ICSE 2024

Last week, I visited the beautiful city of Lisbon, Portugal to present my first lead-author paper, Semantic Analysis of Macro Usage for Portability, at ICSE 2024.

I attended both the main conference and the CHASE workshop, and while most of the speakers and their presentations did not impress me, there were a number of exceptional presentations. Note that I wasn’t able to attend all the talks (that would be impossible since multiple talks were begin given simultaneously throughout the event), but out of the ones I attended, these were by far the best:

• A Journey Into the Emotions of Software Developers by Nicole Novielli: Dr. Novielli gave the opening keynote for CHASE, and it did not disappoint. Her insights into how developers’ emotions affect their productivity were both interesting and intuitive. For example, in her research she has found that developers tend to perform better when they’re happy and worse when they’re sad. I am currently helping organize an HCI study on the biometrics of developers while they debug code in group settings, and wanted to ask Dr. Novielli for some advice, but unfortunately did not get the chance.

• The Surprising Implications of Realism for Human Factors Research by Paul Ralph: It was surprising that one of the CHASE keynotes was a talk about philosophy! Dr. Ralph unfortunately arrived late, and his talk was initially hampered by technical difficulties, but despite these issues he still managed to give a captivating presentation. I honestly felt like I was listening to a Jordan Peterson talk at certain points, and I am now definitely more curious to learn more about Critical Realism and other schools of philosophy.

• Why People Contribute Software Documentation by Deeksha M. Arya: This was the first non-keynote talk at CHASE that really impressed me, and it ended up being one of the best talks I saw at all of ICSE. The presentation was carefully crafted with fun yet simple animations, and Deeksha had clearly practiced giving her presentation beforehand. I had the chance to talk with Deeksha after the talk and connected with her on LinkedIn, she was very nice!

• Code Impact Beyond Disciplinary Boundaries: Constructing A Multidisciplinary Dependency Graph and Analyzing Cross-Boundary Impact by Gengyi Sun: This talk stood out to me because Gengyi managed to be funny while also explaining her work very well.

• The Devil Is in the Command Line: Associating the Compiler Flags With the Binary and Build Metadata by Gunnar Kudrjavets: I’ll be honest, I’m not sure if I liked this talk or not. On one hand, Gunnar’s slides consisted almost entirely of text, which I normally don’t like because it divides my attention between listening to the speaker and reading the text on the slides. On the other hand, Gunnar is a clear communicator, and the text that was on her slides was short, simple, and obviously relevant to what she was saying. I didn’t really understand the point of this talk (can’t just export compiler flags to a compile_commands.json file by using CMake or by intercepting a build system with bear or scan-build?), but I am curious to attend more of Gunnar’s presentations.

• Classifying Source Code: How Far Can Compressor-based Classifiers Go? by Zhou Yang: Zhou was a great and humorous presenter.

• An Ensemble Method for Bug Triaging using Large Language Models by Atish Kumar Dipongkor: Full disclosure, Atish and I are friends and work together in the same lab at UCF. Personal bias aside, Atish is still an amazing presenter. He is passionate about his work and can convey complex topics clearly and concisely.

• Using an LLM to Help With Code Understanding by Daye Nam: Daye’s presentation was excellent. Similar to Deeksha’s presentation, Daye’s was masterfully crafted and she presented her work very well.

• Predicting open source contributor turnover from value-related discussions: An analysis of GitHub issues by Jack Jamieson: I really liked Jack’s presentation because it got straight to the point and was readily comprehensible.

• My own talk!

I would recommend keeping an eye on all these presenters, and if you get the chance to attend their future talks, go for it! I know I will :)

Finally, I realize that in this post I haven’t really described what I consider to be a good presentation. That’s because I plan on dedicating a future post to that topic. I plan to have that post out by the end of May, and when I do I’ll link it here. Basically, all the talks I’ve listed in this post follow most, if not all, of the tenets I’m going to outline in that future post.