Pasteur's Cube

I like using Twitter. Here are some policies I want to aim to follow in order to make Twitter more useful for myself and others. Please hold me accountable! * "Not a hater" policy - I will avoid dunking on people, or amplifying dunking on people, unless I really

Content warning: discusses race, links to a piece that contains an unredacted racial slur, written while I am very angry Nick Bostrom is not a person I have ever spent much time thinking about and is a person I’ve never interacted with. I am grateful to him for his

OVERALL: * Dem Senate control -- Unclear but slight no (46% chance) * Dem House control -- Somewhat leaning no (27% chance) BALANCE OF POWER: * D House + D Senate: Somewhat leaning no (21% chance) * R House + D Senate: Somewhat leaning no (24% chance) * D House + R Senate: Likely no (6% chance) * R

OVERALL: * Dem Senate control -- Unclear but slight no (49% chance) * Dem House control -- Somewhat leaning no (33% chance) BALANCE OF POWER: * D House + D Senate: Somewhat leaning no (26% chance) * R House + D Senate: Somewhat leaning no (23% chance) * D House + R Senate: Likely no (7% chance) * R

My wife, Samantha Wildeford, passed away this week. While I'm sad that she's gone, I'm so happy she could pass away with myself and all the others who loved her. Her life wasn't always kind to her, but she filled it with

So what exactly can and can’t language models do? What's the least impressive thing GPT-4 won't be able to do? What will GPT-4 be incapable of? BIGBench is kind of a way to figure this out. BigBench, aka “The Beyond the Imitation Game” Benchmark, is

Wikipedia has a cool feature where they summarize various decades in pictures. These are really fun to see, to get a sense of history - especially a sense of what events were happening at the same time. 2020s (so far) 2010s 2000s 1990s 1980s 1970s 1960s 1950s 1940s 1930s 1920s

Here are some examples of questions I forecasted on last year, my rationale for the forecasts I made, and how I feel like the forecast turned out. All forecasts were made on 2021 Aug 5. I think doing this kind of practice (writing your rationale out and then examining how

Post co-written with Amber Dawn - In 1950, Alan Turing suggested that machines might one day be able to deceive a human in an ‘imitation game’ - a game where a human judge interacts with both a computer and a human and has to distinguish which is which. Since then,

“It Doesn’t Have to be Crazy at Work” by Jason Fried and David Heinemeier Hansson is a brazen ad for working at Basecamp, the company founded by the book’s authors. But if you can get past that, it has some good advice for those managing companies and designing

Cal Newport previously wrote “Deep Work”, which outlined a rule for success - you do your best work when you are in a large uninterrupted distraction-free block of time called a “deep work block”. Newport’s thesis is that a single uninterrupted 3-4 hour block of focused work on your

Chess is about conventional warfare. But conventional warefare is boring. We need some… nukes. Battle Chess is like Chess, but every piece has a special ability. One of these abilities might involve a nuke. Disclaimer: I made this game when I was in Elementary school. It sounded cool. I make

I like Star Trek: Deep Space Nine and want to watch it. But there are several problems: (1) I don’t want to commit the time to watch 176 episodes (2) I’ve seen about a third of DS9 already but not in any particular order, so watching would be

Alec Stapp and Caleb Watney, in founding their Institute for Progress, declared that "Progress is a Policy Choice". I agree, but I'm also worried about something a bit more simple - in addition to making progress, we also have to survive[1]. This survival is a

Summary * I find there have been ten accidental nuclear incidents in history where I think there was some non-negligble chance that, had things played out differently, there could've been a nuclear exchange. * I think the chance of another accidental nuclear incident happening in the next decade (2022-2032) is

I've been blogging since I was 17 (I'm 30 now[1]) and it has always been a really critical part of my intellectual development. Blogging helped me sort out my personal beliefs and philosophy, bringing me through atheism, and then an interest in philosophy, and then

The book “Managing to Change the World: The Nonprofit Manager's Guide to Getting Results” by Alison Green and Jerry Hauser comes highly recommended from a wide variety of top executive directors of non-profits, and after reading it, I can say these positive recommendations are entirely justified. Don’t

It's been awhile since I've posted some sort of links post. Maybe this can come back more for 2022? I read a lot of articles in the past six months. Here I’ve decided on my top ten favorites (in descending order). I present each one,

Throughout the New York City mayoral race, between January and June 2021, I made multiple predictions about the fate of various political candidates. I think these predictions are my most successful set so far in my forecasting career, greatly outperforming the conventional wisdom and netting >200% ROI on prediction

Ultralearning: Master Hard Skills, Outsmart the Competition, and Accelerate Your Career by Scott Young is all about how to learn things well and quickly. In summary, ultralearning = specific goal + specific benchmark + specific timeframe + significant upfront and periodic investment in learning how to learn + intense amount of time + spaced schedule based

“Working Identity: Unconventional Strategies for Reinventing Your Career” by Hermina Ibarra is a book oriented around case studies of people transitioning careers. At its heart, it has a reasonable methodology and it contains nuggets of wisdom that I think are useful …and maybe even “unconventional”. However, it shrouds this in

Peoplware is a popular book about management for software projects, though I imagine it is applicable to nearly every project. The core idea is that managers are managers of people, not technology, yet most people don’t manage that way.  Most software projects that go wrong go wrong because of