All of Yair Halberstadt's Comments + Replies

Newtonian mechanics is a bunch of maths statements. It doesn't predict anything at all.

The students constructed a model of the world which used Newtonian mechanics for one part of the model. That models predictions fell flat on its head. They are right to reject the model.

But the model has many parts. If they're going to reject the model, they should reject all parts of the model, not just pick on Newtonian mechanics. There's no such thing as gravity, or pendulum, or geometry, or anything at all. They should start from scratch!

Except that's obviously wrong... (read more)

That's clearly not true in a general sense. Here's a pattern that points to a different sum:

1 + 2 + 3 + ... = 1 + (1 + 1) + (1 + 1 + 1) + ... = 1 + 1 + 1 + ... = - 1/2

Now the problem is this pattern leads to a contradiction because it can equally prove any number you want. So we don't choose to use it as a definition for an infinite sum.

So you need to do a bit more work here to define what you mean here.

In precisely the same sense that we can write 1 + 1/2 + 1/4 + ... = 2, despite that no real-world process of "addition" involving infinitely many terms may be performed in a finite number of steps, we can write 1 + 2 + 3 + 4 + 5 + ... = -1/12

I think this is overstating things (which is fair enough to make the point you're making).

The first is simply a shorthand for "the limit of this sum is 2", which is an extremely simple, general definition, which applies in almost all contexts, and matches up with what addition means in almost all contexts. It preser... (read more)

1Shankar Sivarajan13d
It doesn't need to be! It can be more generally something like "the unique value that matches patterns," where what counts is extended first beyond integers, and then to infinite series, and then to divergent infinite series.

Submission: Turing

MMAcevado simulates a Turing Machine in his mind, itself running a lossily compressed simulation of base MMAcevado. The simulated mind runs at 1/10,000th speed, and MMAcevado routes all IO through to the simulated mind.

On the assumption we have self navigating drones that can detect the weakest point in a tank as soon as it gets live of site, and head straight towards it, we would presumably have developed the ability to detect such drones via cameras on the tank as soon as they have line of site.

Than all you need is a bunch of pretty weak guns on turrets mounted on the tank to shoot the drone as soon as they are detected.

Most of these pieces already exist - modern Merkavas have cameras with 360 degrees view, the software to detect a moving drone quickly from a camera is... (read more)

-5Gerald Monroe23d

To a mother drone located farther from the enemy at higher altitude, but not high enough to be engaged. Using laser or directional (phased array) RF.

If it has line of sight to the drones, then it has line of sight to the target, and can be engaged by them.

3Gerald Monroe24d
The mother drone produces the tactical plan and gives the plan to every drone in the swarm. It doesn't need to have los to execute it although there would be a lack of flexibility during the attack. The tactical plan would consist of things like waypoints, references images for positioning, and targets. Each drone for that hypothetical "70 vs 3 tanks" engagement has a track to follow, a period of exploiting the terrain, a maximum speed final approach, a specific place on a specific tank to target etc. The plan can fail if say the target vehicles have moved and additional guns are brought in but the entire process will be a few minutes. Getting the plan would be done with disposable scouts, these are mid altitude drones with a good laser link to the mother. They upload images of the target area before the shots to shoot them down hit. Ultimately you are not going to beat this with conventional forces with more air defense. You need a rapidly relocatable force to counter drone swarms that is cheap. Meaning a defense in depth with your own drones.

I feel like the goalposts keep changing. This is not what was described in the original post.

So a few questions:

How do these drones communicate? Low on the ground P2P communications will have awful range, as will most low energy communication systems. Are they so autonomous they don't need to communicate at all?

What's their range? Existing drones only fly for about 20 minutes, and at a speed of about 70 km/h. Their range is usually about 10 to 20 km. Flying low to the ground and having to navigate will imply much lower speeds, and less efficient flight, as... (read more)

