stuff that happened:
visited jmm, a large math conference that happened to be held in sf. there were some interesting talks and it was good to see a few of my math friends again, but the conference also drove home the point that i really, really don’t belong in math academia. it’s simply not a good fit culture-wise or interests-wise. in the last few years i’ve occasionally wondered what if i kept pursuing math research in college and became a maybe-successful academic and it’s nice to get some more assurance that, actually, my life is better now and it was worth leaving a thing i might’ve been very good at
while at jmm i listened to a talk by terence tao on a subject i knew nothing about (the parity problem in sieve theory). it was incredible - he is extremely good at explaining things just enough for you to begin understanding a complicated topic, without getting slowed down by extra details. listening to him made me think more about the goal of these kinds of talks - i suppose it is to take the audience to the edge of the world as quickly as possible, and then keep them there
started reading nonviolent communication at my boss’s suggestion. i am optimistic that it will help me address some of my longer-running issues and have many thoughts but writing them is taking too long so i will put them in a later post
my social life improved a lot, for two main reasons
stopped trying to meet new people at tech parties. i mostly don’t go to tech parties now, and when i do go i just talk to my friends there. it is simply not worth talking to strangers at most tech parties because the proportion of people looking to network is too high (though strangers at other events are fine)
my friends started hosting weekly coworking sessions and the routine is very stabilizing! i no longer feel obligated to schedule hangouts all the time because i know that regardless of what happens i’ll see people on sunday morning
stuff related to work:
went through the lectures for cathy wu’s rl class. it was helpful to understand the historical context and theory for many of the developments in rl, which otherwise feel a bit unmotivated
started doing some work related to implementing 3d parallelism at my job. it’s very interesting and it’s also the first sizable project i’ve had a lot of ownership over and as a result it sort of took over my life for a few weeks. i’ve realized there is a kind of programming - very detail-oriented and somewhat mathematical in nature, in the same vein as performance engineering or writing zk circuits - that i find to be a lot of fun and am quite good at, and which most people avoid. on one hand it’s convenient that i enjoy a thing that is useful and that very few people can do; on the other hand i don’t really think this is a skill i want to keep getting better at and it’s easy for me to spend a lot of time on low-impact projects containing this kind of work. or, to put it differently, i don’t think i was put on this planet just to move tensors around faster and i’d be a bit sad if that was the main thing i worked on, even though i do like doing it. i’ll try to take a step back soon
stuff watched:
watched harvey, a longstanding john green recommendation. it was very sweet and timeless - it’s been a while since i enjoyed a black-and-white movie this much
rewatched kung fu panda on a flight and was pleasantly surprised by how good it was. every line from oogway is gold
watched shortcomings, a hilarious movie with one of the most (deliberately) unlikable protagonists of all time
finally watched good will hunting and cried a lot. the portrayal of math and mit was very amusing but aside from that i thought the movie was close to flawless. after the intro it’s just an avalanche of extremely charged scenes, every single one concerning a topic i’ve thought a lot about: discussion of knowledge vs lived experience. tomorrow i’ll be fifty. academics arguing about success. it’s not your fault. god what a masterpiece