Posts
Comments
And for me, the (correct) reframing of RL as the cherry on top of our existing self-supervised stack was the straw that broke my hopeful back.
And o3 is more straws to my broken back.
Could it be worth it to buy 23andme stocks if you want some of their user's data?
Naively, the sticker price of 75M USD (today's market cap) for all of their user data might seem cheap - all together for genomes of roughly 12 million users. It seems reasonably cheap to me on a replacement cost and opportunity-cost basis.
However, the 49% basically-majority shareholder is CEO Anne Wojcicki and a "possibility is that Wojcicki has unreasonable plans to take the company private at a bargain-basement price[1]". If you takes this path forward as a decidedly important but minority shareholder, the only clear way forward is to hope for a shareholder oppression lawsuit in your favour. And instead of cashing out, you hope you can turn your skill set of "being annoying enough" into data in-lieu of cash (Note: if this is done through bankruptcy process, I can't see the trustee agreeing to this plan.)
Even given the obstacles, is this a deal? Or is this just a bad price for this particular bag of information? Is 23andme's information sufficiently low fidelity that it might not be useful at all except to relate yourself to fractional vikings?
- ^
Per Matt Levine - Money Stuff, Sep 19, 2024
A super silly heuristic I often use is "What media do you consume?". Intuitively, it kinda make sense as a sort of an informational parallel to the old adage "You are what you eat." Look at their spotify/RSS/blogging/shortform-video consumption habit tends to inform me whether A or B would at least have a decent first conversation. But this seems much better at matching friends than partners - presumably common interest and shared consumption of information is a slightly less important factor in continued romantic life (because there are so many other things!).
Is it Moskovitz's "irrational" responses that got him or a set of rather legible needs like "avoiding funding anything that might have unacceptable reputational costs for Dustin Moskovitz"?
we irrationally expect everyone else to be rational..
I am not sure that we do? I think we are not immune to the typical mind fallacy. But there is plenty of talk around these parts about optimal strategies when confronted with irrational opponents - the correct decision is not always throwing away your own rationality. Communicating with emotional people with the languages they can resonate with seems like a fine practice of rationality.
Now, that is strawmanning a little bit. Perhaps this is talking about a maximum exploitative strategy against irrational/imperfect agents?
Does this some right to you as a toy model?:
You are a chess grandmaster. One day, you are challenged by a youth of hubris - the youth is clearly going to benefit from been taking down a peg or two for his development.
If you play as if you are facing a near-peer opponent, you will likely win but that is not optimizing enough for humiliation (and you knew humiliation is good for his devlopment). Been the chess grandmaster you are and the fact that you had been a callous youth in your haydays, you know you can easily model your opponent's next moves; they will likely make more inaccuracies than you and you are loathed to not take advantage of the situation. So you do, and you make somewhat inaccurate moves in anticipation of them playing badly and thus, maximizing the humiliation potential.
Woud the chess example be a case where a rational agent is rationally expecting the opponents to be irrational?
Thanks! I have built these before I ran into cleanairkits and the school of thought that "lower efficiency with higher throughput is better" - I think per dollar, their CADR is likely quite a bit better! Looking at their Exhalaron line up - we have what is essentially two of my style of filter glued and tension-ed down in a portable package.
And a similar HEPA filter is used here as well. With two fans that are each nominally figure of 75 CFM. 92cfm / 2 / 75cfm = 61% - instead of the 80% figure I handwaved! (the new numbers should be up soon that is based on a more economical P14 setup instead of the P12.)
I think Thomas's "Instead of using HEPA to 'one-shot' (original design intention) the air filtration task, the 'few-shot' approach with much higher through put with a MERV 13ish lvl of efficency is generally better" is mostly correct. I see that Dynomight's IKEA filter investigations have also made a similar conclusion (although it is more in the case of HEPA vs MORE HEPA).
