My pet conspiracy theory is there is a fair amount of coordinated manipulation to get political posts on the HN front page. Fortunately, they are often quickly flagged to the abyss.
That’s not a conspiracy theory. Anyone who doesn’t realize at this point that online discourse is heavily engineered and manipulated is an unthinking rube.
I tried AP Pro 3 but decided to return them in favor of my v2. The v2 felt more comfortable in my ear and were good enough. I was also disappointed the heartbeat monitor only works while exercising.
Researchers analyzing 100,000+ transients in 1949-1957 astronomical plates found they were 45% more likely during nuclear tests and showed small but statistically significant correlations with UAP reports. The findings argue against simple plate defects and add to earlier work [1] that found puzzling Earth shadow deficits and aligned transients suggesting reflective objects at high altitudes.
Totally agree with your view on the symbolic context injection. Is this how things are done with code/dev AI right now? Like if you consider the state of the art.
Agreed, it would be cool, but. From that article, with my commentary (disclaimer: IANAA, I Am Not An Astronomer):
1) "The retrograde orbital plane... of 3I/ATLAS around the Sun lies within 5 degrees of that of Earth... The likelihood for that coincidence out of all random orientations is 0.2%." Not sure where he comes up with 0.2%. 5/180 = 2.8%. (I use 180 degrees, rather than 360, because I suspect that if it were not retrograde, he'd use the same argument.)
2) "the brightness of 3I/ATLAS implies an object that is ~20 kilometers in diameter (for a typical albedo of ~5%), too large for an interstellar asteroid. We should have detected a million objects below the ~100-meters scale of the first reported interstellar object 1I/`Oumuamua for each ~20-kilometer object." Huh? We barely detected this object because it's so dim. Why should we be detecting interstellar objects two or three orders of magnitude smaller?
3) "No spectral features of cometary gas are found in spectroscopic observations of 3I/ATLAS." An article today (22 July, https://astrobiology.com/2025/07/spectroscopic-characterizat...) says "Spectral modeling with an areal mixture of 70% Tagish Lake meteorite and 30% 10-micron-sized water ice successfully reproduces both the overall continuum and the broad absorption feature... 3I/ATLAS is an active interstellar comet containing abundant water ice, with a dust composition more similar to D-type asteroids..."
4. "For its orbital parameters, 3I/ATLAS is synchronized to approach unusually close to Venus (0.65au where 1au is the Earth-Sun separation), Mars (0.19au) and Jupiter (0.36au), with a cumulative probability of 0.005% relative to orbits with the same orbital parameters but a random arrival time." This probability is harder to compute (although 0.65au from Venus is nearly the radius of Venus' orbit, 0.72au, i.e. not close). In any case, so what? Why would an interstellar probe travel close to Mars or Jupiter, if they're interested in Earth? (see next point) Later (his point 8), he says the probe comes close enough to these planets to launch ICBMs at them. Ok...
5. "3I/ATLAS achieves perihelion on the opposite side of the Sun relative to Earth. This could be intentional..." Sure, if they're interested in Earth, stay away from it.
> "The retrograde orbital plane... of 3I/ATLAS around the Sun lies within 5 degrees of that of Earth... The likelihood for that coincidence out of all random orientations is 0.2%." Not sure where he comes up with 0.2%.
This part of the calculation, at least, is basically correct. The orientation of a plane in space is defined by its normal vector, so the right way to look at probabilities is in terms of solid angle. The normal of 3I/ATLAS's orbit falls within a cone around Earth's normal vector, having a half-angle of 5 degrees, and that cone's solid angle occupies about 0.2% of the full sphere.
Of course, this is only the chance of a retrograde alignment. Presumably, if the comet's orbit was prograde aligned with the Earth's to within 5 degrees, Loeb would be making exactly the same claim. So really, the relevant probability is 0.4%.
Nevertheless, I agree that the article is basically just a bunch of cherry-picked probabilities and insinuations that don't add up to much.
Also:
> "the brightness of 3I/ATLAS implies an object that is ~20 kilometers in diameter (for a typical albedo of ~5%), too large for an interstellar asteroid."
But to justify this, Loeb cites his own work showing that the object is either a large asteroid, or a comet with a small nucleus. And then he seems to have looked at some earlier spectra and jumped to the conclusion that 3I/ATLAS couldn't be a comet, so it must be a large asteroid. But of course, follow-up observations have debunked this point and clearly shown it to be a comet.
I think there's also a sampling bias here? ATLAS, the survey that discovered the comet, is specifically looking for potential Earth impactors. One assumes that would involve looking close to Earth's own orbital plane.
