Abstract
Due to the characteristics of the maintenance process followed in open source systems, developers are usually overwhelmed with a great amount of bugs. For instance, in 2012, approximately 7,600 bugs/month were reported for Mozilla systems. Improving developers’ productivity in this context is a challenging task. In this paper, we describe and evaluate the new version of NextBug, a tool for recommending similar bugs in open source systems. NextBug is implemented as a Bugzilla plug-in and it was design to help maintainers to select the next bug he/she would fix. We evaluated the new version of NextBug using a quantitative and a qualitative study. In the quantitative study, we applied our tool to 130,495 bugs reported for Mozilla products, and we consider as similar bugs that were handled by the same developer. The qualitative study reports the main results we received from a survey conducted with Mozilla developers and contributors. Most surveyed developers stated their interest in working with a tool like NextBug. We achieved the following results in our evaluation: (i) NextBug was able to provide at least one recommendation to 65% of the bugs in the quantitative study, (ii) in 54% of the cases there was at least one recommendation among the top-3 that was later handled by the same developer; (iii) 85% of Mozilla developers stated that NextBug would be useful to the Mozilla community.
Highlights
Due to the characteristics of the maintenance process followed in open source systems, developers are usually overwhelmed with a great amount of bugs
We present in this paper the following improvements over our first work on NextBug (Rocha et al 2014): (a) new features introduced in NextBug version 0.9, including recommendation filters and logging files; (b) an extended section presenting the tool and its architecture in more detail; (c) more tools presented and analysed in the related tools section; (d) a new quantitative study using the full dataset of 130,495 bugs; (e) a new section describing the qualitative study
The following comments show a sample feedback received from 66 Mozilla developers: “It would not be bad, as long as it could be configured to do the recommendations according to a set of parameters.” – R.C.* (We show only the initials from survey subjects to preserve their anonymity)
Summary
We evaluated the new version of NextBug using a quantitative and a qualitative study. We applied our tool to 130,495 bugs reported for Mozilla products, and we consider as similar bugs that were handled by the same developer. The qualitative study reports the main results we received from a survey conducted with Mozilla developers and contributors. Most surveyed developers stated their interest in working with a tool like NextBug
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
