Software Archeology @ RIT

[ar·che·ol·o·gy] n. the study of people by way of their artifacts
Week 3 - Continuing Chromium

13 Sep 2014

Recent Team Changes and a New Semester

It’s the start of another great fall semester doing research! Since last fall there have been a lot of changes to the project, including changes in research team members. There have been members graduating or moving on to different projects, and new members enthusiastically coming in to pick up the work! As of this fall, the team now is comprised of a total of 8 team members including Professor Meneely. Current members are Brian Spates, Kayla Davis, Richard Kalimba, Felivel Camilo, Alvaro Pareja-Lecaros, Rachel Catipovic, and myself. The members funded by CREU include Alvaro, Rachel, Kayla, and myself.

This is a much larger team than the project has had before. We hope that having more people on the project will bring more perspectives, knowledge, and experiences. In the end having this many people on the project will allow us to accomplish even more research than we have in past semesters.

Over the Summer

This summer a lot of progress had been made on the project by Professor Meneely and Kayla. Some of their hard work included but is not limited to updating the code review parser, working with sheriff data, and more than doubling the amount of ‘verify’ unit tests our project contains.

Additionally, our team was able to submit and receive approval on a paper, which is extremely exciting! The paper was submitted to the SSE’14 workshop at FSE’14. It is entitled “An Empirical Investigation of Socio-Technical Code Review Metrics and Security Vulnerabilities” and authored by Andrew Meneely, Alberto C. Rodriguez Tejeda, Brian Spates, Shannon Trudeau, Danielle Neuberger, Katherine Whitlock, Christopher Ketant and Kayla Davis.

These Past Few Weeks

During this and previous weeks at the start of the semester, our team has begun to spin back up and jump back into research. Newer members are becoming increasingly familiar with the Chromium project and our code base, and are joining the more senior members in continuing research on various topics/issues in our GitHub backlog.

Brian has been working on research into how natural language processing (NLP) can be used on the project to come up with more interesting metrics and perhaps measure transfer of knowledge, a topic discussed last year but not deeply researched. Richard and myself are looking into OWNERS files in the chromium project, and have been brainstorming metrics for that. Additionally, we have been working on the notion of ‘teams’ within Chromium. We are still figuring out how to organize those and think of useful metrics from that. Felivel has worked on parsing the relationship between commits and bugs. Kayla has been working on matching ellipsized developer emails scraped on reviews to full developer emails in our database. Finally, Alvaro and Rachel have been working on updating our vulnerability data set.

Plans for this Semester

Our long-term goals for this semester include (but certainly are not limited to) advancing metrics on OWNERS files, implementing NLP on the project, updating our data set, improving code performance, and continuing to come up with new, interesting metrics and relationships within our data.

« Home