Software Archeology @ RIT
Recent Progress
In the past week, our sub-teams have continued to make progress in the areas of NLP, first-ownership, and bugs research. Although work has been somewhat slower than in the past, we have continued to develop our code base in these areas and are optimimistic about what interesting metrics our research in these topics will yield.
Recently in the first-ownership sub-team, there has been progress as far as completing the script and loading the data into our database. Our team is currently working on refining the script further as well as writing verify tests on the data. An example refinement is changing the script output to include a date. Initially the script output was a CSV (Comma-separated value) file of developer, filepath, and commit hash, where the commit hash was the reference to the commit in which the developer was first added as an OWNER on the filepath. We are including the date on which they were added in order to decrease the number of calls we have to make later. By that I mean that when we are doing comparisons of when a developer made their first ever commit and when they were added as an OWNER, we can retrieve the stored date they were added as an OWNER instead of having to make a call and get the date associated with the commit hash. Small changes like this improve the run time and overall quality of our code base.
Plans for Spring
Our plans for continuing this research in the spring include more agressive research pace and a variety of papers. Firstly, our advisor Professor Meneely will be much more available to assist and meet with our team members. Additionally, many team members that were brought on as new this past semester will still be on the project and will be able to continue research without having to be spun up. The team members that were new this semester have done a great job becoming acquainted with the project, and will no doubt develop some great findings to be published.
The minor changes to the team in the coming semester are as follows. Richard, Kayla, Alvaro, Rachel, and Brian are all still going to be here continuing to work on the project. Felivel will be out but still expects to be active in research. For myself, I will be gone on co-op and will not be participating in the research. Another new team member may be brought on depending on funding in order to balance out the team.
In the spring our team also has ambitious plans as far as publications and papers. We plan on putting out a journal paper in spring on our work so far. This paper may or may not include our research into OWNERS, depending on how long the paper is without OWNERS. The OWNERship research may be developed into a separate paper. Decisions on this are still pending. Brian’s work on the NLP portion of the research will also have its own paper, which will delve in-depth into his work using NLP on the project this entire semester. NLP and OWNERS may end up being combined if there are interesting metrics/data, but this is also up in the air. Finally, Felivel will have a separate paper on his work with the bugs and vulnerabilities on this project. Overall we are very excited about the number of papers we expect to be written, as well as the breadth of the paper topics.
Until Next Time
As I am leaving for the semester and will not be returning to school until next fall, I would like to thank CREU for their funding and support. I would also like to say it’s a been a pleasure being a part of this research team. Over the different semesters it has included a variety of brilliant, motivated team members including our advisor, without whom none of this would have been possible. I have grown and learned a lot as a part of this team, and I look forward to seeing all of the unique and innovative results of this research project.
This project is sponsored by the Collaborative Research Experiences for Undergraduates, funded by the National Science Foundation.