K^2

If that's too cliché, we can flip the script. Dwarf planets are just planets, and IAU definition planets are orthoplanets.

Yeah, but what's the actual utility here? Mercury-sized object out in the Kuiper Belt would be just another snowball. A big snowball, but the thing is, one might actually exist, likely, even, based on some models, and we haven't found it yet, and we aren't really losing any sleep over that. For all intends and purposes it'd be just another KBO. Describing such an object out in the KB as a planet would be silly. In contrast, had the Mercury been as small as the Moon, or even as small as Pluto, it'd still be a significant contribution to inner system. Besides, we already have a number of categories based on physical characteristics. We have icy objects, terrestrial objects, ice giants, gas giants, and brown dwarves. Many of these can take a number of possible roles, from dwarf planets, to actual planets, to moons. Saying Pluto is an icy dwarf planet says volumes more than "protoplanet" in your definition. While for Titan and Ganymede the defining trait is being icy moons of a gas giant, despite very nearly making the cutoff for a planet in your definition. And the Moon is nowhere near the same kind of object despite being smaller, because it is orbiting the Sun directly. Describing Earth-Moon system as terrestrial binary planet puts it into the correct perspective. And then we ought to revisit the problem of measuring. We don't have good numbers for a lot of exoplanets out there. We either have lower bounds on mass if we discovered them from spectroscopy methods or estimate from size based on transit method. The situation is going to be way worse with the exomoons. So were we to define things based on mass, whe we will be discovering objects in other star systems, we will neither be able to tell whether they are moons or planets or protoplanets, nor will we actually care, because it doesn't really tell us anything about that particular star system nor, indeed, about conditions on that object. On the other hand, we can tell if the object orbits the star directly or if it's going around a parent object, basically, so long as we can observe it at all. So we'll immediately know if we're dealing with a moon or binary planet, and that distinction tells us something about evolution of that star system, which is what we're really after here. P.S. I personally have no objections to renaming "dwarf planet" to something else. "Protoplanet" is still a bit misleading, because it kind of implies that it can become a planet - e.g. protoplanetary disk. But I honestly don't care what the term is, so long as we have one. Likewise, I don't particularly care if the binary planets are called double planets instead. I prefer the term binary, but to me personally, the actual choice doesn't make much of a difference.

Applicable to what? Are you going to apply your binary definition to stars? So what do you call a star orbiting another star, when barycenter is inside the later? Clearly, you don't want to call it a binary. The planet-moon vs binary planet distinction only makes sense in context of discussing star-planet-moon interactions, because it's actually tells us things about the evolution of a star system. The normal process is for planets to form from protostellar disk and moons to form from protoplanetary disk. That makes the later always gravitationally bound to the planet more than to the star. When we see an exception, something interesting happened. Either the planet migrated into the inner system from where it formed or there was an impact event. If I tell you that Earth has a moon, you say, "So what? So does almost every other planet." But that misses the mark. The Moon is unlike every other body in Sol that we've studied. Having taxonomy that lets us identify these things in other star systems is crucial. When we start being able to distinguish some of the known exoplanets into multiple bodies, under current definitions, all we're going to say is, "They're moons, probably," because there is no way for us to measure barycenter and most of these really will be just moons, so we'd be just guessing. But we need to inform not just the scientific community but the public when we find something interesting, and finding pairs of objects that are pulled by the parent star more than by each other is interesting. These are discoveries worth knowing about. And that means it's worth distinguishing them in nomenclature. If we find systems like Earth-Moon out there, we should be identifying them as binaries from the moment they are resolved, because it's not just another moon that we expect to be there. It's a signature of a catastrophic event that took place during formation of the system and tells us more about the development of planetary system around that star than any moon ever could. As for rogue planets, honestly, who cares? While space is probably filled with them, without a proximate star, truth is, we aren't going to find any any time soon. And I would give good odds that we'll never observe a rogue system composed of two similarly-sized objects where calling one the moon of the other would seem odd. Sure, they must statistically exist, but we aren't going to be studying or cataloguing them. So, again, who cares? I understand wanting to make definition that cuts across all cases everywhere. But that's just stamp collecting, in my opinion. We need practical definitions for practical cases where naming things differently or the same tells us something useful. The definition of the planet was updated to exclude Pluto because calling all sufficiently massive objects planets was becoming useless. We had to change the definition to reflect how planet interacts with the rest of the star system. Just like the tug-of-war definition I'm arguing for, our current official definition of a planet cannot be applied to a rock drifting through interstellar space. It only makes sense for an object orbiting a star. And how massive an object has to be to be considered a planet depends on how close it is to the parent star. Being a planet isn't an intrinsic property of an object. Neither should it be of the planet-moon system. The useful definition of a moon is one that acknowledges interaction with parent star, because it lets us distinguish boring moons from exciting binary planets.

