Tuesday, April 8, 2014

ASPLOS Firsts

This was not my first trip to ASPLOS, but it did have a few other firsts I noticed:

I believe this was the first ASPLOS with an "approximate computing" session, a topic that has a soft stop in my research heart. There were two talks—from James Bornholt of ANU/MSR and Mehrzad Samadi of Michigan—both which more than justified their placement in the first session slot. Here's hoping that this session topic stays at ASPLOS for the next few years.

This was also the first time I can remember that the WACI session actually made anyone laugh. There was a healthy mix of ridiculousness and legitimately provocative ideas. Notably, Josh Smith's keynote was  shockingly ambitious while remaining  legitimately focused. In recent years, the session has listed a little too much toward the serious side—I hope the lightheartedness this year carries forward to future WACIs.

I also think this was the first ASPLOS with a lightning session and poster for every paper (something I first saw when Onur Mutlu chaired ASPLOS 2012). The lightning session is an incredibly valuable addition to the conference. It's impossible to catch every talk (even at a single-tracked conference, I contend!). So having one session where everyone sees every paper once has benefit disproportionate to the time taken. The poster session was perhaps less valuable since there was plenty of unstructured time already to follow up with authors. But I hope the lightning session is here to stay.

No comments:

Post a Comment