I answered the question in my other reply, but to reiterate, I did observe surface marring with a 3/32″ hole. I didn’t notice any specific difference with a 1/8″ hole, but I didn’t test it rigorously enough to see if the hold the screw had on the board was weaker.
quanticle
Karma: 2,876
Later on, out of curiosity, I did try a 3/32″ pilot hole, on a separate scrap piece. I did see the phenomenon that ChatGPT and Claude were warning about, where the surface of the material mushroomed up and ruined the surface finish. A 1/8″ pilot hole seemed to work fine, but I didn’t conduct a rigorous stress test to determine whether it was significantly weaker than the 7/64″ hole. It’s entirely possible that I could have used a 1/8″ bit and it would have worked fine.
Yes, my usage of 6 LLMs for this was gratuitous. It was less about getting practical advice and more about seeing how different LLMs would handle a question for which there wasn’t a single “correct” answer based on internet sources, and how they’d respond to feedback from the user about other approaches. I learned that ChatGPT and Gemini tend to fixate more and push back against the user, while Claude is less committed to its answer. ChatGPT, Gemini and Claude all seemed to have a similar level of detail while the quality of the answers provided by Meta AI, Deepseek and Kimi was a step below.
I reference Said Achmiz’s post specifically, because his post has an elaboration on Alicorn’s approach, where he considers what the various approaches he’s researched reveal about the structure of the underlying system and combines that with his background knowledge to improve the improvisation step.
In my case, I used my (admittedly limited) background knowledge of the behavior of wood to try to determine what the most important considerations were when choosing the screw type. In this case, it was the diameter of the solid root, and whether undersizing the hole would place undue strain on the board.
I find it confusing that the linked article claims that “output randomness does not belong in a strategy game”. Strategy games, originally, were intended to mimic warfare, where there is a significant amount of “output randomness” due to what Clausewitz called “friction”. The purpose of injecting randomness via dice rolls is to simulate that friction. Sometimes the weaker unit makes a heroic last stand and drives off a force many times more powerful than it.
In fact, one of my critiques of most “strategy” games is that they don’t have enough randomness. Where are the units getting lost? Or orders being misunderstood or deliberately disobeyed?
You have an entire section titled “Human values as weak evidence”, which discusses how humans diverge from their evolutionary goals, but then you don’t address the obvious counterargument that an AI is not going to be a product of evolution, it will be the (indirect) product of a deliberate design process.
Why should a deliberately designed system work like one that has evolved?
You seem to be making the claim that any sufficiently intelligent system will reject “semantically thin” goals, like maximizing paperclips. However, the argument you put forth in support of that claim appears to be that humans are sufficiently intelligent systems and humans reject semantically thin goals, and therefore the orthogonality thesis is incorrect.
But why should we expect an AI to think like a human? Our aeroplanes do not fly like birds do. Our submarines do not swim like fish do. Why should we expect an AI to think like a human does?
You can love your life, society, its norms and the freedoms they afford you. But the claims about the intrinsic goodness of your system, without a shared basis of evaluation between that to which you aim to compare it (under its own criteria), make them epistemic-ally thin as a viking screaming of his love for Valhalla. And while I think it is that vikings right to live and die for Valhalla, if that way is threatened, that love does not bubble up to something as equivalent to an excuse for external imposition of the viking way.
Ironically, this does more to justify colonialism than it does to defeat it. By your logic, if a modern military force were somehow transported back to the 1600s, they would have no moral basis to prevent European settlers from colonizing North America. For these settlers, the act of settlement was as much driven by morality, by a religious need to go forth and multiply and spread the light of God, as it was by rational calculation of material needs.
So I ask you, by what right would you stop these settlers from going forth and bringing the light of the Lord to the heathen darkness?
Capital holders would still have income, and if machines replaced labor on a vast scale, profits, dividends, and asset values could rise sharply, leaving owners with even more spending power.
The marginal value of more stuff declines pretty quickly. Someone like Elon Musk or Mark Zuckerberg can only enjoy so many houses, so many cars, so many new clothes, so many vacations, etc., because there are only 24 hours in a day. The empirical evidence we see from the behavior of actual billionaires is that they make money practically automatically, because the interest on their existing assets accumulates faster (often hundreds of times faster) than they can actually spend it. The only way these fortunes actually decline is if they commit to massive (sometimes wasteful) investments or pledge to give their wealth away. Even then, as was the case with Bill Gates and Warren Buffett for a while, their wealth may still accumulate faster than it can be dissipated.