That would be 0 atm. :p I'm not feeling like looking for a no-paywall version right now, but their abstract hints at expectations of limited metastability with the right sort of mixture. We're certainly not talking 1atm, but something in a few GPa range could already be handled with a metal shell. You still need to get material to that pressure first, but maybe creating small sealed pellets is viable, which you can already find uses for. The holy grail, of course, would be figuring out how to make flexible cable out of this stuff. Then you can shape it into whatever shape you need for particular application.

Haven't done a lot of reading on this yet, but superficially looks legit. Room temperature superconduction, albeit, in a diamond anvil cell. https://phys.org/news/2020-10-room-temperature-superconducting-material.html

You keep making the definition more and more complex. "A body is a moon if and only if there exists another body qualifying as planet or dwarf planet, such that the barycenter of the two is on a quasistable trajectory entirely contained within the surface boundary of the later." This sounds like a theorem from convex geometry. When your definitions for categories start to sound like that, you have a bad taxonomy. All just to try and squeeze the Moon into a definition of the moon. It's not helpful. Compare to a clean taxonomy. An object is a planet if it meets current criteria of a planet and is pulled by the primary star(s) stronger than by any other object. Done. Most importantly, as I've indicated earlier, the above is something we can easily measure for exoplanets and exomoons once we detect them. In contrast, measuring sizes of exoplanets is very, very hard. And in many cases, estimating barycenter precisely enough to know whether it falls within an exoplanet or not is going to be impossible. Think about it, the first extrasolar planetary systems we are going to detect are going to involve rather large objects. Perfect candidates for being double planets. Under current definition with barycenter we won't be able to classify them as moon or double planet, because we simply won't be able to tell with enough precision where barycenter falls. If we go with tug-of-war definition, we won't need to. We'll be able to catalogue them as double exoplanets or exomoons as appropriate. The case with the Moon is the same as with Pluto. Taxonomy should be useful for studying the world around us, not just give things convenient labels everyone is used to. For all the reasons IAU pulled the trigger and made Pluto not a planet, they should update definitions for moons and make Moon a planet.

It seems like we ought to be able to tell if a bunch of asteroids suddenly came from outside Solar System during some specific time period in the past. We have found some debris that came from long-ago impacts. I don't know if any go quite that far into the past, though. So maybe there really is nothing for us to analyze. But if we have some rocks that we can identify with specific impact events as likely being from that impactor, we ought to be able to analyze their composition which should tell us where they came from. It's not an exact science, as you can find odd rocks with odd things in them, but if you get a large enough sample, there are patterns. Sol formed from a remnant of a supernova which seeded the system with heavy metals. Relative concentrations of all the things the asteroids and comets are made out of, while varying greatly across the system, taken together form a sort of signature of origin of our Sol. It would be highly unlikely that, given a large enough sample, we would not be able to establish if a bunch of asteroids originated outside Sol. That said, I don't know if we have a large enough sample size or if anyone tried to look for these particular anomalies.

That makes the definition even more arbitrary than it already was. Tug-of-war is straight forward. Does Moon primarily orbit Earth or Sun? In this case, the Sun. Our Moon's trajectory around the Sun is convex. Unlike every moon in Solar System, it never accelerates away from the Sun, because Sun's gravity is the dominant force on the Moon. Earth's effect is secondary, and qualitatively, Earth's and Moon's trajectories around the Sun look the same. Whereas all the moons of other planets follow a complicated curve with twists as their parent planet pulls them to accelerate away from the Sun. If you list all the ways in which Moon is like any other planet of Sol, it's going to be a fairly extensive list. The list of similarities to moons is, "There exists another body, such that the barycenter of the two happens to fall inside the later." That's the only thing they have in common. It's entirely artificial and makes for bad categorization.

That's an easy one. Together they are binary planetary system called Risk. I don't care if it's not official, there is objectively no other option. And yes, point taken on systems evolving, but it's a difference between a significant qualitative change and, "Well, it drifted out, so it's not a moon anymore." Worst part is that Moon's orbit is elliptical. So we know that there will be a period in Earth's history, even if we're not around, when you'd have to check your calendar to know if the Moon is a moon today or a planet, because it would depend on whether it's near its apogee or perigee.

This is the current IAU definition, and I've outlined the problems with barycenter definitions just above.