However, I didn't want to 3D print/jerryrig an enclosure to fit in the recommended filters, and where I am, I couldn't source a nice self-supporting (non-HEPA) filter that I can easily plop a fan onto. But if I had some more spare time, the cleanairkits folks that Zac mentioned built basically a "quieter rectangular box with fans on the small sides, and non-HEPA filters on the big sides" - and that's what I want eventually.
For those interested in Chinese philosophy, I'd suggest 韩非子 (Han Feizi), which offers a thoughtful meta-analysis of earlier philosophers like Laozi and contemporaries such as Xunzi, in so far as their thoughts applied to statecraft. (The first Emperor was a big fan of the work.). Note: Avoid the Burton Watson translation.
This recommendation assumes some basic knowledge of Chinese history.
For those new to the subject, [Recommendation to come, I am trying to find the English version for a children's book to Chinese Philosophy and History] might be a better starting point. -> No translation found. Will update if I run into anything better.
My recommendations are based on several key considerations:
- Quality of available translations
- Relevance to Western readers
- Focus on state-building as both an entry point and a way to understand how philosophy shaped Chinese governance
- Historical impact on actual governance
Chinese statebuilding is a very relevant lens because how early the Chinese started to concentrate executive power. The legalist school makes sense because the state was able to codify law staffed with court officers instead of relying on customary law enforced by local notables (thou the latter still happened a lot). And for Confucianism, for almost all of its existence, saw the best way to enact their worldly vision was through influencing the Emperor/King and their imperial/monarchial apparatus; Chinese philosophy is very much Chinese Political Philosophy. (Besides things like Xuanxue, ofc.)
While Mozi and Xunzi are often recommended here, I've found they might not be the most accessible entry points. Mozi, while interesting, had relatively limited historical influence compared to other schools of thought - being more well known to lay Chinese audience by his portrayals in historical dramas than his works. It would be somewhat like introducing Western philosophy through a lesser-known PreSocratic philosopher – potentially interesting but perhaps not the most representative starting point.
I've found that Chinese philosophy becomes more accessible when viewed through the lens of practical governance, as this provides concrete examples of how these ideas were implemented. Governance, being a universal concept, offers familiar ground for readers from any background and which ever cardinal direction. Additionally, many philosophical classics about governance have been reinterpreted over time, making them more approachable for modern readers.
TLDR: Chinese philosophy is firmly attached to Chinese history just like Western philosophy to the history of the Church. Gonna know both and let them bounce off each other.
I did read the original. It was long and I skimmed it. It was better in the coherence-sense that the OOP didn’t post a probability on whether it is true or not. Hell, the OOP hedged it by saying “ Do I believe what I’m saying? Well, yes and no”.
I guess the core of my confusion is the radical mismatch in confidence projection in its explicit form and implicit form (through tone and context setting). [Note: the updated wording definitely tempers the expectations in the right direction, thou still a bit bonkers at first glance.]
50% is extremely high. And lighthearted tones are often used to convey a sense of “I know this is farfetched theory. But I hold this strong claim very/appropriately weakly”.
Though not meant as derision, it is absolutely wild to read “Though I don't know that much about orcas” and “50% that orcas could do superhuman scientific problem solving” in the same paragraph.
My uneasiness with this post is that I am not sure how serious/joking the post is. It has some of the hallmark of a relatively lighthearted post written in a serious way. (The interaction with the IP, for example) And tones of conversation is light at parts. Yet the call to actions are confusing - it is not really motivating and seems to offload responsibility too eagerly for someone that actually believes what they are writing about.
I am very confused about the post and not sure what to think about it.
Stephen puts it elegantly. Though for me who is more of a code monkey, I'd like to think of it as "Runtime Non-Zero cost type safety through some const generics".
I can see how the article can be convincing. But it is worth it to keep in mind that Hunterbrook is also a hedge fund that trades on their own news - an obvious case of potential alignment failure if there ever was one. Though I am not sure if they are shorting this one.
