Tiled With Pentagons

(Epistemic status: Opinions, but justifiable ones. For RD, and to a lesser extent RK and DN/LT; with thanks to JM and PR.) ...then what would I make absolutely sure that the new blood read, played, or otherwise interacted with? And why? This list is not meant to be exhaustive, but I've tried pretty hard to cover a lot of ground very fast. You may assume that this is in addition to classics, like "A List of Lethalities", excerpts from Bostrom, and "Ten Levels of AI Alignment Difficulty". Accordingly, this is the things that I would personally add to that curriculum, or maybe bump some marginal things in favor of. It's aimed all over the spectrum of what "new AI safety researcher" means; some of them are for totally new people, some are for people who have a sense of what subfield they want to attack, and some could benefit literally everyone including established researchers. I've tried to pick things that are specifically underutilized and a re...

(Epistemic status: A prophecy of the shape of things to come, if one that seems well-attested. "The sun will rise around this time a year to the day from now" is a prophecy, too.) Taiwan is toast. There, I said it, roll credits. The Republic of China is one of a handful of strange carveouts in the world of Westphalian sovereignty and the (supposedly) rules-based international order, alongside such curiosities as the Vatican, Portugal's exclaves in Africa, Singapore, Lesotho, Cootch-Behar, Bir Tawil, Transnistria, and French Guiana. Unlike the rest of these, it has been a locus of geopolitical struggle since its effective founding. A rump state of the post-Qing Chinese state, it was founded in 1912, but only took on its current form after the vicious betrayals and grand atrocities of the Chinese Civil War, the brief pause to fight off the Japanese, and not-quite-final defeat of Kuomintang forces followed by flight to Taiwan, where it has remained ever since. Until the earl...

(Epistemic status: a riff on existing cogtech. May not land for you unless you're a huge puzzle game nerd like me. Mild spoilers for The Witness.) There's a practice you may have heard of or even use yourself, in order to break through blocks and make progress on hard problems that you face. It's called a Yoda Timer, and it goes like this: First, fix the problem in your mind. Understand it. Face it. Then, set a timer for 5 minutes. Start the timer. Then, simply... solve the problem. Completely. In 5 minutes. Don't try to do it! There is no try! Do, or do not! Usually you will fail to actually totally solve the problem, but what you will have done is make serious progress on it. Maybe you've made a solid plan for tackling it. Maybe you've solved a significant subproblem and are now unblocked. Maybe you've spent time figuring out what you need to even make a serious attempt; you were underprovisioned from the start, and those 5 minutes weren't wasted, but ...

(This will make much more sense if you read "11. Why the First "High Dimension" is Six or Maybe Five" first. As before, epistemic status: morally correct, in the mathematician’s sense; here to give flavor and intuition without too much rigor. Frankly, I'm going to move even faster and more cursorily than last time. With thanks to IL, and to everyone I've tried to shill 4D Golf to.) After talking with IL a bit more and doing some thinking for myself, I realized with creeping delight that the long heuristic argument that I gave for why we should consider the first qualitatively high dimension to be six or maybe five is not remotely the only one. In fact, a shocking number of strong heuristic arguments all converge on the same figure: very specifically, "six or maybe five". That is: five is marginal, and six is definite. To give a very quick recap of the heuristic argument from the original post, the idea is that we can operationalize what it means fo...

Because I feel like it, I'm putting up a list of every single project I have not just an idea that I'd like to do, but even have a reasonably well-worked-out plan for what I'd do, how I'd do it, and even a vague sense of how much it'd cost. Feel free to steal these ideas for your own; I'm a consequentialist about them, and I'd largely rather that they come into existence by someone else's hand than that they never arise. I've grouped some of them together when it seemed natural, and then given a sketch of what I want to do and why. Assorted microtonal arrangements: there's a few decently well-known songs that I've listened to and gone "yes, delightful, but what if they were weirder and better?" and I still haven't got around to arranging, let alone recording. Several songs in 31-edo, which I often prefer to call "modernized quarter-comma meantone". (Because that's what it is.) I think that something like the richne...

(Epistemic status: earnest thoughts about a piece of functional jewelry I wear daily; contains both practical utility and more symbolism than is strictly warranted; heavy towards the end.)   Every morning I put on my necklace. On it are four Favors, my class ring, a tritium safety light (ice blue), and a small metal pendant that tells me where the moon is. That's not meant metaphorically. At any given time of night, I can look down at it to turn a rough sense of the phase of the moon and its position in the sky into a pretty accurate compass heading. It's made by Wndsn, a company specializing in precision analog instruments - tools that work without batteries, without satellites, without any infrastructure at all past your eyes and brain. The pendant is a simple lookup table based on a trick you might already know: line up the points of the moon's crescent or draw a line tangent to its changing edge, and draw the great circle on the dome of the sky. Wherever it hits the gro...