Visit to Northeastern University

A few months ago, I was pleased to receive an e-mail from professor Jan Vitek (chair of the ACM SIGPLAN) inquiring about the status my basic block versioning work. He had been on the committee of a conference to which my advisor and I had submitted a paper (which was unfortunately rejected). Following some discussion, Jan Vitek invited me to visit Northeastern University to meet his research group and give a talk about my PhD work.

This Monday, I flew to Boston to visit the Programming Research Laboratory. I had the opportunity to exchange ideas with several PhD students as well as established researchers in the PL and compiler research fields, including Jan Vitek himself, Matthias Felleisen and Olin Shivers. Topics discussed were quite broad, ranging from graduate studies to PL design to data-parallel programming to compiler verfication to actors and JIT compiler imlementation.

The talk itself went quite well and elicited much participation. I have to say that the act of creating and organizing the slides for the talk was pretty helpful for me. I’ve been able to come up with better examples to explain basic block versioning and the way incremental code generation is implemented in Higgs. It also gave me an opportunity to talk about my recent work on typed shapes. The general feeling I’ve been having lately is I explain my research much more effectively when giving talks than when describing it in papers.

I exchanged ideas with Jan Vitek regarding hybrid type analyses for optimizing dynamic languages (he is working on an optimizing VM for the R language). Jan Vitek also gave me precious feedback on how to improve my basic block versioning paper. I’m currently working on an updated submission for the upcoming ECOOP conference. Regardless of whether the new paper gets in or not, it will be much improved with additional data to support our claims.

My hosts were very well-organized which made the whole experience quite pleasant. A meeting schedule had been planned for me and a hotel room reserved when I arrived. I got a little scared Tuesday evening when my taxi was over 30 minutes late, putting me in serious danger of missing the last flight back to Montreal, but I was lucky enough to run into Carla E. Brodley, dean of CCIS, who offered me to share an uber ride with her. My only regret regarding this short trip is that I did not get to try Boston clam chowder. Maybe next time.