PLDI 2022

This past week I traveled to sunny San Diego to attend the 2022 edition of the Programming Language Design and Implementation conference (PLDI 2022). I spoke with many intelligent individuals and sat in on some great presentations.

I originally signed up for PLDI so I could attend PLMW, the collocated programming languages mentoring workshop for new PhD students, and to meet Mike Hicks. My advisor and I met with Dr. Hicks virtually a few times last year to brainstorm ideas for my current research project (about automatically transforming C preprocessor macros to C functions), and I wished to ask him in-person for some guidance on how to formally verify the correctness of transforming macros to functions. He recommended that I collaborate with others who are well-versed in formal verification, and was kind enough to give me some specific names to seek out.

I also met other well-known academics in the PL community, such as John Regehr and Matthew Flatt. I asked Dr. Regehr how he would go about proving the correctness of my project’s transformation, and he said he would extensively test it (or, as Dr. Hicks had advised, rely on someone more knowledgeable in formal verification to prove that it is correct). This response makes sense coming from Dr. Regehr, since his research largely focuses compiler testing and fuzzing. I was also very excited to meet Dr. Flatt; he understands macros extremely well, so I was hoping he would be able to give me some novel insights into my problem. While he was able to immediately recognize some of the hidden challenges of my problem that have taken me weeks to realize, he could only warn me that if I wanted to formally prove that correctness of transforming macros to functions, I would probably be stuck doing formal verification forever. Oh well, it was worth a shot anyway. On the plus side, I got to hear Dr. Flatt talk about his work on Rhombus, the successor to Racket.

Finally, met some other great PhD students such as Aurèle Barrière, who researches just-in-time compilation at IRISA, and Davis Silverman, who is currently developing the Sin Scheme compiler.

In the evenings I had time to relax and explore San Diego with my girlfriend, who accompanied me for this trip. On Friday we walked down to Pacific Beach, and on Saturday we traveled to Balboa Park. We didn’t visit the San Diego zoo, which is located in the park, but we did go to the Japanese Friendship Garden, which was absolutely beautiful.

Overall, I really enjoyed my time in San Diego. Now that I’m back in Orlando though, it’s time to get back to work - programs and papers don’t write themselves!