No. Earth only has to clear its orbit of asteroids to be considered a planet. Same goes for Moon. IAU is perfectly happy with the idea of binary planets. They just use barycenter definition, which I find to be faulty. Primarily, because moons shift in orbit due to tidal interaction. Case in point, Moon is drifting away from Earth, and barycenter of the Moon will end up outside of Earth's surface. So an Earth-Moon system is bound to become a binary planet by IAU definition. And it's not one of these weird, "Oh, something might happen to knock an object out of its orbit, changing its category." This is in progress, and I don't think a question of whether Moon is a planet or a moon should be a matter of timing. The tug-of-war definition is much better, because you're pretty much guaranteed that it's going to stay consistent for the majority of the system's life. You can still end up with edge cases where orbital drift is going to change the tug-of-war balance, but majority of these are going to be for moons drifting inward, and in these cases the two will eventually become a single planet, so a change of status is unavoidable there. It's also consistent with origin of the body. Moons that form from planetary proto-disc are going to be tightly bound to the planet they've formed with, and so will be moons by tug-of-war definition. Our Moon has formed from an impact. It's effectively a lithobraking capture. It originated as a planet and ended up in orbit that a naturally formed moon is extremely unlikely to occupy. If it did not have energy and angular momentum during impact to take that high of an orbit, it probably would have merged with Earth producing just some moonlets in lower orbit from the debris. In other words, it's pretty unlikely that a capture that would otherwise be qualified as a planet would end up in a low enough orbit to be a moon by tug-of-war definition. It doesn't matter a whole lot for Sol. It literally is just a matter of whether we want to rename the Moon, because it won't make a difference to any other celestial body I'm aware of. But consistent system of definitions is important for extrasolar bodies. We are discovering more and more exoplanets out there. Soon we will be discovering exomoons. And estimating planet sizes is tough. Figuring out if potential exomoon's barycenter is within or without the parent exoplanet might be difficult or impossible. And in many cases, the orbits might be elliptical enough for it to vary! The tug-of-war definition is unambiguous and measurable. If we can estimate an orbit a potential exomoon takes around its primary, we can estimate the gravitational pull. And that means we can say with high degree of confidence, or at least, way better than with current definition, whether we're looking at a planet-moon system or a binary planet. And if we are going to be studying the star systems out there to get better understanding of how they form and evolve, we do need a consistent nomenclature that we can actually apply. And applying it to every star system out there except for Sol would be a little silly. Changing moon definition to be based on relative gravitational pulls of the star and planet is going to provide the least inconsistency. And by that definition what you see in the sky is no moon.

Pluto definitely can't be a planet or we'll have like a dozen more we'll have to add. Moon, on the other hand, totally a planet. It's silly that we count it as Earth's satellite, when Sun pulls on it a lot stronger than Earth does. Moon orbits the Sun and happens to share the orbit with Earth. So we should bring the count back to 9, with Earth-Moon system counted as a binary. I know, I know, some people might complain that Moon not being a moon would get confusing. But if Moon is reclassified as a planet, it would need to be promptly renamed after a Roman deity, and honestly, the only sensible choice is to call it Luna. So that should resolve the confusion. Well, in English. Other languages will have to find their own workarounds. P.S. I have one final argument. If we say that Moon/Luna is a planet, humanity is technically an interplanetary civilization.

You are already making an assumption that rocket's TWR is maintained constant, so it's moving up with a constant acceleration. That's not strictly speaking true for typical ascent profile, but if you are going to make that simplification, then you can use the formula you already have for falling with drag and use it for ascent with drag. Simply substitute a = (Δv - g t1) / t1 in place of g and flip the sign so that you're ascending.

That is very clearly a joke article.

There are a lot of theories out there that reformulate existing mathematical basis and aren't technically wrong, but also entirely useless. E.g., I can technically use Gauge Theory to state "Theory of Everything," which will produce a Lagrangian that must be satisfied by the universe to account for all known physics, including relativity and quantum effects. But if you try to derive anything from it, the integrals diverge even if you try to regularize them, so that formulation is useless, and this is why we still consider quantum gravity an open question. So @Kerbart might have actually gotten it right... Not that I'm willing to spend time to figure out if this is actually valid and useless or just BS through and through for reasons outlined by @mikegarrison

The CFI I had teaching me to fly taildraggers didn't know they were also called "conventional gear". So yeah, that name's definitely going out of fashion. Though, it's still in a lot of manuals.

Even if you just assume that light is pulled by gravity like a massive object, you'll quickly find that the angle by which light is bent as it passes massive objects is wrong. And yeah, we WOULD have derived SR. I mean, it's basically what Einstein did. But GR is another matter. And GR is what's responsible for gravitational lensing, orbit of Mercury, and GPS satellite's clock running fast. And the math on which we would have been able to build GR properly, rather than via the geometric ansatz Einstein pulled out of who knows what inspiration, was only developed in late 50s. Based on when papers were published properly describing GR in terms of Gauge Theory, I estimate we'd have actual GR in the 90s if it weren't for Einstein's work, even if we knew about all of the aforementioned effects. On the plus side, we wouldn't have that messy differential geometry based theory they still teach in universities. I mean, it works, and it's not wrong, but it feels very arbitrary, and leads to some bad intuition, because people try to think in terms of embedded geometry no matter how well they know that solutions to Einstein Field Equations cannot be embedded in general.