3Gerald Monroe24d
To a mother drone located farther from the enemy at higher altitude, but not high enough to be engaged.  Using laser or directional (phased array) RF.  This is how drones are fighting in Ukraine right now. See: https://www.businessinsider.com/russian-military-reservist-describes-flock-ukrainian-fpv-drones-2024-1 Most drones are short range, yes.  Some are hybrid with gas engines, like the mother drones or like the below, from https://news.mit.edu/2017/hybrid-drones-carry-heavier-payloads-greater-distances-0804 .  These will have hundreds of kilometers of range.    > Again, how are you actually destroying the tank? Firstly ADS systems are likely to be extremely effective against drones. Secondly tank armour is actually really really difficult to pierce. Standard package.  Switchblade 600 has the same warhead as the javelin missile.  It's listed at 33 lbs.   Tanks have a lot of weak spots, for instance from this video  you can just look for yourself at the very thin plate on the Abram's roof that appears to just be rolled homogeneous steel.  It would be pretty difficult to armor tanks to withstand attacks from all sides, for one thing someone will just build slightly larger drones. Currently the cost when built without the inefficiencies of USA contractor system is under $1000.  https://www.economist.com/interactive/science-and-technology/2024/02/05/cheap-racing-drones-offer-precision-warfare-at-scale  from the article : "A simple FPV drone costs perhaps $400" These have every element you mentioned except the GPUs for full autonomy.   And ok, it's consumer grade, let's go to milspec, and suppose it does cost 100k a drone*.  Well, how much does 3 tanks cost?  Each Abrams is listed at 24 million dollars (present day export cost).  So 70 drones will be 7 million dollars, killing 72 million in tanks and 9 crewmembers who needed time and money to train, and who will have lost their combat experience. Note the drones don't lose their combat experience when destro

When is talking about kinetic energy weapons it's referring to armour piercing sabots, because that's what needed to pierce tank defences. I don't know how effective ADS would be against other kinetic energy weapons because there's never been any need to try, they're useless against tanks. These rounds are so heavy, and fly so fast, that's it's practically impossible to fire them from anything weighing less than a few tons. Not relevant for a drone/Javelin. Also notice how you're creating epicycles upon epicycles here. A drone that fires a Javelin, that fi... (read more)

3Gerald Monroe24d
I think you're really really badly off base, and let me show you why: That's what they are capable of under human control.  The actual drone hardware is under $1000. None of your radar proposals are any use.  The drones will fly inches from the ground and are a lot of plastic, and their speeds are low, and will keep terrain between them and the enemy.  These are generally not going to be detectable on radar until it's too late.  Radar is not effective now. The biggest objection to why you won't see these drones immediately is you need powerful onboard GPUs to run a realtime AI control model to get human (or insect) level flight control, and to actually coordinate a drone assault you essentially need much more expensive "mother" drones that have combustion engines and racks of GPUs running a transformer or newer model. What needs to happen is the drones can implement strategies with respect to waypoints, and as the drone encounter possible enemy forces their models need to select and upload to the mesh network high resolution pictures of what they perceive, as well as lots of compressed metadata. The model aboard the "mother" drone - or human commanders - have to decide on a tactical solution to engage the enemy assets they can see/estimate exist.  Each of the objections you mention has a tactical solution, and in most cases it will be hugely asymmetric, killing armored vehicles and trained soldiers for a fraction of the cost.  It will never be 1 drone against a defense, but enough drones calculated to overwhelm the known defenses. For example, if the approach path chosen, hugging the terrain, exposes the drones to 3 phalanx guns, and there is a total of 5 seconds of exposure time, and the gun can change targets every 250 ms, then up to 60 drones can be engaged. If 70 drones are sent, is the cost of 70 drones worth trading for 3 armored vehicles?  It the answer is no, you don't engage and instead seek weaker targets like supply trucks.
1RussellThor24d
Yes - fair enough and I hope you are right - I would be happier if defense wins too. I hope soon  Europe/USA develops such a system, including the ability to mass produce it in the quantities needed.

Like everything you have layers of defence. Phalanx takes out all drones in an area. Against any ATGMs you use trophy or an equivalent ADS.

Also once you have a javelin/anti armour carrying drone it's going to set you back hundreds of thousands of dollars and be a suitable targets for iron dome style defences, which can cover a larger area and where each missile costs some 75000 dollars.

