Week 11 - Identifying Questions Current Efforts and Related Work

Current Questions

As we draw near to the end of the semester, we are beginning to narrow down the questions we aim to answer with the statistical data and modeling we have generated thus far. The questions include the following:

• Was a correction made as a result of the code review?
• What types of words are used in inspection security fixes?
• What are the measurable components of an effective code inspection?
• Did a reviewer overlook a problem in the code?
• What is the experience of any particular developer?
• Are the invited reviewers participating?
• Who are the security experts in Chromium?
• What is the optimal amount of conversation surrounding a particular code review?

The answer to these questions involve a set of metrics we have or are currently generating. While this is not a comprehensive list this represents the direction we are heading with our current efforts.

Question Development and Finalizing

The multitude of data that the Chromium project provides is so vast and grand that the possible metrics are almost endless. In the same breath, the amount of questions are also endless. In the beginning of our research study we generated preliminary questions that we would initially aim to answer. As time went on we naturally began to produce more. Later down the line, as we drew closer to the end, we began a filtering process for the questions. We excluded questions that were maybe irrelevant to the current iteration and put on ice questions that may involve more effort than allowed in the current iteration.

Related Work

Another professor from our Software Engineering department referred a related research article entitled, The Impact of Code Review Coverage and Code Review Participation on Software Quality. The research article was authored by students from Queen’s University in Canada and Kyushu University in Japan. The paper was based on study that is very similar to ours. They were looking at the Qt, VTK,and ITK projects while we are using the Chromium project. The paper assisted us in identifying many metrics and even answered questions we were aiming to answer already.

This project is sponsored by the Collaborative Research Experiences for Undergraduates, funded by the National Science Foundation.