Perhaps more damningly:
Jiangsu Pacific Quartz Co., Ltd. (SHA: 603688) produces HPQ in China. Earlier this year, state legislators evaluated North Carolina House Bill 385, which could ban ownership of local quartz mines by foreign entities from countries designated as adversarial to the U.S., such as China.
Per the Hunterbrook article.
PS: It is likely critical, but I am more uncertain about it being a single point. Unless we are limiting ourselves to the allegorical West.
A quick sanity check on the Chinese side of the web had revealed a couple of manufacturers for semi-conductor grade quartz, allegedly with manufacturing and processing centres in Jiangsu, CN.
My prior on this product type actually being a critical single point of failure is low.
See below: http://www.quartzpacific.com/api/upload/uploadService/dowloadEx?fileId=1113&tenantId=147391 ^Product spec (one of many semiconductor grade product shape) http://zj.people.com.cn/BIG5/n2/2023/0316/c186327-40338436.html ^investment news on new sites and manufacturing capacity
It doesn’t seem like you are arguing that breastfeeding is universally more convenient than formula. But breast feeding can be very inconvenient:
- It is often painful
- Elevated chance of inflammation
- Public spaces are not setup for mothers to breast feed; some may not value it, but a lot of people value privacy.
Formula’s convenience lays in enabling asynchronous feeding of the baby - by separating the role of the producer and the role of the feeder, the other partner can take care of the baby whilst the mother sleeps.
Another compromise to make is store breast milk and reheating it on demand!
On Lesswrong being a dispersed internet community:
If the ACX survey is informative here, discussing local policy works surprisingly well here! I’d say a significant chunk of people are in the Bay Area at large and Boston/NYC/DC area - it should be enough of a cluster to support discussions of local policy. And policies in California/DC has an oversized effect on things we care about as well.
I am curious, what were other "visions" of this workshop that you generated in the pre-planning stage?
And now that you have done the workshop, which part of the previous visions might you incorporate into later workshops?
I hope the partial unveiling of a your user_id hash will not doom us all, somehow.
I am not everyone else, but the reason I downvoted on the second axis is because:
- I still don't really understand the avoidant/non-avoidant taxonomy. I am confused when avoidant is both "introverted... and prefer to be alone" while "avoidants... being disturbing to others" when Scott never intended to disturb Metz's life? And Scott doesn't owe anyone anything - avoidant or not. And the claim about Scott being low conscientious? Gwern being low conscientious? If it "varying from person to person" so much, is it even descriptive?
- Making a claim of Gwern being avoidant, and Gwern said that Gwern is not. It might be the case that Gwern is lying. But that seems far stretched and not yet substantiated. But it seemed confusing enough that Gwern also couldn't tell how wide the concept applies.
There is some good stuff here! And i think it is accurate that some of these are controversial. But it also seems like a strange mix of good and “reverse-stupidity is not necessarily intelligence” ideas.
Directionally good but odd framing: It seems like great advice to offer to people that going straight for the goal (“software programming”) is a good way to approach a seemingly difficult problem. But one does not necessarily need to be mentored - this is only one of many ways. In fact, many programmers started and expanded their curiosity from typing something like ‘man systemctl’ into their shell.
It seems like, instead of asking the objective lvl question, asking a probing “What can you tell me about the drive to the conference?” And expanding from there might get you closer to desired result.
Witty, but I feel like that is not actually true?
It is likely that the rationality oft named is not the true name of the thing. Or “just be a perfect bayesian agent lol” is not practical. But that does actually mean anything legible is immediately false?
TLDR: This is an long metaphor to draw parallels between the Hanseatic League and the broader EA/LW communities. It is OK to not be a {corporation, societas, collegium, universitus} with common/top-down/bottom-up violence-monopolizing system. The price to implement a resolution system people would find satisfactory might be too high.
The fact it is hard to resolve conflict is because it is an integral part of the bargain, not an isolated bug. I personally don't want us to become a chimera with nine heads - a chimera for the sole purpose so we can utter that "we are a state".