1RussellThor25d
Trophy sounds less effective than phalanx for missile defense in this situation: https://en.wikipedia.org/wiki/Trophy_(countermeasure) "The system is currently incapable of defeating kinetic energy anti-tank weapons." So a Javelin type missile that released a rail gun type slug when on the outside of the Trophy defense  range would destroy the target. "In the ATGM's case, the EFP will affect the shaped plasma jet, dramatically decreasing its penetration ability." - This sounds a normal missile will still cause damage, and since Phalanx is not as armored as a tank, probably destroy it.  I don't see why https://newatlas.com/drones/huntress-turbojet-drone/ should cost >$100K when mass produced. In the context of this article, before we talk about strengths/weaknesses of Iron dome type defenses, the vast majority of countries don't have them currently deployed and can't afford to.

Why compare with a Javelin, and not e.g. a Kornet which exports for 25,000 dollars (similar to a top range GPU), and can be produced for much cheaper (as evidenced by the fact Hamas is perfectly capable of producing them).

The chances of the vision you're espousing being viable (autonomous air based drones replacing the majority of all existing army units in the near future) is extremely low.

I think you're looking at a war between two seconds rate armies, which have not adapted to a new technology and are suffering a large rate of attrition due to it, and assuming based on that that such a system is unbeatable.

I'm not going to go into detail about all the issues with this article, I'm just going to focus on one specific issue.

If you manage to build this entire system, how w... (read more)

1RussellThor25d
For the phalanx or similar - check this link this link. The consensus seems to be that the gun can't take out multiple missiles. Wikipedia gives a Javelin (surface to air) with a max speed of "Mach 1.7+ approx." EDIT the ground one is much slower, probably because its has more penetrating power. I don't think we need that against the gun, it is not as armored as a tank. I don't doubt that a fixed gun can take out drones, but its missiles vs gun that matters here. At 115 degrees per sec, 2 Javelin 120% apart going Mach 1.7 is going to be a serious problem for it. I get a cost quoted of about US$5M for the land based gun vs $78K for the missile. https://www.thedefensepost.com/2023/03/02/us-uk-javelin-missile/ If it was made cheaper, would it be less effective? Also note that the land based system needs to withstand armor piercing rounds from the heaviest gun something like  this drone  could carry. Sure you could destroy the drone but probably not before it can get some rounds off. Especially if it engaged from >1500m out. That drone could also carry the Javelin and I doubt someone  (soldier) could shoot it down easily. See this comment thread about jamming. 
1Lalartu1mo
Anti tank FPV drones? Almost certainly not long term as they’re more expensive than ATGMs, That is not true at all, anti-tank fpv cost is about 1/100 of a Javelin missile. It is not obvious how much autonomous guidance would add to a drone cost, but probably less than 10000%.

Ok, that makes sense, targeted killings from a distance greater than a sniper rifle does seem like a good choice for a drone.

Just realised I misunderstood this section of the post, going to delete, rewrite and repost.

They coordinate with missiles to defeat countermeasure such as flares and chaff from slow moving aircraft (helicopters etc) by observing and transmitting the position of the target from somewhat further away and transmitting that info to their missile.

What is an example of a terrorist attack drones enable where with the same cost and effort terrorists couldn't do something similar already?

3Vaniver1mo
To elaborate, it's pretty easy to kill someone important if you are willing to be arrested/executed afterwards; the main thing a suicide drone might enable is killing someone important and being able to escape afterwards. This could already be done with dupes, like the 2017 killing of Kim Jong-nam, but I think the nerve agent involved was more expensive than a handmade gun.

Now I don't doubt we'll be seeing incremental changes here, and more uses of drones and autonomy, but I don't think this is going to rewrite the rules of war anytime soon.

9Daniel Kokotajlo1mo
Drones already rewrote the rules of warfare. The recent invasion of Ukraine would have gone very differently without drones. Primary effect they are having is on sensors/surveillance; every assault is seen building up miles before it reaches the front line, and artillery is targeted on it in real time. Big buff to artillery and big nerf to armored assaults. Secondary effect is kamikaze drones -- now the #1 weapon by monthly kills maybe, or maybe #2 after artillery. It says a lot that I'm not even sure which is doing more killing now, since artillery is far and away the obvious contender for a static conflict. Even ww2 which was mobile saw most deaths from artillery I think.  Special mention to naval and long-range drones (bayraktar, shahed, etc.) which are not exactly revolutionizing anything but which are having a noticeable effect. All in all I think yes the war has been revolutionized by drones. But the biggest changes are yet to come.
3Shankar Sivarajan1mo
I believe it will, but not in the way described here. What drones facilitate is terrorism, and maybe assassinations. If developed to fruition, they would "rewrite the rules" of irregular, asymmetric warfare.

