Testing: The ART of Diversity in Applications</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yXFBSJU9FzWimN-dOSSysA.jpeg" /><figcaption>Distribution of Applications of Adaptive Random Testing</figcaption></figure><p>ART has been adopted in numerous application domains. A plausible reason is that RT can be used alone or with other testing methods. It does not need the source code. As ART is an alternative to RT, it can be adopted in many application domains as well. The only limitation is the need for test oracles. With the increasing acceptance of metamorphic testing [8] as an effective means to alleviate the oracle problem, the applicability of RT and ART can be extended to the domains involving oracle problems.</p><p>Huawei has applied ART to an industrial case study involving 50 million test cases [27]. Originally, it took 50 days to test and analyze. It was necessary to reduce the number of test cases while revealing more failures. At the same time, control flow, data flow, or code instrumentation could not be applied for reasons of privacy and safety. Based on the linear time algorithm ARTsum [3], category selection-based ART (CSBART) was proposed, where partial categories were selected to compute test case distances. Experimentation showed that CSBART outperformed a more complex K-means clustering strategy.</p><p>Simula Research Laboratory has successfully applied ART to an industrial real-time embedded system [2][13][19]. Despite Simula’s initial observation regarding the inefficiency of ART, the findings from the industrial system were positive. “Results on artificial problems and one industrial case study show that Adaptive Random Testing performs best.” [13] “One main advantage of ART and SBT is that [they] can be tailored to whatever time and resources are available for testing: when resources are expended and time is up, we can simply stop their application without any side effect.” [2] “Adaptive random output diversity algorithm is able to identify different failure types better than the search-based algorithms when test suites are above a certain size.” [19]</p><p>Other notable practical applications include object-oriented programs [5], web services [24], configurable systems [3][7], mobile applications [16], database applications [25], regression testing [6], combinatorial testing [12], and mutation testing [22].</p><p>Furthermore, based on the new research direction proposed in the 2010 paper, a collaborative project between the industry and the academia is being conducted to study how failure patterns can help detect obstacle perception problems in automatic driving systems.</p><h3>4. Support, Support, Support: Three Further Things that Matter in ART</h3><p>There are three further things that matter in adaptive random testing: your support, your support, your support.</p><blockquote>• If you are a researcher, we need your support in proposing better ART approaches and empirically evaluating the family of such techniques.</blockquote><blockquote>• If you are a practitioner, we need your support in applying various ART methodologies to large-scale industrial applications and sharing your experience with us.</blockquote><blockquote>• In either case, we need your immediate support in clicking the “applause” button at the end of the blog entry.</blockquote><p><strong>References</strong></p><p>[1] Ackah-Arthur, H., Chen, J., Towey, D., Omari, M., Xi, J., Huang, R., 2019. One-domain-one-input: Adaptive random testing by orthogonal recursive bisection with restriction. IEEE Transactions on Reliability 68 (4), 1404–1428.</p><p>[2] Arcuri, A., Iqbal, M.Z., Briand, L., 2010. Black-box system testing of real-time embedded systems using random and search-based testing. In: Testing Software and Systems: Proceedings of the 22nd IFIP WG 6.1 International Conference (ICTSS 2010), Lecture Notes in Computer Science, vol. 6435. Springer, Berlin, Germany, pp. 95–110.</p><p>[3] Barus, A.C., Chen, T.Y., Kuo, F.-C., Liu, H., Merkel, R.G., Rothermel, G., 2016. A cost-effective random testing method for programs with non-numeric inputs. IEEE Transactions on Computers 65 (12), 3509–3523.</p><p>[4] Bidgoli, A.M., Haghighi, H., 2020. Augmenting ant colony optimization with adaptive random testing to cover prime paths. Journal of Systems and Software 161, 110495.1–110495.17.</p><p>[5] Chen, J., Kuo, F.-C., Chen, T.Y., Towey, D., Su, C., Huang, R., 2017. A similarity metric for the inputs of OO programs and its application in adaptive random testing. IEEE Transactions on Reliability 66 (2), 373–402.</p><p>[6] Chen, J., Zhu, L., Chen, T.Y., Towey, D., Kuo, F.-C., Huang, R., Guoa, Y., 2018. Test case prioritization for object-oriented software: An adaptive random sequence approach based on clustering. Journal of Systems and Software 135, 107–125.</p><p>[7] Chen, T.Y., Kuo, F.-C., Liu, H., Wong, W.E., 2013. Code coverage of adaptive random testing. IEEE Transactions on Reliability 62 (1), 226–237.</p><p>[8] Chen, T.Y., Kuo, F.-C., Liu, H., Poon, P.