“the [Efficax Altruismus] is not a societas: (a company) for it knows neither a common ownership of goods nor shared ownership of the good, since in the [Efficax Altruismus] there is no joint ownership; nor is it a company formed for certain commercial transactions, since in the [Efficax Altruismus] each individual makes transactions for himself, and the profit and loss falls on each individual…
It is also not a collegium (a college)….since it is formed from separate [communities]. It is also not a universitas (a corporate body), because…for it is required that it has property, a common treasure, a common seal, a common syndicus and a recognised leader.
“the [Efficax Altruismus] is … a firm alliance of many cities, towns and communities for the purpose of ensuring that enterprises on water and on land have the desired and favourable success and that effective protection is provided against pirates and highwaymen, so that the merchants are not deprived of their goods and valuables by their raids.” - The Hanse in Medieval and Early Modern Europe, I think...
Oh my, I hope your sanity is holding.
In a sort of morbid way, seems like things are working as intended - the "sharp" fella is winning social battles (invented or not) and keep exploiting the ever widening strategy space. Emboldened, he quickly gets to the "this is the line and no further" boundary of his current strategy. But instead of modifying it and keep his old strategy as a tool in his arsenal, he over-exploit it and disrupts the equilibrium so much he gets kicked out.
It is very possible that it works - though I am somewhat doubtful and I don’t have a unit to test it.
A quick way for us to learn more would be to I guess duct tape the screen to the laptop at the angle/height your want - and work with it for a bit. Might be able to get more experimental data than our theory crafting.
Indeed! But these are side loads instead of directly above the hinges.
Imagine this... you are a hinge. You are designed to take loads that roughly matches the motion of opening and closing of the lid + a bit of additional tolerance. But when someone mounts something heavy on the side of the laptop, you are mostly annoyed but OK with it because the side load will try to rip out the hinges out of their respective housing in different directions - the housing is usually plastic on cheaper computers, but perhaps aluminum on macs?
The problem with have that much weight on the top of your laptop would be... that your hinges would want to close when you don't want them to - they not designed to be stable while holding onto that much weight + the leverage given to the new screen from being far away from the hinge assembly.
It might work, but it seems like the main difficulty would be the laptop hinge.
The hinge would be taking way more force than it is intended to take. And from my experience, it is somewhere around 3kg for macbook air 13 (your model is likely different) - so a M156 mounted 35cm above the hinge might produce quite a bit of stress.
Just going off a hunch, mostly the asymmetry of risk and award?:
Award: Spread the gospel of probabilistic truth, personal intellectual growth potentially
Risk: retaliation (especially if the author is in the field), harassment, threats, law suits, wait… just more kinda of retaliation really. And potentially been seen as some one against the field of anti-aging despite an attempt at doing good science.
Hiya Yovel! Q1: How have you been impacted by the recent hostilities? Q2: What do you think are the potential end goals of this newly re-escalated conflict for the Israeli government? (As an naive observer, seems like [occupying Gaza / leave a power vacuum / let the Hamas reorg after] are all rather bad outcomes)
Always welcome more optionality in the opportunity space!
Suggestion: Potential Improvement in Narrative Signalling by lowering the range of RAs to hire (thus increasing pay):
- If I were applying to this, I'd feel confused and slightly underappreciated if I had the right set of ML/Software Engineering skills but to be barely paid subsistence level for my full-time work (in NY).
- It seems like the funding amount is well corresponded to how much the grant is. I am rather naive when it comes to how much ML/engineering talent should be paid in pursuit of alignment goals. But it seems like $70k spread across 4 people at full-time (for half year each) is only slightly above minimal wage in many places.
- Comparisons: At 35k a year, it seems it might be considerably lower than industry equivalent even when compared to other programs
- Ex: Lightcone has a general policy of paying ~70% market pay for equivalent talent.
