@norvid_studies To put it another way:
- In HATETRIS, the branching factor is high. Almost all moves are horrible, one or two are great, very difficult to predict which.
- In Burgiel (alternating S/Z), the branching factor's low, most moves are mediocre-to-decent, and there are no great moves.
@davidbessis@aretae "Every theorem I proved increased my ability to prove the next one."
Yes, and for somebody less good, every problem he *doesn't* solve will make it harder to solve the next one. Small differences in innate capacity magnify in math, just like small differences in motivation etc.
@davidbessis@aretae Any activity involving multiple steps will be compounded across each step. You can get way more than a 10x difference if you look at the physical activities properly analogous to mathematics.
@davidbessis@aretae But math isn't a single step - it's a series of min-threshold steps. If you put me and and a runner and Usain Bolt on a track, and counted up how many times we could run a 11-second 100m dash, Usain Bolt could do it once every few minutes, the runner once an hour, and I never.
@HiawathaAvenged@captivedreamer7@acteduweininger I much preferred Ørberg over Wheelock; whenever I tried Wheelock, I fell into the 'recipe for disaster' failure mode: https://t.co/B9CeRywo36
I also typed out each chapter of Ørberg I read. @ScorpioMartian went a step further and *handwrote* each chapter. Effort, but worth it.
@gwern@losten8 @idiot_tanuki @Miles_Brundage@slatestarcodex I showed this to somebody familiar with Lem's work in Polish (Solaris, specifically), but unfamiliar with the Cyberiad or with Michael Kandell's English translations of Lem. When I removed the extra quotation marks, she was not able to tell the difference between Lem and GPT-3.
@qntm If you want to look at a cellular-automaton-like environment in which the behavior of an infinite amount of sparsely-populated space over an infinite amount of time is unknown, I recommend primordial particle systems.
https://t.co/Jp44cqT5Az
https://t.co/SS36dF6SId
@qntm I can't prove it mathematically, but information in CGOL can't travel faster than c/2 in vacuum, and disorder CAN travel faster than c/2 in occupied space. To resist perturbations, I think you'd need a way for the environment to spread order faster than disorder when perturbed.
@qntm The problem is that when spacefillers encounter any noise at all, the noise propagates. Try putting the Max pattern file (https://t.co/D9R6XC8v8W) into a simulator (https://t.co/CbKWZ0RBIS) with a single 2x2 still life somewhere in space and watch the agar dissolve into chaos.
@dpmattingly Gotcha; was wondering if things like current possession and field position played into it. Even with no refinements, the bot is pretty good.
@MyNameIsSome0ne@jon_bois Yes, but J. Howell's site doesn't list either as Division I-A until 1880. Also, my scraper just assumed scores to be integers. Silly me.