I think this is wildly off base.

Cheap drones are far easier to destroy than pretty much anything else on the battlefield, and are highly susceptible to electronic measures. Their only advantage is they are cheap, and current tactics and equipment hasn't yet adapted to them. Once every vehicle contains a cheap jammer, and every unit carries them around, the cheapest drones will be far less useful (except for reconnaissance).

You suggest various countermeasures, but these end up taking us back to where we started. For example you suggested reconnaissance dron... (read more)

[This comment is no longer endorsed by its author]Reply
4Gerald Monroe1mo
So the OP addressed this, and I believe the OPs described mitigations will make the drones almost completely immune to electronic measures: Laser light between drones is not easy to jam, it can be done by shining a brighter laser on the receiving drone at the same frequency, but that requires the defense system to be tracking the drone. This will also not stop the drone, see "given they are autonomous". This means that the drone has circuit cards that likely host a multimodal transformer neural network that analyzes video frames for valid targets. Once a target is found, coordination is needed to deploy the optimal weapon on a given target and to avoid wasting munitions such as multiple suicide drones attacking the first infantry soldiers perceived. But without coordinating the drone still fly and still fight, just less effectively. Note that the reason why the drones are difficult to attack with microwaves would be the use of metal shields over the electronics, forming a faraday cage. Theoretically this defense is perfect, although in reality there have to be things like exposed gps antenna and camera sensors. Finally I think you are not fully updating on the consequences of drones as seen in an actual battlefield. In reality drones are so fast that were they autonomous and built in greater numbers, I suspect they would be strictly dominant against most land forces. Any defense you propose is too expensive when there are hundreds to thousands of drones in low flying swarms, traveling at 100-300 mph. They also can react extremely quickly and the drone commanders can order them to concentrate forces in key areas, bypassing defenses and defeating conventional ground forces.
2the gears to ascension1mo
Is this what you meant to say?
4Yair Halberstadt1mo
Now I don't doubt we'll be seeing incremental changes here, and more uses of drones and autonomy, but I don't think this is going to rewrite the rules of war anytime soon.

I'm interested in how much processing time Waymo requires. I.e. if I sped up the clock speed in a simulation such that things were happening much faster, how fast could we do that and still have it successfully handled the environment?

It's arguably a superhuman driver in some domains already (fewer accidents), but not in others (handling OOD road conditions).

Reminds me of the risk from mirror organisms. Basically you create cyanobacteria using right handed amino acids instead of left handed ones, and it outcompetes everything else because nothing can predate it (it's indigestible to normal organisms).

2Gerald Monroe2mo
Right.  It's the same idea.  The one I gave above is essentially that https://academic.oup.com/nar/article/34/1/e7/2401668 there are other amino acids nature didn't have a choice between.  Probably a photosynthetic bacteria that was 'designed', by accident or evolution, from a larger set of amino acids or on a different world would have an efficiency or defense advantage.  So slowly over time with each generation, a small advantage means slightly more of the algae in an ever expanding bloom is this alien life.   I think the reason this doesn't normally happen is within the possibility space of earth life, there are rarely truly dominant advantages, so eventually this bloom would hit a limit of the biomes it's optimized for, or other creatures would evolve to eat it, or it would have it's advantages copied by gene transfers. But in theory an alien organism could have an efficiency advantage in all conditions that earth has, and obviously it can't be eaten and it's genes are not intercompatible with earth life.

Just read in Morris's Righteous Victims:

Agriculture was primitive, with little irrigation. During the first half of the nineteenth century, land was usually owned by the villagers privately tely or collectively. The second half of the century saw the growing impoverishment of the villagers, in large part owing to more efficient Ottoman taxation, and a great deal of rural land was bought up by urban notable families (in Arabic, a'yan), who had accumulated their new wealth as Ottoman agents, especially in tax collection, and through commerce with the West. By the early twentieth century, villagers in dozens of localities no longer owned their land but continued to cultivate it as tenant farmers.

