Breakfast. Read some papers for my project with Sid. Sanjoy Dasgupta continues to impress with terrific writing. I really enjoyed his paper on active learning. Before the meeting, wrote up a little summary of ideas for Sid and Artur, then scarfed down some leftover coq au vin and half a burrito.
Meeting was good (I think!). Still not clear how to assess. Sid and Artur hadn’t been able to read the email beforehand, but if I make a habit of these little summaries, maybe it can save us a little time so we can jump right in. (Apparently Amazon makes you write up your discussion points and ideas before a meeting, which I think is very cool.) Sid corroborated that the Dasgupta paper (as I’d unfortunately realized) wasn’t immediately applicable to our nearest neighbors problem. After that, pretty open ended discussion. We still haven’t converged on a topic, but hopefully we’ll be able to talk to the law prof I contacted soon and get some feedback on our ideas.
After the meeting, stepped outside into a downpour for a walk. Ended up enjoying the walk so much that I went all the way to BTC, did a little meditation on the beach, and came back nearly 2 hours after I’d left. Worked out, took a shower, caught up on some reading, replied to Joanna about a waitbutwhy explainer which seems to be missing the key points of Elephant in the Brain, and replied to an email thread with Alexey Guzey about why there aren’t more Elon Musks out there. Alexey’s model: 99th percentiles of raw brainpower, energy, ambition, persistence and creativity. I agree with that assessment, but i also argued environmental factors and the uniqueness of Elon’s ambitions (managing to avoid typical status games, get-to-Mars ethos). Alexey is the fellow who debunked Why We Sleep among other heroics, and so it’s pretty cool to be just chatting with him.
Popped in on Dad who was writing a Smalltalk parser. Had a fun discussion about BNF grammar vs. regular expressions (he’d reinvented a mixture of the two (with the expressivity of BNF)) while hacking. We complained about the poor naming of new ideas in computer science and how this creates an implicit barrier to thought (VC dimension anyone?) Can’t we ever rename anything?
Dinner: Mom’s delicious pot roast and sis M’s delightful potato crisp thingies. Really fantastic. Afterwards, zoned out and read some more of the Simons book. Watched Parasyte and the Office and hit the hay.