Dark energy was assumed to be a constant force in the universe, both currently and throughout cosmic history. But the new data suggest that it may be more changeable, growing stronger or weaker over time, reversing or even fading away.
HN discussion here.
The Feynman Technique is a study method designed to help you gain a deep understanding of any topic. It consists of four steps: study, teach, identify gaps and simplify. The thinking model was originally developed by Richard Feynman, the Nobel prize-winning scientist and author of Surely You’re Joking, Mr. Feynman!. Today, his notes have found their way into popular culture and numerous articles on how to study better.
…
If you’re doing well, your experience will be interactive, and your study results will be challenged. You’ll be forced to clarify, explain in less or more detail, answer questions, unpack jargon and consider perspectives you didn’t even think of. If it’s not going so well, you’ll be talking to a wall for an hour because you’re still struggling too much with what you teach and how you teach it. This is why I think this step is a good place to ask yourself: Is your primary goal to learn about a subject with your student being a mere means to your end? Or do you actually want to teach them so you both benefit from the experience?
Step two, teaching the subject you want to learn to someone else, seems to be the big lever here.
But the joke goes that you do Bayesian reasoning by doing normal reasoning while muttering “Bayes, Bayes, Bayes” under your breath. Nobody - not the statisticians, not Nate Silver, certainly not me - tries to do full Bayesian reasoning on fuzzy real-world problems. They’d be too hard to model. You’d make some philosophical mistake converting the situation into numbers, then end up much worse off than if you’d tried normal human intuition.
For example, does Putin have cancer? We start with the prior for Russian men ages 60-69 having cancer (14.32%, according to health data). We adjust for Putin’s healthy lifestyle (-30% cancer risk) and lack of family history (-5%). Putin hasn’t vanished from the world stage for long periods of time, which seems about 4x more likely to be true if he didn’t have cancer than if he did. About half of cancer patients lose their hair, and Putin hasn’t, so we’ll divide by two. On the other hand, Putin’s face has gotten more swollen recently, which happens about six times more often to cancer patients than to others, so we’ll multiply by six. And so on and so forth, until we end up with the final calculation: 86% chance Putin doesn’t have cancer, too bad.
This is insanely long but really fascinating, like watching a 10-part Netflix documentary play out in a blog post. I’d love to see more in-depth cross-examination of debates like this.
And just look at this chart! Bayesian reasoning applied to the covid lab leak hypothesis:
Most social problems can be divided into two types: wide problems and narrow problems.
Many social ills are caused by wide problems being treated as though they’re narrow problems. When that happens, we have mistakenly engineered an avoidable disaster. And we do it all the time.
Wide problems have weak self-selection, weak selection pressure, creating a group that ends up being a broadly representative sample of the population. For example, a jury of peers, which is supposed to be representative of the general population.
Narrow problems require strong self-selection and strong selection pressures, turning into a group that ends up being extremely skewed relative to the population. For example, the Los Angeles Lakers basketball team, a highly selected group of elite players.
Consider the worlds of politics and the police, which are responsible for significant abuses of power from bad apples that harm society. Both realms are, generally speaking, best treated as wide problems.
In other words, it would be good for society if politics was a big tent, in which people from all walks of life—and a relatively normal set of traits—were those who made decisions about how we should be governed. Sure, you might want some weak selection pressures to ensure that politicians were more intelligent, educated, and compassionate than the general population. But you wouldn’t want the group to be too skewed, especially toward undesirable, anti-social traits.
What have we done instead? We’ve engineered social systems that have rabid self-selection and selection pressures on steroids for the worst possible traits to attract and promote all the wrong kinds of politicians.
This made me think a lot about how I know a lot of folks (including myself) who are interested in broad social ills like loneliness and the lack of community, and then point to successful communities like The Commons and ask why can’t we scale that to the broader society at large. But what if The Commons is a “narrow” solution that we’re trying to apply to a “wide” problem? That is, it’s a highly selected - both self- and purposefully - group of people who are quite unusual relative to the general population (highly agentic, motivated by the questions of community and belonging, lifelong-learner types). The same structures, incentives, approaches that work for The Commons wouldn’t necessarily work for different groups of people. “Scaling community” is practically an oxymoron imo because the whole point of community is that they are distinct in some way, such that a group of people have a reason for being together.
https://www.mayerowitz.io/blog/mario-meets-pareto
Loved this! Cute explainer of the Pareto curve through Mariokart choices.
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.