I think this exactly describes our neighborhood. Our garden backs on to a park. On the other side of the park is a foundation school. Next to the foundation school are kindergartens and daycares of all different ages. Shops are all a seven minute walk away (closer ones are under construction).

The pavements are extremely wide and beautifully planted, with cycle paths everywhere. There's a play area every few hundred metres, as well as benches, shades, exercise machines etc.

It is dense (approximately 25000 people in 2.5 Square Kms, but that's what makes it p... (read more)

The point isn't that chatbots are indistinguishable from humans. It's that either

  1. Chatbots are already conscious

Or

  1. There'll be no way to tell if one day they are.

Both should be deeply concerning (assuming you think it is theoretically possible for a chatbot to be conscious).

1GradualImprovement2mo
Yair, you are correct.    Point 2) is why I wrote the story. In a conversation about the potential for AI rights, some friends and I came to the disconcerting conclusion that it's kinda impossible to justify your own consciousness (to other people). That unnerving thought prompted the story, since if we ourselves can't justify our consciousness, how can we reasonably expect an AI to do so? 

This is one of the reasons I think people should read a lot more local news. Fatal car crashes are rare enough in an area of 100,000 people that they'll usually be reported. Also positive and more relevant news is much more common - this restaurant just opened, that development just started, local job ads, deep dives into the mayoral candidates, etc. Generally end up better calibrated, less depressed, and more focused on issues that actually effect you.

It seems you've stepped on quite a land mine here, and the following is mostly just vague guesses.

As far as I can make out it dates back to the Ottoman land code of 1858 where for various reasons a lot of land was declared owned by the government, which would collect a tax in lieu of rent.

So in one case the Ottoman empire sold a large tract of land to a Lebanese Effendi, who then sold it to the Yishuv. There was an village on this land which had been settled for some 60+ years, and despite protests to the Ottoman government the villagers were all evicted b... (read more)

My impression reading the book is that whilst the Israelis definitely had ambitions on far more, they were prepared to grudgingly accept the partition plan. They were heavily dependent on international support for their cause and far weaker than the surrounding Arab States so would have had no incentive for war in the founding days of the state if it wasn't forced on them.

Certainly there weren't any concrete plans by the Yishuv or Haganah to take over any land pre the UN partition vote. This is despite the Haganah generally being quite well organised in th... (read more)

Another error - whilst chat GPT is correct that 41 is a centred square number, it's formula is wrong.

2VipulNaik3mo
Thanks, fixed now! Sorry I missed that.

To be blunt, I think this post completely, and possibly wilfully, misrepresents Scott's post. Every single one of your points appears to be a strawman:

1. The post implies it's important not to change the words we use to refer to minority groups, but Scott doesn't say why.

The post makes the claim hyperstitious cascades are bad, where previously innocent words that noone took offense to become taboo, not changing the words we use to refer to minority groups. He also explains his reasoning perfectly clearly, as gjm points out.

2. The post needlessly drags its ... (read more)

1[deactivated]3mo
A major claim I’m making is that this has never actually happened in history, and certainly not in any of the examples Scott uses. Words become taboo because they are used offensively.
-2[deactivated]3mo
I don't think such an attempt has ever happened and succeeded. I'm open to counterexamples, though. I think Scott's account of the history of the term "Black" is dubious.

Demilitarisation would be necessary initially for this to be acceptable to Israel (apart from such small arms as are necessary to maintain law and order, and a coast guard to prevent smuggling), but could be removed long term given continued peace and economic integration.

Also it should be acceptable for Gaza to enter into peace treaties with e.g. Egypt to defend itself from Israeli aggression, and Israel to defend itself against Egyptian aggression.

Finally it should be expected that Gaza will maintain authority over it's airspace and waters.

I think this w... (read more)

He was responding to the selfish reason - "my children will take care of me in my old age", not the social reason - "my children will be a small part of keeping the economy running in my old age".

In a world where Jews have so little cultural identity that they're happy to relocate Israel to Moldova, Palestinians and Israelis might as well have so little national identity that they're happy to live together in a one state solution.

2Arcayer4mo
Cultural identity, in any reasonable world, is about the people around you and your way of life, not where you are on a map.

This hasn't historically always been the case - there was widespread public acceptance of homosexuality in the first 500 years of Islam's existence, with homoerotic poetry being a staple of their culture - see e.g. here.