-L., Towey, D., Tse, T.H., Zhou, Z.Q., 2018. Metamorphic testing: A review of challenges and opportunities. ACM Computing Surveys 51 (1), 4:1–4:27.</p><p>[9] Chow, C., Chen, T.Y., Tse, T.H., 2013. The ART of divide and conquer: An innovative approach to improving the efficiency of adaptive random testing. In: Proceedings of the 13th International Conference on Quality Software (QSIC 2013), IEEE Computer Society, Los Alamitos, CA, 268–275.</p><p>[10] Huang, R., Liu, H., Xie, X., Chen, J., 2015. Enhancing mirror adaptive random testing through dynamic partitioning. Information and Software Technology 67, 13–29.</p><p>[11] Huang, R., Sun, W., Xu, Y., Chen, H., Towey, D., Xia, X., 2019. A survey on adaptive random testing. IEEE Transactions on Software Engineering. DOI: <a href="https://doi.org/10.1109/TSE.2019.2942921.">https://doi.org/10.1109/TSE.2019.2942921.</a></p><p>[12] Huang, R., Xie, X., Chen, T.Y., Lu, Y., 2012. Adaptive random test case generation for combinatorial testing. In: Proceedings of the IEEE 36th Annual International Computer Software and Applications Conference (COMPSAC 2012). IEEE Computer Society, Los Alamitos, CA, pp. 52–61.</p><p>[13] Iqbal, M.Z.Z., Arcuri, A., Briand, L.C., 2011. Automated system testing of real-time embedded systems based on environment models. Technical Report 2011–19, Simula Research Laboratory, Lysaker, Norway.</p><p>[14] Liu, H., Chen, T.Y., 2016. Randomized quasi-random testing. IEEE Transactions on Computers 65 (6), 1896–1909.</p><p>[15] Liu, H., Xie, X., Yang, J., Lu, Y., Chen, T.Y., 2010. Adaptive random testing by exclusion through test profile. In: Proceedings of the 10th International Conference on Quality Software (QSIC 2010). IEEE Computer Society, Los Alamitos, CA, pp. 92–101.</p><p>[16] Liu, Z., Gao, X., Long, X., 2010. Adaptive random testing of mobile application. In: Proceedings of the 2nd International Conference on Computer Engineering and Technology (ICCET 2010), vol. 2. IEEE, Piscataway, NJ, pp. 297–301.</p><p>[17] Mao, C., Chen, T.Y., Kuo, F.-C., 2017. Out of sight, out of mind: A distance-aware forgetting strategy for adaptive random testing. Science China Information Sciences 60, 092106:1–092106:21.</p><p>[18] Mao, C., Zhan, X., Tse, T.H., Chen, T.Y., 2019. KDFC-ART: a KD-tree approach to enhancing Fixed-size-Candidate-set Adaptive Random Testing. IEEE Transactions on Reliability 68 (4), 1444–1469.</p><p>[19] Matinnejad, R., Nejati, S., Briand, L.C., Bruckmann, T., 2015. Effective test suites for mixed discrete-continuous Stateflow controllers. In: Proceedings of the 10th Joint Meeting on Foundations of Software Engineering (ESEC/FSE-10). ACM, New York, NY, pp. 84–95.</p><p>[20] Merkel, R.G., Kuo, F.-C., Chen, T.Y., 2011. An analysis of failure-based test profiles for random testing. In: Proceedings of the 35th Annual International Computer Software and Applications Conference (COMPSAC 2011). IEEE Computer Society, Los Alamitos, CA, pp. 68–75.</p><p>[21] Omari, M., Chen, J., Ackah-Arthur, H., Kudjo, P.K., 2019. Elimination by linear association: An effective and efficient static mirror adaptive random testing. IEEE Access 7, 71038–71060.</p><p>[22] Patrick, M., Jia, Y., 2017. KD-ART: Should we intensify or diversify tests to kill mutants? Information and Software Technology 81, 36–51.</p><p>[23] Shahbazi, A., Tappenden, A.F., Miller, J., 2013. Centroidal Voronoi tessellations: A new approach to random testing. IEEE Transactions on Software Engineering 39 (2), 163–183.</p><p>[24] Tappenden, A.F., Miller, J., 2014. Automated cookie collection testing. ACM Transactions on Software Engineering and Methodology 23 (1), 3:1–3:40.</p><p>[25] Zhang, L., Zhang, D., Wang, C., Zhao, J., Zhang, Z., 2019. ART4SQLi: The ART of SQL injection vulnerability discovery. IEEE Transactions on Reliability 68 (4), 1470–1489.</p><p>[26] Zhang, X., Xie, X., Chen, T.Y., 2016. Test case prioritization using adaptive random sequence with category-partition-based distance. In: Proceedings of the 2016 IEEE International Conference on Software Quality, Reliability and Security (QRS 2016). IEEE Computer Society, Los Alamitos, CA, pp. 374–385.</p><p>[27] Zhang, Z., Wang, Y., Wang, Z., Qian, J., 2019. How to effectively reduce tens of millions of tests: An industrial case Study on adaptive random testing. IEEE Transactions on Reliability 68 (4), 1429–1443.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/262/1*sgi3sT8sySBy1WJl-pmy0g.jpeg" /><figcaption><a href="https://thtse.medium.com/"><strong><em>T.H. Tse</em></strong></a></figcaption></figure><p><em>Written by </em><a href="https://thtse.medium.com/"><strong><em>T.H. Tse</em></strong></a><em>, Honorary professor at The University of Hong Kong, for </em><a href="https://medium.com/jss-editors-selection/diversity-diversity-diversity-ed96828f2aeb"><em>JSS Editor’s Selection</em></a><em> on February 26, 2021.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e972cec64acd" width="1" height="1" alt=""><hr><p><a href="https://medium.com/jss-papers-of-the-year/diversity-diversity-diversity-e972cec64acd">Diversity, Diversity, Diversity:</a> was originally published in <a href="https://medium.com/jss-papers-of-the-year">JSS Papers</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A family of code coverage-based heuristics for effective fault localization]]></title>
