Student at Caltech. Trying to level up.
Submit anonymous feedback here (try to be nice): https://www.admonymous.co/tkwa
Thomas Kwa
Karma: 286
I’m thinking of a situation where there are subspecies A through (say) H; A can interbreed with B, B with C, etc., and H with A, but no non-adjacent subspecies can produce fertile offspring.
Are there ring species where the first and last populations actually can interbreed? What evolutionary process could feasibly create one?
Videos have irremovable noise, but in some domains there is none and compression is more useful. In my experience, one example where the prediction-compression duality made problems much easier for humans to understand is in code golf (writing the shortest possible programs that perform various tasks). There’s a Stack Exchange site dedicated to it, and over the years people have created better and better code-golf languages. Solution sizes shrunk by a factor of >3. Now they’re so good that code golf isn’t as fun anymore because the tasks are too easy; anyone fluent in a modern golfing language can decompose the average code golf problem into ten or twenty atomic functions.
I’m on a very similar path; DM me if you want my thoughts.
That was it, thanks!
“throwing the baby out with the bathwater”?
I’ve been searching for a LW post for half an hour. I think it was written within the last few months. It’s about how to understand beliefs that stronger people have, without simply deferring to them. It was on the front page while I was reading the comments to this post of mine, which is how I found it. Anyone know which post I’m trying to find?
Could the mechanism be that your note-taking rate is limited to a certain number of WPM? One way I think about note-taking is that I need to force my brain to “compress” the information to encode it into my memory, which is easier if I’m forced to actually compress the information by taking notes of a limited size.
It isn’t perfect, and it probably won’t actually work as a Gömböc either, because the shape is reported to be very sensitive to dimensional tolerances. But I spent quite a bit of time and energy on this and thought somebody might use it as a starting point for improvements. Or better yet, that it would persuade the Gömböc discoverers that it is time to finally publish the details of the commercial Gömböc shape in the open literature.
Thingiverse doesn’t appear to have a working 3D model, nor does there appear to be one elsewhere. See this link: https://www.thingiverse.com/thing:10190
There’s a design patent on the particular Gomboc shape, but that doesn’t apply to mono-monostatic shapes in general, and it expires in a few years anyway. I’ve added a bounty on creating/finding a working 3d model of a mono-monostatic shape.
I just wrote up a quarterly life review and will edit in relevant information soon. For now, I can say that I spend much more of my time feeling busy (e.g. reading LW, writing quarterly life reviews, checking emails) but the vast majority of this time is not highly productive. I want a more regular sleep schedule (already take melatonin at 9:30pm daily), and a way to get exercise that is pleasant, I can do every day, and takes less than an hour (baseline is walking on hilly streets). I occasionally miss important administrative things, but my system is improving fast enough that I might not benefit from suggestions.
I’ve been looking into this and find it not worth my time, though I plan to try it anyway to gain familiarity with investment.
First I have to get a margin account. This is not too much trouble. Then I upgrade this to portfolio margin; TD Ameritrade says I need $125k, “full options trading approval, and three years of experience trading options”. Investing for myself is out; what about my parents? They have a passing interest in finance, so they can likely pass the test after I discuss it with them for a few hours.
Then I need to figure out how box spreads work. Being justifiably cautious I should first try it on paper, then with 20% before selling the full number of boxes. Finally I need my parents to set up three bank accounts in an online bank to maximize FDIC coverage, adding administrative work.
What do they gain from this? 3-year Treasury yield is 0.23%, but the Facebook thread you linked suggests I should be unlikely to write a box for less than 0.6%. Savings and CD rates are 1.15%, so the difference is 0.65%. I will not go remotely close to 10x margin on someone else’s life savings, and puts are expensive. If I keep 40% in the brokerage account, this means they actually gain .65% * .6 = .39%. Interest on savings is taxable as ordinary income, so until they retire, over half of this is eaten up by taxes (assuming the capital loss from the box spread is used to offset long-term capital gains). Considering that this will take a couple of weeks of free time plus intermittently checking in to prevent margin calls, and there’s still a risk of screwing up somehow, the after-tax benefit is currently less than what either I or my parents value our time at.
I haven’t heard this. What’s the strongest criticism?
Say I need to publish an anonymous essay. If it’s long enough, people could plausibly deduce my authorship based on the writing style; this is called stylometry. The only stylometry-defeating tool I can find is Anonymouth; it hasn’t been updated in 7 years and it’s unclear if it can defeat modern AI. Is there something better?
$5 for any productivity intervention (mental technique, habit, reasonably priced physical item, etc.) that I think is worth trying. An additional $50*X if it leads to an estimated X% increase in my productivity in the year after trying it.
My current bottleneck areas are sleep, focus/energy, consistency, and self-image in that order, with food and infohazards as additional issues. I have a comprehensive writeup available upon request.
I’ve already read the LW threads on physical objects and one-shot life improvements, as well as a few others. A list of the physical items I use is here.
$800 for a gömböc made of a tungsten alloy with a density greater than 16 g/cm^3 that weighs at least 1000 g. As far as I know, this would be the first such item in the world; if I become aware of others this will decrease. Knowledge of how to create one gets partial award if I decide to create one, depending on how useful it is.
$50 for a 3D model of non-gömböc mono-monostatic shape which has been verified to work. There’s a design patent on the gömböc; the European patent may be overturned in court and the American one expires in 2024, but this would be nice anyway.
Edit: fixed typo in weight
$20 for writing a Spacechem puzzle that remains unsolved for 14 days when posted publicly and publicized on the Spacechem forums. Add another $20 if you write a detailed post on LW or crossposted from elsewhere explaining how you designed the puzzle and how one would go about solving it. Partial award for impressive attempts.
Some possible approaches:
If you’re extremely good at space optimization, write a single-reactor puzzle on the edge of your ability. Cram everything into the reactor and hope no one else can.
Encode some kind of public-key cryptography in Spacechem. Might be impossible.
More inspiration here.
Thomas Kwa’s Bounty List
This seems suspect to me; it violates EMH and AFAICT the backtest was only carried out to 2017 (it should be possible to backtest to at least 1993, when VIX started being tracked).
Though, as I also commented, there are some general structures that make me sit up and take note; probably the strongest is “These people have ignored their own carefully gathered experimental evidence for decades in favor of stuff that sounds more intuitive.” (Robyn Dawes/psychoanalysis, Robin Hanson/medical spending, Gary Taubes/dietary science, Eric Falkenstein/risk-return [...])
Eric Falkenstein wrote in 2016 that he converted to Christianity and doesn’t believe in evolution via natural selection. This doesn’t automatically mean that he was a crank in 1994, but to me it’s moderate evidence, plus a warning that we can’t trust people who have good excuses for contrarianism on some subject to have a good opinion on any other subject.
Before October 2019, I didn’t floss; as of July 2020 I floss on about 98% of nights. Flossing tends to remove minute amounts of food or plaque in my teeth such that my mouth feels cleaner in the morning and my breath smells better, so I would do it independently of any dental health benefit.
I’m not flossing particularly well, either: I floss both sides of every tooth but with just enough care to barely avoid slamming the floss into my gums every time; I don’t unwind the floss to get a fresh piece on every tooth as is recommmended; I floss after brushing my teeth, not before; flossing takes me under a minute.