Judaism also unequivocally rejects homosexuality, yet many modern orthodox synagogues happily have openly gay members of their congregation. So this doesn't seem quite as impossible as you make out.

2Jay4mo
In 14 centuries of Islamic history from Spain to Indonesia, with limited travel and much regional variation for most of it, there will be many opportunities to find examples that match our own culture's Current Thing.  Some Muslims are hypocrites; some Westerners look for homosexual subtext where none was intended.  Many Muslim empires have risen in vigor and fallen in decadence.  Still, the orthodox position is clear - homosexuality is both sinful and illegal.  I've seen a Jew eat pork and laugh it off; it would be a mistake to make pork a key component of an appeal to militant Jews. More to the point, in this era gay rights are associated with the West at its most liberal, which is exactly what Islamists oppose.  Activity that might have been tolerated a thousand years ago is now perceived as a Western obsession and its practitioners as enemy sympathizers.

From what I know, the international peacekeeping force in Lebanon does precisely nothing. They leave whenever there's fighting, and have no interest stopping Hezbollah from rearming. I literally do not know if it would make any difference if they didn't exist at all.

This is to be expected. UNIFIL has no skin in the game, and would take significant risk if they attempted to stop Hezbollah operations.

As Yovel said, wildly off the mark.

Just one minor extra point - the poorer Israeli working class is mostly supportive of Netanyahu. It's the middle and upper classes who oppose him, who they see as corrupt.

Also the youth is generally more right wing in Israel, and the older generation more left wing.

In general you can't copy your model of politics from one country into another and expect it to accurately predict what's going on.

Isn't that the point? Where we stand now, we have to make a decision without knowing if there will or won't be a treacherous turn...

But wouldn't that be the case for any organelle, even one which is inherited from both parents?

3Malmesbury5mo
I would guess that when organelles are inherited from both parents, the traitor organelle is disadvantaged by its burden on the host, but advantaged by it's ability to be the predominant organelle in the offspring. If the cost-benefit is favourable, then the traitor organelle will take over. OTOH, if only one parent transmits the organelle, the advantage disappears but the burden remains. So I'd expect that it makes it more difficult for traitor mitochondria to invade. Hopefully that makes sense!

But by making more copies of itself/poisoning other mitochondria isn't it more likely to end up in the female gamete?

2Malmesbury5mo
Not quite, if it's less efficient at doing the normal mitochondria work, it puts a big burden on the cell, who is then less likely to reproduce.

Thank you very much for this excellent post!

Would you be able to give a more detailed explanation of Organelle competition? I'm afraid I didn't understand at all how having different types prevents it.

2Malmesbury5mo
It's not so much the different types in themselves that prevent competition, but having multiple types make it possible to have a mechanism that forces all organelles to come from only one pre-selected parent. If all organelles come from the female, then a rogue mitochondria cannot take over by making more copies of itself or by poisoning other mitochondria, because the only way to make it to the next generation is to be in the female gamete, period. In other words, there's not much an organelle can do to increase in frequency, aside from improving the overall fitness of the organism. Does that make more sense?

I think this is a suitable mechanism for grants for well defined mathematical problems, but not for more vaguely defined ones. To be frank, the well defined ones tend to be less interesting.

For example the Cray problems are all well defined, and are certain very interesting, but mostly not actually that fundamental to mathematics.

The Hilbert problems are in some ways far more important, and most of these are more vague questions or directions for research than an actual concrete conjecture.

Nobody would have been trading shares in godels incompleteness theo... (read more)

4DaemonicSigil6mo
However we currently do it, I guess; I don't have any improvements in this particular direction. Though do note that P vs NP, and P vs PSPACE, and BQP vs P or NP are all examples of precisely-defined problems that are also very significant. These are problems that I would much rather have an answer for than the Riemann hypothesis. Though even there, the process isn't guaranteed to generate an interpretable proof.

Hi, and welcome to Less Wrong!

It's great to see you trying to improve your decision making process, and being prepared to put your work in front of the public!

Here's a few thoughts on how to improve what you wrote:

Firstly, a terminology note. A theorem usually refers to a statement about the world that is either true or false, like "1+1=2" or "humans have 4 legs". What you have is not a theorem but a technique to improve decision making.