The judge has dismissed Ann Altman’s case against Sam Altman, on narrow procedural grounds. The statute that Ann Altman was filing her case under had a statute of limitations that expired in 2008. Judge Zachary Bluestone has said that Ann can re-file her case using a different law, the Missouri Childhood Sexual Abuse law, which carries a looser statute of limitations.
The judge is also allowing a defamation counterclaim filed by Sam Altman to proceed.
That’s unfairly dismissive. I can’t speak to retail, but manufacturing absolutely does require “deep work”. Machining requires concentration and technique in order to ensure parts have the right tolerances, surface finish, etc. Assembly work often involves deep thinking in order to ensure that the machine is correctly assembled and properly configured.
It’s not all routine “mind numbing” assembly line work, just as not all IT is routine mind numbing data entry.
No, we don’t wait hours or days, we text you asking, “Hey, when would be a good time to follow up?”
Working remotely is the rub, even for software developers (archive):
We appear to be reaching remote-work equilibrium, after years of conflicting trends and predictions. The share of open jobs listed as remote on the career site Indeed has held steady between 8% and 8.6% for the past six months.
That’s roughly triple what the rate was in 2019 but markedly less than it was in 2022. Back then, more than 10% of jobs were advertised as remote, and many others were understood to be for obvious reasons.
[...]
Forty percent of applications submitted through LinkedIn are for remote roles, even though those jobs represent only 8% to 9% of listings.
JobHire.AI, which makes an artificial-intelligence tool for job hunters, recently analyzed more than 600,000 users’ applications. It found remote roles are roughly four times harder to get than in-office or hybrid positions.
These are fictional examples. In reality, like at the Battle of Cannae, for example, the Carthaginian force, led by Hannibal Barca, was outnumbered by the Romans by nearly 2:1 (roughly 40,000 Carthaginians against around 80,000 Romans). And yet the Carthaginians won by trapping the Romans in a double-envelopment and attacking them from all sides, preventing the Romans from concentrating their superior numbers, allowing Hannibal’s forces to slaughter the Romans. Although Hannibal’s forces may have outnumbered the Romans at the actual line of contact, no one claims that the Carthaginians had numerical superiority at Cannae.
Similarly, at Agincourt, the English were outnumbered by the French by approximately 2:1. The French were defeated because the English used terrain cleverly, forcing the French forces to attack down a narrow path surrounded by forest on both sides. This, combined with the superior firepower of English longbows, enabled the English to decimate the French, and win a major victory in the Hundred Years War. Once again, although the funnel effect of the terrain may have meant that the English enjoyed numerical superiority at the point of contact, Agincourt is widely regarded as an example of an underdog victory, where a smaller English army won a major victory against the odds against a much larger French foe.
At the battle of Narva, Charles XII of Sweden used a snowstorm to hide the approach of his 10,000 troops, suprising the 35,000 Russians laying siege to Narva and putting them to rout. Do you claim that the Swedes had numerical superiority over the Russians merely because some of the Russian troops could not see the Swedes until it was too late?
At the Battle of Trafalgar, Nelson had 27 British ships, against Villenueve’s combined Franco-Spanish fleet of 33. However, Nelson was able to maneuver his fleet such that it split Villenueve’s line into thirds, allowing his fleet to defeat Villenueve piecemeal. Although Nelson had numerical superiority at specific, crucial points in the battle, overall Villenueve’s fleet was larger, and Trafalgar is seen as a victory of British elan and seamanship, not numerical superiority.
The reason I bring up all these battles is because numerical superiority is but one part of victory, and, arguably, not even the most important part. Logistics, terrain, morale, and tactics all have parts to play as well. Redefining numerical superiority as you’ve done elides those factors, as you can always say, “Well, actually, at the point of contact, the victorious forces had numerical superiority over the vanquished.” That’s true, but it’s in some ways as much of a tautology as, “The team that wins the game is the one that scores the most points.” It doesn’t inform any predictions about the future, nor does it indicate when you should be surprised by an outcome.
This concept stems back to Adam Smith, a man who lived with his mom all his life, never held a job other than in Academia, and lived in an England sustained by colonial commodities’ extraction. It makes sense, then, that to him, Economy looked like a Divine Mystery. He didn’t know how enterprises were planned, where goods came from, how they were processed, who’d buy them and why. That is to say, he wasn’t aware of the political planning behind raw goods’ acquisition, the orchestration and science behind manufacturing, the original realities and incentives of the places that made English surplus possible, nor the realities and incentives of the very concrete people who everyday made decisions at the bazar. Removed from all this, the market was indeed “a God” to him, and in the same situation, you’d probably have a similar perspective.