            <link>https://medium.com/jss-papers-of-the-year/a-family-of-code-coverage-based-heuristics-for-effective-fault-localization-373b50a17ed5?source=rss-4f58beca6fe3------2</link>
            <guid isPermaLink="false">https://medium.com/p/373b50a17ed5</guid>
            <dc:creator><![CDATA[The Blog of Systems and Software]]></dc:creator>
            <pubDate>Thu, 25 Feb 2021 14:19:52 GMT</pubDate>
            <atom:updated>2021-02-25T22:51:01.787Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*o6UFRPKsNpfNWVAUGV3cnQ.png" /></figure><p>Fault localization (which aims to reveal locations of faults in software in automatic or semi-automatic ways) has received considerable attention from academia and industry for the last two decades. Our paper, which was published in the prestigious Journal of Systems of Software right at the beginning of the last decade, focuses on spectrum-based fault localization, i.e., leveraging the useful data provided by test case execution and code coverage to narrow down the search domain when looking for faults in software. More specifically, we compare and contrast the code-coverage information (which portion of the code is covered and which is not) provided by successful and failed test cases to rank the portions of code (at the statement level) in terms of their likelihood (or <em>suspiciousness</em>) of containing faults. The general idea is that programmers can examine the code from most suspicious to least when locating faults, thereby providing order and direction to what would otherwise be an extremely large search domain.</p><p>In a period when many such papers were exploring the possibilities of applying different techniques (such as similarity coefficients, machine learning, etc.) to provide better rankings, and then demonstrating that the technique proposed was more effective than other proposed techniques by virtue of empirical evaluation, our paper was seminal in that it provided four crucial novelties:</p><ol><li>It started out by boiling down the computation of a suspiciousness (of a piece of code) to its simplest intuition — generically, the more failed test cases that execute a statement the more likely it is to be faulty.</li><li>It then added on the intuition that the more successful test cases that execute a statement the less likely it is to be faulty.</li><li>Next, it posited that test case/code-coverage data should not be considered in isolation; rather, each additional test case provides additional information that should be taken into account in tandem with prior information such that the weight assigned to a piece of information (that is used to assign a suspiciousness score) is fluid and evolutionary.</li><li>Finally, it introduced the most important concept that not all the failed (or successful) test executions provide the same contribution towards fault localization. More precisely, with respect to a piece of code, the contribution introduced by the first failed test that executes it in computing its likelihood of containing a bug is larger than or equal to that of the second successful test that executes it, which is larger than or equal to that of the third successful test that executes it, etc. The same applies to the successful.</li></ol><p>This is why our paper is titled as such — it <em>evolves from simple intuition to more complex computations</em> and in doing so provides ‘a family of heuristics’ to achieve fault localization. When our heuristics were compared to the state-of-the-art at the time across multiple types of programs, spanning multiple programming languages and faulty versions, they outperformed those fault localization techniques compared to. As a result, it became the state-of-the-art. This is one of the reasons for it being cited as much as it has been — it has represented/represents a benchmark that newly proposed techniques must assess themselves relative to (especially because of the comprehensive results we provided across multiple programs from varied sources). Additionally, our paper facilitated other research in the area because our heuristics offered extension points both in terms of manipulating weights as well as adding on to the basic intuitions.</p><p>To conclude, not only can this family of heuristics help programmers quickly (efficiently) locate bugs by examining a small percentage of code (effectively), but it also gives them the flexibilities to customize the heuristics that are appropriate for their projects in their working environments.