Secondly, you present a framework with some seemingly arbitrary calculations. Why is it (a*b+c)/(d+e), and not (a+b*c/d*e... (read more)

I think the biggest risk of DACs is that it incentivises people to fund contracts they don't actually want fulfilled to milk the proposer for cash.

My expectation is that if this becomes mature you'll get traders which try to predict which contracts won't be fully funded, and then push them up to say 50% (after which pushing them further risks them actually getting fully funded).

This not only discourages proposers from putting up contracts (to easy to lose money), but also makes it harder for users to easily see which contracts are worth funding and which a... (read more)

1teageegeepea2mo
Isn't whether a project is "worth funding" depend on whether you think it's proposed output is an actually valuable public good? If it is, then you shouldn't mind not getting a refund that much. If it isn't, there are other places to look for arbitrage.

I don't think this actually solves the freerider problem. It solves the coordination problem.

In your toy example Bob was able to specify that he wouldn't build anything unless everyone signed the contract - if this is possible you don't really need the dominant assurance part, as it's still worth it for the freerider to sign up, so long as they assume there's a decent chance others will as well (and if they don't, that's going to discourage non-freeriders as well).

In real life though, this is impossible. Instead the contract is along the lines of "I will d... (read more)

Love the idea (contributed 10 dollars), but I don't love the title - it feels a little clickbaity (I expected an article about asteroid deflection, not about DACs).

3Yoav Ravid6mo
Yeah, something descriptive like "A platform to fund public goods using dominant assurance contracts" or "A refund bonus crowdfunding platform to subsidize public goods" would be better in my opinion (and the jargon wouldn't be a problem in this community).

Usually ends up making a lot of mess as frozen chips fly all over the counter, then quickly melt.

I think to answer this you need to break things down by region.

A look at our world in data shows that China was a 5th of the world population at the start of the first millennium, and it shrunk over the course of the millennium. Europe also shrunk, coinciding with the fall of Rome.

Africa, the Americas, and Australia gre at what looks like a fairly normal rate, but they are a much smaller percentage of the population.

So the explanation is likely to be local to the most populous regions, rather than some global event.

2Ege Erdil6mo
McEvedy and Jones actually discuss a regional breakdown in the final section of the book, but they speculate too much for the discussion to be useful, I think. They attribute any substantial slowdown in growth rates to population running up against technological limits, which seems like a just-so story that could explain anything. They note that the 3rd century AD appears to have been a critical time, as it's when population growth trends reversed in both Europe and China at around the same time: in Europe with the Crisis of the Third Century, and in China with the fall of the reconstituted Han dynasty and the beginning of the Three Kingdoms period. They attribute this to technological constraints, which seems like an unsupported assertion to me. The other important population center is India, where we have very few records compared to Europe and China. Datasets generally report naively extrapolated smooth curves for the Indian population before the modern period, and that's because there really isn't much else to do due to the scarcity of useful information. This doesn't mean that we actually expect population growth in India to have been smooth, just that in the absence of more information our best guess for each date should probably be a smoothly increasing function of the date. As McEvedy and Jones put it, "happy is the graph that has no history". I agree that locations isolated from Eurasia would most likely not show the same population trends, but Eurasia was ~ 75% of the world's population in the first millennium and so events in Eurasia dominate what happens to the global population.

Sorry, could you elaborate what you mean by all the way up?

1Jeffs6mo
All the way up meaning at increasing levels of intelligence…your 10,000 becomes 100,000X, etc. At some level of performance, a moral person faces new temptations because of increased capabilities and greater power for damage, right? In other words, your simulation may fail to be aligned at 20,000...30,000...

I would be skeptical such a proof is possible. As an existence proof, we could create aligned ASI by simulating the most intelligent and moral people, running at 10,000 times the speed of a normal human.

1Jeffs6mo
Okay, maybe I'm moving the bar, hopefully not and this thread is helpful... Your counter-example, your simulation would prove that examples of aligned systems - at a high level - are possible.  Alignment at some level is possible, of course.  Functioning thermostats are aligned. What I'm trying to propose is the search for a proof that a guarantee of alignment - all the way up - is mathematically impossible.  We could then make the statement: "If we proceed down this path, no one will ever be able to guarantee that humans remain in control."  I'm proposing we see if we can prove that Stuart Russell's "provably beneficial" does not exist. If a guarantee is proved to be impossible, I am contending that the public conversation changes. Maybe many people - especially on LessWrong - take this fact as a given.  Their internal belief is close enough to a proof...that there is not a guarantee all the way up. I think a proof that there is no guarantee would be important news for the wider world...the world that has to move if there is to be regulation.

