None of the talks were that exciting to me today. Martin Bligh talked about surviving out of memory situations--NFS and iSCSI, for example, both may need to allocate memory in order to handle writes, so if the reason we're writing is in order to free memory, we can easily deadlock. It seems to a hard problem that no one's really worked on enough yet, but I suppose it's at least good to know that people are aware of it....
This is the third year I've attended OLS. Every time they've had an opening night reception. They always attempt to have a couple of speakers, too. But this is in a big evenly lit room containing a bunch of people who haven't seen each other recently, with plenty of drinks and snacks. The result--people snacking and talking throughout the speeches--is completely predictable. Will the organizers ever learn anything from this?
Afterwards Trond, Olaf and I met Beepy at a nearby bar. The only other people there were the bartender and her mom and dad. They all had Russian accents, and there was a bunch of Russian memorabilia on the walls that I didn't look at closely. Beepy showed us some of his photographs after the mom helped him hook up to the wireless network.