</p><p>We are very grateful for the recognition that this paper has received over the last few years and are hopeful that this trend continues. In fact, we fully expect the general trend to continue given the advances that have been made lately in related areas, such as cloud computing and internet-of-things to name a few. Storage capacities have greatly increased and the speed at which data can be collected and processed has also increased, especially with ever-expanding global connectivity. Nevertheless, the fundamental problem still exists — given this wealth of data, how do we use it to pinpoint the faulty code when things go wrong? Our proposed heuristics offer a way of looking at things that is almost as fundamental as the very problem itself. That is why we believe this paper, and related works both published by us and by others, will continue to have a considerable impact in the years to come.</p><p>We thank very much the Journal of Systems and Software for publishing this paper and nominating it as one of the finalists for the 2020 Most Influential Paper.</p><p>Remark: A conference version of this paper, <em>Effective Fault Localization using Code Coverage</em>, with preliminary results was published at the 31st IEEE International Computer Software and Applications Conference<em> </em>in 2007 and received the Best Paper Award from the conference.</p><p><a href="https://dblp.org/pid/28/7718.html">Vidroha</a> and <a href="https://dblp.org/pid/w/WEricWong.html">Eric</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=373b50a17ed5" width="1" height="1" alt=""><hr><p><a href="https://medium.com/jss-papers-of-the-year/a-family-of-code-coverage-based-heuristics-for-effective-fault-localization-373b50a17ed5">A family of code coverage-based heuristics for effective fault localization</a> was originally published in <a href="https://medium.com/jss-papers-of-the-year">JSS Papers</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[About: “A systematic and comprehensive investigation of methods to build and evaluate fault…]]></title>
            <link>https://medium.com/jss-papers-of-the-year/about-a-systematic-and-comprehensive-investigation-of-methods-to-build-and-evaluate-fault-364942cb8a35?source=rss-4f58beca6fe3------2</link>
            <guid isPermaLink="false">https://medium.com/p/364942cb8a35</guid>
            <dc:creator><![CDATA[The Blog of Systems and Software]]></dc:creator>
            <pubDate>Wed, 03 Feb 2021 23:04:08 GMT</pubDate>
            <atom:updated>2021-02-25T13:49:13.736Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="A systematic and comprehensive investigation of methods to build and evaluate fault prediction models" src="https://cdn-images-1.medium.com/max/1024/1*yXTwOnHjNtcgKPC4-rxyIA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/140/0*mUXjThC73mAQCAkB.jpeg" /></figure><p><strong>Finalist and candidate for JSS MIP grand prize</strong></p><p>At the time we worked on this project, around 2008–2009, there were many papers on predicting fault-proneness of software components. However, publications tended to focus on specific and sometimes narrow aspects, often using public databases or open source projects. In this paper, in collaboration with Telenor, a major Telecom player in Europe, we wanted to investigate the best ways to build and apply such fault-proneness models in a realistic industrial context and in a comprehensive manner.</p><p>We therefore systematically assessed three aspects on how to build and evaluate fault-proneness models in the context of large Java legacy system development projects: (1) we compared many data mining and machine learning techniques to build fault-proneness models, (2) we assessed the impact of using different metric sets such as source code structural measures and change/fault history (process measures), and (3) we compared several alternative ways of assessing the performance of the models.</p><p>The results of the study indicated that the choice of fault-proneness modeling technique has limited impact on the resulting classification accuracy or cost-effectiveness. There are however large differences between the individual metric sets in terms of cost-effectiveness, and although the process measures are among the most expensive ones to collect, including them as candidate measures significantly improves the prediction models compared with models that only include structural measures and/or their deltas between releases. Further, we observe that what is considered the best model is highly dependent on the criteria that are used to evaluate and compare the models.</p><p><em>Originally published at </em><a href="https://lionelcbriand.medium.com/about-a-systematic-and-comprehensive-investigation-of-methods-to-build-and-evaluate-fault-56a900a2382b?source=---------0----------------------------"><em>https://lionelcbriand.medium.com</em></a><em> on February 3, 2021.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=364942cb8a35" width="1" height="1" alt=""><hr><p><a href="https://medium.com/jss-papers-of-the-year/about-a-systematic-and-comprehensive-investigation-of-methods-to-build-and-evaluate-fault-364942cb8a35">About: “A systematic and comprehensive investigation of methods to build and evaluate fault…</a> was originally published in <a href="https://medium.com/jss-papers-of-the-year">JSS Papers</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>