- Recalling from memory of LTFF/similar grants that experienced researchers were granted 70k ~ 200k for their individual research.
- A quick glance at 80k job-board for RAs nets me a range of 32,332 ~ 87,000.
- Ex: Lightcone has a general policy of paying ~70% market pay for equivalent talent.
- Of course... money is tight: The grant constraint is well acknowledged here. But potentially the number of RAs expected to hire can be further down adjusted as while potentially increasing the submission rate of the candidates that truly fits the requirement of the research program.
- Hiring more might not work as intended: Also, it might come as a surprise that fewer people to manage will turn out to be a blessing rather than a curse - hiring one's way out of something is tempting but should usually be tempered with caution.
- Thing I might have got wrong:
- The intended audiences and people of the right skill-set will not be in countries where the salary is barely above subsistence level.
- The Research Project has decided that people who possesses an instinct for "I'd like to work here, but please give me X because I think I am worth that much and can offer at least that much value" is generally a poor fit for the project.
- A misunderstanding the salary distribution and the rate of deterioration of the current financial climate within the scene.
Overall, I am glad y'all exist! Good luck :)
Viktor has a point here - the title is informative, but not well optimized (perhaps intentionally) for attracting eyeballs.
Something akin to:
Military and AI Compute: DoD's 100 million cheque and what did it get for them?
Might do the trick a bit better.
*Not actual advice
Blow the matters up in an election season, concentrate media focus with minimal cost. Contact local political activists and famous NIMBYs, mass pamphlet style mobilization. Silent protest (of even just one) outside the Berkeley Department of Transportation.
- Carwash Forum for repair advice, purchases and general industry gossips about carwashes.
- RedFlagDeals for Canadian, especially Ontario, localized advices. Think advices that an average family would ask their friends for. Ex: "How much is it to hire a plumber to install the new water meter?"
- OpenCorporate if you are doing the lowest effort of corporate due-diligence and KYC *NOT LEGAL ADVICE.
Agreed, there can be a optimum. But I think the intuition here is that it is exceedingly rare enough to run into a situation where it is local optima in all "directions".
It is only an "optimum" when all 175 billion parameters are telling you to screw off and stop trying.
My instinct is that, the lotteries odds were not truly random or close to truly random. Or the odds for the specific lotteries were a lot better than assumed.
Or in other words, the prior for the lotteries being fair is low.
Quick note on “Ukraine…has not trained its people in guerrilla warfare.” I am sure that Ukraine has not engaged in public programs to turn a significant percentage of its population into capable guerrilla fighters.
However, from my sources in the NATO deployments, the Ukrainian irregulars and volunteers have been rigorously trained in “…irregular warfare” in significant numbers - to quote my sources. Will provide more rigorous and structure info shortly.
Today, there is a distinct lack of faith.
This lack of faith has permeated all pairs of 16oz denims, event merch hoodies and smartwool socks. The lack of faith is insidious, bad, and all things molochian.
Actually, screw this manifesto writing style. I am about to meta criticize myself for a style that has NOT aged well and tends to be way too authoritarian in tone but measly in effective information communicated.
Actually, screw this meta-criticism. In all good faith, I didn't have to write the criticism. I could have simply done the internal self-criticism and just deleted what I have written to start anew. Instead, I have now left it as a slightly well-disguised monument to my apparent self-improved and evolution. But really, just to stroke my ego and fill the word count.
Actually, have I written anything of content yet? Have I put my pen to paper and allow the ink to betray my trust when my back is turned? (also known as drying)
Oh no... Better start counting the trains that leave Datong for the trail of coal crumbs.
Real Content Soon™
[APPRENTICE] I'd like to take Isusr on what is being offered. Writing directly as well. Following in danohu's good idea in leaving a message to show that Isusr's proposal and the apprentice system is attractive.
It is likely this one: “吃得苦中苦，方为人上人”. Lit. Eat the most bitter of the bitters, become the person above the people.