But that's not affected by any actual in-game information because you don't really get any information.

You keep on asserting this, but that's not actually true - do the maths. A player who doesn't update on them still being alive, will play fewer rounds on average, and will earn less in repeated play. (Where each play is independent).

The reason is simple - they're not going to be able to play for very long when there are lots of bullets added, so the times when they find themselves still playing are disproportionately those where bullets weren't added, so they should play for longer.

1dr_s7mo
But the thing is, updating on being still alive doesn't change anything - it can never drive your estimate of n up and thus save you from losing out. It could convince you to play if you aren't playing - but that's an absurdity, if you're not playing you won't get any updates! All updating gives you is a belief that since you're still alive, you must be in a low-bullets, high-probability world. This belief may be correct (and then it's fine, but you would have played even without it) or wrong (in which case you can never realise until it's too late). Either way, it doesn't swing your payoff. In your added bullets scenario thinking about it there's a bit of a difference because now a strategy of playing for a certain amount of turns can make sense. So the game isn't time-symmetric, and this has an effect. I'm still not sure how you would use your updating though. Basically I think the only situation in which that sort of updating might give a genuine benefit is one in which the survival curve is U-shaped: there's a bump of mortality at the beginning, but if you get through it, you're good to go for a while. In that case, observing that you survived long enough to overcome the bump suggests that you're probably better off going on playing all the way to the end.

You don't get told no, you just guess from the fact you're still alive.

but I picked the game description very deliberately to show the effect I was talking about, so obviously changing it makes things different.

On the contrary, it doesn't show any such effect at all. It's carefully contrived so that you can update on the fact you're still alive, but that happens not to change your strategy. That's not very interesting at all. Often a change in probabilities won't change your strategy.

I'm simply showing that with a slight change of setup, updating on the fact your still alive does indeed change your strategy.

1dr_s7mo
If you don't get told I don't think it does then, no. It just changes the probabilities in a more confusing way, but you still have no particular way of getting information out of it. As a player, who by definition must be still in play, you can't deduce anything from it. You only know what whatever your odds of surviving an extra round were at the beginning, they will go down with time. This probably leads to an optimal strategy that requires you absolutely quit after a certain number of rounds (depending on the probability of the bullet being added). But that's not affected by any actual in-game information because you don't really get any information.

The addition is permanent. Updating on the fact that you're still playing provides evidence that the bullet was not in fact added in previous rounds, so it's worth carrying on playing a little bit longer, whereas if you didn't update, even if it was worth playing the first round, you would stop after 1 or 2.

-1dr_s7mo
OK, so if you get told that a bullet was added, then yes, that is information you can use, combined with the knowledge that the drum only holds maximum 6 bullets. But that's a different game, closer to the second I described (well, it's very different, but it has in common the fact that you do get extra information to ground your beliefs). Even simpler, something similar would happen simply if you didn't spin the chamber after each turn, which would mean the probability of finding a bullet isn't uncorrelated any more. These details matter. I'd need to work it out to figure how it works, but I picked the game description very deliberately to show the effect I was talking about, so obviously changing it makes things different.

See my comment from earlier below, which highlights how this information is in general useful, even if in this case it happens not to be:

To give a concrete counterexample.

Let's say each round there's a 50 percent probability of adding an extra bullet to the gun.

If I didn't update based on the fact I'm still playing then I would quickly stop after a few rounds, since the probability I would see a bullet constantly increases.

But if I do update, then it's worth it carrying on, since the fact I'm still playing is evidence there still are plenty of empty chambers

2dr_s7mo
Not quite getting it - is the addition permanent or just for each round? Seems to me like all that does is make the odds even worse. If you're already in a game in which you should quit, that only is all the more reason; if you're not, it could tilt the scales. And in neither case does updating on the fact you're still playing help in any way, because in fact you can't meaningfully update on that at all.
Load More