Case hardening. Making something which if propelled fast enough (secondary issue) and with a G force resisting detonator (secondary issue) which has sufficient integrity and inertia to penetrate as deeply as possible before exploding. Materials science in making aerodynamic rigid, shock tolerant materials to fling at the ground.
I am sure the materials science aspects have come along since ww2, as has delivery technology, but I'd say how it goes fast, hits accurately and explodes is secondary to making a case survive impact and penetrate.
I would posit shaped charges could be amazing in this, if you could make big ones to send very high energy plasma out. I'm less sure depleted uranium would bring much to the table.
(Not in weapons engineering, happy to be corrected)
I'm not sure you would want a shaped charge unless you guarantee it was pointing in the right directionatthe right time. Modern bunker design usually includes deflection tactics.
I was guessing either tungsten or depleted uranium, as for APDS, but the bomb's average density is only about 5 g/cc (14 tonnes in 3.1 m³). Length of 6.2 m times 5 tonnes per cubic meter gives a sectional density of 31 tonnes per square meter, which is about 15 meters of dirt. So Newton's impact depth approximation would predict a penetration depth one fourth of the reported 60-meter depth.
I don't know how to resolve the discrepancy. The plane wouldn't fly if the bomb weighed four times as much. Maybe most of the bomb's mass is in a small, dense shaft in the middle of the bomb, which detaches on impact?
> Length of 6.2 m times 5 tonnes per cubic meter gives a sectional density of 31 tonnes per square meter, which is about 15 meters of dirt. So Newton's impact depth approximation would predict a penetration depth one fourth of the reported 60-meter depth.
This seems to assume that the weapon would penetrate until it displaced an equal amount of dirt by mass, which seems like nonsense. Why would that be the case?
A system described in the 2003 United States Air Force report called Hypervelocity Rod Bundles[10] was that of 20-foot-long (6.1 m), 1-foot-diameter (0.30 m) tungsten rods that are satellite-controlled and have global strike capability, with impact speeds of Mach 10.[11][12][13]
The bomb would naturally contain large kinetic energy because it moves at orbital velocities, around 8 kilometres per second (26,000 ft/s; Mach 24) in orbit and 3 kilometres per second (9,800 ft/s; Mach 8.8) at impact. As the rod reenters Earth's atmosphere, it would lose most of its velocity, but the remaining energy would cause considerable damage. Some systems are quoted as having the yield of a small tactical nuclear bomb.[13] These designs are envisioned as a bunker buster.[12][14] As the name suggests, the 'bunker buster' is powerful enough to destroy a nuclear bunker.
Shape can change it to be arbitrarily bad; 14 tonnes of 5-micron-thick Eglin steel foil spread over a ten-block area wouldn't penetrate anything, just gently waft down, although it could give you some paper cuts. I suspect it can't make it much better, except in the sense of increasing sectional density by making the bomb longer and thinner, which we already know the results of.
Velocity doesn't enter into Newton's impact depth approximation at all. It does affect things in real life, but you can see from meteor craters that it, too, has its limits.
Target characteristics, no idea, but in a fast enough impact, everything acts like a gas. It's only at near-subsonic time scales that condensed-matter phenomena like elasticity come into play. Even at longer time scales the impact can melt things. This of course comes into conflict with the design objective of the bomb acting solid, so that it penetrates the soil instead of just mixing into it, and can still detonate when it comes to rest. I feel like buried plates of the same metal would have to be able to deflect it? And there are plenty of other high-strength alloys.
I did some quick calculations: The energy of the impact from the stored kinetic energy gained by falling fro 15,000m is about the same as half a kiloton of TNT going off. That's focused into a circle just 80cm in diameter.
Your calculations appear to be off by a factor of ~1000. Not half a kiloton, but half a ton (~500kg), assuming fall in a vacuum (upper bound on impact energy):
Python 3.10.12 (main, May 27 2025, 17:12:29) [GCC 11.4.0] on linux
Type "help", "copyright", "credits" or "license()" for more information.
MOP_potential_energy = 13607*9.8*15000 # E = m*g*h
MOP_potential_energy
2000229000.0
TNT_specific_energy = 4.184e9/1000 # joule/kg
TNT_specific_energy
4184000.0
MOP_potential_energy/TNT_specific_energy
478.0662045889101
Yet setting off half a tonne of TNT on the ground, or even just under it, won't penetrate 60 meters deep, or even 15; it will just blast open a shallow crater. A shaped charge will do only a little better.
No real secret sauce, the weapon weighs almost 30,000lbs and most of it is just hardened metal to make it heavy. The warhead is only ~5,300lbs of explosive
reply