SUMMARY The architect of a large, evolving system may wish to revise its decomposition from time to time; for instance, because the structure has deteriorated over time, certain components need to be outsourced to another site. One way to assess the current decomposition is to consider the past evolution, searching for components that often changed together. We iteratively devised and implemented a process for doing so at Philips Healthcare MRI. In this paper, we describe the lessons learned on how to effectively support architects to improve their system decomposition. Copyright © 2011 John Wiley & Sons, Ltd.

Several defect prediction models have been proposed to identify which entities in a software system are likely to have defects before its release. This paper presents a replication of one such study conducted by Zimmermann and Nagappan on Windows Server 2003 where the authors leveraged dependency relationships between software entities captured using social network metrics to predict whether they are likely to have defects. They found that network metrics perform significantly better than source code metrics at predicting defects. In order to corroborate the generality of their findings, we replicate their study on three open source Java projects, viz., JRuby, ArgoUML, and Eclipse. Our results are in agreement with the original study by Zimmermann and Nagappan when using a similar experimental setup as them (random sampling). However, when we evaluated the metrics using setups more suited for industrial use -- forward-release and cross-project prediction -- we found network metrics to offer no vantage over code metrics. Moreover, code metrics may be preferable to network metrics considering the data is easier to collect and we used only 8 code metrics compared to approximately 58 network metrics.

Changing source code in large software systems is complex and requires a good understanding of dependencies between software components. Modification to components with little regard to dependencies may have an adverse impact on the quality of the latter, i.e., increase their risk to fail. We conduct an empirical study to understand the relationship between the quality of components and the characteristics of their dependencies such as their frequency of change, their complexity, number of past failures and the like. Our study has been conducted on two large software systems: Microsoft VISTA and ECLIPSE. Our results show that components that have outgoing dependencies to components with higher object-oriented complexity tend to have fewer field failures for VISTA, but the opposite relation holds for ECLIPSE. Likewise, other notable observations have been made through our study that (a) confirm that certain characteristics of components increase the risk of their dependencies to fail and (b) some of the characteristics are project specific while some were also found to be common. We expect that such results can be leveraged for use to provide new directions for research in defect prediction, test prioritization and related research fields that utilize code dependencies in their empirical analysis. Additionally, these results provide insights to engineers on the potential reliability impacts of new component dependencies based upon the characteristics of the component.

The study of Software Configuration Management (SCM) has so far focused on the use of tools, SCM models, strategies, patterns or best practices. There are few industry studies on how an organization actually practices file branching and merging. In this empirical study at Océ, we have observed that some developers create branches freely without regards to their consequences on merging. This contradicts recommended best practices and SCM processes. So we investigate if there are hidden costs in propagating or merging changes at Océ. The investigation led us to understand that branching and merging can be done freely under certain circumstances to provide development concurrency. However, some files cannot be branched freely and it may be better to use recommended practices to edit them. Some roles were also noted to be more affected by branching of files.

In an agile environment, where the architect's role is not crisply defined, the architect may have to deal with a number of issues that arise during development and are considered architecture-related. Such issues range from feature requests and enhancements through to defects found during testing. Architect specifications that turn into issues boomerang the architect to resolve them. In a case study we looked at issues entered into the issue tracking system that were assigned to architects to understand the root causes for their inception. We found that many of the avoidable issues can be circumvented by better communication between the architect and other stakeholders, and by small adaptations in the development process.

For many software projects, bug tracking systems play a central role in supporting collaboration between the developers and the users of the software. To better understand this collaboration and how tool support can be improved, we have quantitatively and qualitatively analysed the questions asked in a sample of 600 bug reports from the MOZILLA and ECLIPSE projects. We categorised the questions and analysed response rates and times by category and project. Our results show that the role of users goes beyond simply reporting bugs: their active and ongoing participation is important for making progress on the bugs they report. Based on the results, we suggest four ways in which bug tracking systems can be improved.

A widely shared belief in the software engineering community is that stack traces are much sought after by developers to support them in debugging. But limited empirical evidence is available to confirm the value of stack traces to developers. In this paper, we seek to provide such evidence by conducting an empirical study on the usage of stack traces by developers from the ECLIPSE project. Our results provide strong evidence to this effect and also throws light on some of the patterns in bug fixing using stack traces. We expect the findings of our study to further emphasize the importance of adding stack traces to bug reports and that in the future, software vendors will provide more support in their products to help general users make such information available when filing bug reports.

In software development, bug reports provide crucial information to developers. However, these reports widely differ in their quality. We conducted a survey among developers and users of APACHE, ECLIPSE, and MOZILLA to find out what makes a good bug report. The analysis of the 466 responses revealed an information mismatch between what developers need and what users supply. Most developers consider steps to reproduce, stack traces, and test cases as helpful, which are, at the same time, most difficult to provide for users. Such insight is helpful for designing new bug tracking tools that guide users at collecting and providing more helpful information. Our CUEZILLA prototype is such a tool and measures the quality of new bug reports; it also recommends which elements should be added to improve the quality. We trained CUEZILLA on a sample of 289 bug reports, rated by developers as part of the survey. The participants of our survey also provided 175 comments on hurdles in reporting and resolving bugs. Based on these comments, we discuss several recommendations for better bug tracking systems, which should focus on engaging bug reporters, better tool support, and improved handling of bug duplicates.

In a survey we found that most developers have experienced duplicated bug reports, however, only few considered them as a serious problem. This contradicts popular wisdom that considers bug duplicates as a serious problem for open source projects. In the survey, developers also pointed out that the additional information provided by duplicates helps to resolve bugs quicker. In this paper, we therefore propose to merge bug duplicates, rather than treating them separately. We quantify the amount of information that is added for developers and show that automatic triaging can be improved as well. In addition, we discuss the different reasons why users submit duplicate bug reports in the first place.

In software engineering experiments, the description of bug reports is typically treated as natural language text, although it often contains stack traces, source code, and patches. Neglecting such structural elements is a loss of valuable information; structure usually leads to a better performance of machine learning approaches. In this paper, we present a tool called infoZilla that detects structural elements from bug reports with near perfect accuracy and allows us to extract them. We anticipate that infoZilla can be used to leverage data from bug reports at a different granularity level that can facilitate interesting research in the future.

In software development, bug reports provide crucial information to developers. However, these reports widely differ in their quality. We conducted a survey among developers and users of APACHE, ECLIPSE, and MOZILLA to find out what makes a good bug report.

Developers typically rely on the information submitted by end-users to resolve bugs. We conducted a survey on information needs and commonly faced problems with bug reporting among several hundred developers and users of the APACHE, ECLIPSE and MOZILLA projects. In this paper, we present the results of a card sort on the 175 comments sent back to us by the responders of the survey. The card sort revealed several hurdles involved in reporting and resolving bugs, which we present in a collection of recommendations for the design of new bug tracking systems. Such systems could provide contextual assistance, reminders to add information, and most important, assistance to collect and report crucial information to developers.

Several studies have been conducted to determine if company-specific cost models deliver better prediction accuracy than cross-company cost models. However, mixed results have left the question still open for further investigation. We suspect this to be a consequence of heterogenous data used to build cross-company cost models. In this paper, we build cross-company cost models using homogenous data by grouping projects by their business sector. Our results suggest that it is worth to train models using only homogenous data rather than all projects available.