Most of the posts on this blog are written at least a month before they're posted, sometimes much longer. I try to post things that are worth posting even so, hence the name "Cold Takes."
Here are highlights so far, which should give a sense for what the blog is like.
The "most important century" series
The "most important century" series argues that the 21st century could be the most important century ever for humanity, via the development of advanced AI systems that could dramatically speed up scientific and technological advancement, getting us more quickly than most people imagine to a deeply unfamiliar future. It's available in web, audio, PDF and Kindle formats.
Some other highlights so far
- Why talk about 10,000 years from now? summarizes one of the central attitudes of this blog: I try to imagine myself as a billions-of-years-old observer, thinking of hundred-year phenomena as "short-lived" and writing about the sorts of things that might still matter a very long time from now.
- Does X cause Y? An in-depth evidence review. There's an interesting theory out there that X causes Y. If this were true, it would be pretty important. So I did a deep-dive into the academic literature on whether X causes Y. Here's what I found. (Embarrassingly, I can't actually remember what X and Y are. Fortunately or unfortunately, I think this piece is correct for most (X,Y) that you might try to research.)
- Give Sports a Chance is the first in my "Cold Links" series: links that I like a lot, that are so old you can't believe I'm posting them now. It makes the case for why you might be interested in sports, even if you hate sports. If you're sold, here are my other sports posts.
- Gell-Mann Earworms. "Gell-Mann Amnesia" refers to constantly forgetting how unreliable news sources are; I use "Gell-Mann Earworms" for the opposite condition, in which "I can’t trust this" is constantly ringing in your ears as you read anything. Since this seems correct, how does one live with this condition?
- Summary of history (empowerment and well-being lens) presents a summary of human history in one table, through the specific lens of "empowerment and well-being": I consider historical people and events significant to the degree that they influenced the average person's (a) options and capabilities (empowerment) - including the advance of science and technology; (b) health, safety, fulfillment, etc. (well-being). Through this lens, the wars and power struggles that fill traditional history textbook barely matter; changes in science and technology, health, poverty, gender relations and slavery are all far more significant, leading to a different picture of history.
- Phil Birnbaum's "bad regression" puzzles. If you've ever wanted to see someone painstakingly deconstruct a regression analysis and show all the subtle reasons it can generate wild, weird and completely wrong results, click here.
Has Life Gotten Better? is an in-progress series asking: "What would a chart of average quality of life for an inhabitant of Earth look like, if we started it all the way back at the dawn of humanity?" Most analyses of this topic focus on the last few hundred years, but humanity is at least hundreds of thousands of years old.
"Where's today's Shakespeare?" is an upcoming series, which will argue that:
- Art and science seem to show similar "stagnation" patterns over time: the rate of new significant works doesn't seem to keep up with growth in the number of artists and scientists. "Where's today's Shakespeare?" refers to the puzzle: "Why is no 20th-century writer as acclaimed as Shakespeare, even though there are so many more people in the 20th century who have the education and opportunity to be writers compared to Shakespeare's time?"
- One explanation for this "stagnation" would be that there's a "golden age" in the past, and our society is doing various things to squash/discourage innovation that it didn't used to.
- Another explanation would be a "fruitpicking" model of science: ideas naturally get harder to find, as each idea is harder to improve on than the last. This would mean science is slowing down for "natural" reasons rather than because we're doing something wrong.
- I think the "fruitpicking" model is better supported by evidence than the "golden age" model - for both science and art. But:
- It's somewhat unintuitive that this kind of dynamic could apply to art. I'll discuss a bit how it could.
- The "fruitpicking" model has some unfamiliar implications of its own. It implies that artists and scientists might be best thought of as "discovering," or even (in art's case) "using up" ideas rather than "creating" them. This seems important for our general picture of how such people add value and what kinds of rights they should have to their intellectual property.
- It also implies that we should expect further stagnation by default; that this is very hard to avoid; and that our prospects for more creativity could depend more on things like population growth and intellectual property than on creative individuals or even culture.
Applied epistemology. "Epistemology" usually means having precise academic debates over things like the nature of knowledge. By contrast, I'm interested in topics like: how should we decide what to believe in a world where there's far too much information and analysis available for us to read any noticeable fraction of it?
- One post, "Minimal-trust reasoning," will go through what I have personally found the most helpful tool for this challenge: picking some questions to dig all the way to the bottom of personally, looking only for hard data and not deferring to experts, then using patterns I pick up from these exercises to form heuristics about whom and what to trust in settings where I know less.
- One post will ask: how far should "self-skepticism" go? Should I (a) doubt intuitions and rely on evidence? Or (b) doubt evidence and rely on people who know the evidence better? Or (c) worry that those people form a strange bubble out of touch with most people, and index my views to conventional wisdom? Or (d) forget about optimizing for correctness and instead try for a belief system that is harmlessly wrong 9/10 times and importantly right 1/10 times? (I lean toward a combo of (b) and (d).)
- One post will examine the pros and cons of the "Bayesian mindset" of handling all uncertainty via quantitative probability estimates and "betting."
"Future-proof" ethics. I'm interested in the question of what behaviors and choices about scarce resources today would make me least likely to be condemnable by future generations and most "ahead of the curve" on ethics.
I'm going to lay out the connections between the sort of ethical framework common in the effective altruism community (which often is highly concerned with animals and future generations) and the goal of being "ahead of the curve," and discuss both pros and cons of this school of ethics that I don't think are widely appreciated.