Have you actually read The Wealth of Nations? Smith was well aware of how the enterprises of his day were planned, where goods came from, how they were processed, and how markets worked. His argument proceeds from that intimate knowledge. Smith argued, that, counterintuitively to most people of his era, markets should be allowed to work, and that countries shouldn’t engage in mercantilist policies to hoard “specie” (gold and silver). He argued that the true wealth of nations came from trade and industry, not shiny metals, and that the way to increase the wealth of a nation was to encourage the development of markets and commerce rather than mercantilist policies that accumulated shiny metal coins at the expense of trade. While Smith may have seen divine providence in the working of the market, that doesn’t imply that he was unaware of the practicalities of trade, no more than Isaac Newton’s belief in biblical numerology invalidates his ideas regarding the analysis of infinetesimals.
Furthermore, Smith’s belief in the power of free markets and free exchange led him to be a staunch anti-colonialist. Smith was one of the first British writers to speak out against the depradations of the East India Company, arguing that the company was making money because it had the backing of the British monarchy, and therefore was able to engage in state-like looting via its private armies and navies, rather than making wealth from free exchange. He correctly identified that the East India Company would require ever increasing amount of subsidy from the state, which would have deleterious results for both the company and Great Britain.
You’re free to use your own idiosyncratic definition of “numerical superiority”, but then you shouldn’t be surprised when your interlocutors are persistently confused as to how you can claim an army of 1000 has “numerical superiority” against an army ten times its size.
A thousand Frenchmen may have achieved local overmatch against the forces of the regime. But at that time, the Revolutionary forces most assuredly did not outnumber the entirety of the French military.
Who has the numerical advantage?
Alice, of course. Bob’s army wins via superior cohesion and morale. By your logic there can be no victory against numerical odds, because you can always look closely and find specific points where the victorious side had numerical overmatch against the losing side.
The basic strategy of a revolution leverages an overwhelming numerical advantage to overcome the coercion produced by an existing power structure.
No it does not. Can you name one successful revolution where the revolutionaries had an “overwhelming numerical advantage”? In the American, French and Bolshevik revolutions the revolutionaries, at best, had parity with the forces opposing them. They succeeded not because they had overwhelming numerical advantage, but because they had advantages in coordination and cohesion that enabled them to strike while their opponents were still preparing.
With all of these anecdotes, you should ask yourself, what did people do before they had cell phones? Cell phones aren’t that old! It’s not like the 1990s and early 2000s were some pre-civilizational state where people just couldn’t coordinate or make decisions for lack of instant communication across vast distances.
That’s just what Eliezer is pointing at with “Choosing to be Biased”. In practice, it’s impossible to determine causality between beliefs and “success”. There are lots of successful people with all kinds of utterly insane beliefs. Were their insane beliefs the cause of their success? Or were these people so talented that they were able to achieve success despite those beliefs?
In contrast, determining whether a belief results in good predictions about the world is quite easy, and gets you many of the benefits without any of the potentially catastrophic costs.
I think the general concept provided by Viliam is true, even if some of their specific examples aren’t perfect. Traffic laws are a good example. In the US (and Europe), most people follow traffic laws. They stay in their lane, they stop for red lights, they yield to pedestrians, and generally maintain an orderly movement of traffic. Do they, individually “leave money on the table” by doing this? Absolutely. It’s aggravating to be late and be held up by hitting an unnecessary sequence of red lights.
However, if everyone started ignoring traffic laws, the outcome would not be a better equilibrium. The outcome would be Indian roads, which are both less safe and less efficient than roads in the West, because hardly anyone actually follows traffic laws and the roads devolve into a Hobbesian “war of all against all”.
I don’t think “often going ‘above and beyond’ is just a waste”, I think it’s the price we pay for living in a civilized society. When someone brings free donuts to the office, we don’t swipe the entire box, we take one, and leave some for everyone else. When stores leave things on open shelves, we don’t simply take products and walk out the door with them. When we see a pedestrian waiting to cross the road, we stop and let them cross. When we see a line we join at the end rather than attempting to barge our way to the front. When we take a shopping cart from the store, we return it to the cart corral, rather than leaving them out in the middle of the parking lot.
I have lived in places where people did not follow those rules. Those places were markedly worse places to live than places where people behaved in an ethical, civilized manner. I push back strongly against anyone who claims that these rules are mostly just waste.