The Firefox Temporal Defect Dataset (FTDD)

The bug tracking repositories of software projects capture initial defect (bug) reports and the history of interactions among developers, testers, and customers. Extracting and mining information from these repositories is time consuming and daunting. Researchers have focused mostly on analyzing the frequency of the occurrence of defects and their attributes (e.g., The number of comments and lines of code changed, count of developers). However, the counting process eliminates information about the temporal alignment of events leading to changes in the attributes count. Software quality teams could plan and prioritize their work more efficiently if they were aware of these temporal sequences and knew their frequency of occurrence. In this paper, we introduce a novel dataset mined from the Firefox bug repository (Bugzilla) which contains information about the temporal alignment of developer interactions. Our dataset covers eight years of data from the Firefox project on activities throughout the project's lifecycle. Some of these activities have not been reported in frequency-based or other temporal datasets. The dataset we mined from the Firefox project contains new activities, such as reporter experience, file exchange events, code-review process activities, and setting of milestones. We believe that this new dataset will improve analysis of bug reports and enable mining of temporal relationships so that practitioners can enhance their bug-fixing process.

When using the dataset please cite the paper describing it:

@inproceedings{DBLP:conf/msr/HabayebMMBB15,
author = {Mayy Habayeb and
Andriy V. Miranskyy and
Syed Shariyar Murtaza and
Leotis Buchanan and
Ayse Bener},
editor = {Massimiliano Di Penta and
Martin Pinzger and
Romain Robbes},
title = {The Firefox Temporal Defect Dataset},
booktitle = {12th {IEEE/ACM} Working Conference on Mining Software Repositories,
{MSR} 2015, Florence, Italy, May 16-17, 2015},
pages = {498--501},
publisher = {{IEEE} Computer Society},
year = {2015},
url = {https://doi.org/10.1109/MSR.2015.73},
doi = {10.1109/MSR.2015.73},
timestamp = {Wed, 16 Oct 2019 14:14:52 +0200},
biburl = {https://dblp.org/rec/conf/msr/HabayebMMBB15.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}