Correct, but we don't really need to understand relativity to correct for it. Just measure that time on GPS satellites runs a little faster, observe that the difference is fairly consistent, and compensate for it. Of course, the fact that we are actually using theoretical values predicted by General Relativity and it gives us the necessary precision is one of the proofs of the theory.

There are actually two kinds of vorticity going on, both to do with circulation. Wingtip vortices is one, but you also have a vortex sheet layer. A totally laminar flow cannot generate lift, because it cannot produce circulation in the air. In order for there to be net circulation, there has to be a layer of turbulent air separating flow above and bellow the wing. So even a theoretical infinitely long wing, the kind that can't possibly have wing tip vortices, because it lacks wing tips, and one that produces even lift along its entire infinite length, would still produce a turbulent vortex layer behind it. This is also why the trailing edge is sharp. It's not actually the most aerodynamic shape, as you lose energy to that vortex layer, but you need vortex sheet for lift, and sharp trailing edge is necessary for Kutta condition which you need to shed the vortes sheet. Far behind the airplane, the wingtip vortices and the vortex sheet merge into a single wake.

I'm not sure that's actually strictly true. Certainly, if we treat black hole as static, immovable object, then yeah. No time-like curves lead out of the interior region. But hypothetically, if we start bending space-time with something even more extreme, it might be possible to make parts of the interior briefly accessible. And it might not require anything out of science fiction, either. For example, I'm not sure what will happen to a stellar black hole passing through ergosphere of a supermassive black hole with extreme angular momentum. Intuitively, I expect the event horizon of the former to shift and possibly even shrink, but I ain't doing the math on this one. More relevant to what I was talking about, however, GR breaks down just outside the event horizon. Very, very close to it, but still on our side. And the particle dynamics there can be relevant to the way black holes emit Hawking radiation. So it's actually something physical, potentially measurable, and we are kind of guessing on, because we don't have the math to describe what actually happens there. Hawking himself suggested that whatever quantum soup is simmering just above the event horizon, it's probably key to squaring the fact that black holes emit radiation with cosmic censorship.

Ooof, so that's a lot to unravel, actually. Part of it is that the word "time" means so many different things. First, GR prohibits particles from traveling FTL locally. So if you take two objects close enough to each other that we can disregard space-time curvature, they can move at most with the speed of light relative to each other. Once you either increase the distance or start twisting space-time into a pretzel, you can actually change that. Again, case in point, expansion of the universe. Far-away galaxies are flying away from us at FTL speeds because space itself is stretching out between us at sufficient rate. And it's why we have FTL concepts like Alcubierre Drive in the first place. In QFT, we treat space-time as flat. (Usually. There are field theories in curved space-time. Useful, for example, if you are talking about particle physics inside a neutron star.) So the causality violations in QFT are local, and it looks off the bat like an incompatibility between GR and QFT. And it sort of is, and is part of why Quantum Gravity is so flipping hard, but it's not actually a contradiction. General Relativity is a mean field theory. (Specifically it's a mean field approximation of a gauge field theory on the Poincare group.) That's a smarty-pants way of saying that GR inherently averages out motion of particles to a classical path. You don't have to understand the full mind-melting madness of building a particle theory in curved space-time to gain some appreciation for how this works. Lets take the simplest particle represented as a field we all know. A photon. Light, or just electromagnetic radiation in general, is a great example, because we kind of have intuitive understanding of how light behaves. And it's also relatively easy to picture it as a wave, so that helps. First of all, just a touch of E&M. How does an electromagnetic wave propagate? When an electric field changes at a point, it induces a magnetic field around it. In turn, a change in magnetic field induces an electric field. So if you take a charged particle and start wiggling it, it results in an electric field nearby alternating, which causes the magnetic field to alternate, which causes the electric field further out to alternate, and so on. But the identity of the source doesn't really matter. Every point of space around the emitter results in its own spherical wave echo spreading in every direction. So this is the kicker. What path does the electromagnetic wave take? It looks like it will take absolutely every possible path, as every point in space that gets excited by a wave produces its own spherical wave which excites all the nearby points of vacuum and so on. But we know from our every day experience that light doesn't normally just bend around to reach every corner of space. It appears to propagate out in straight lines, casting shadows. What gives? Well, you have to take into account interference. Electromagnetic wave has a phase. If the phases of two waves reaching the same point match, they add up. If they are exactly out of phase, they subtract from each other. As the wave is re-emitted by every point in space, the overall phase is related to the total distance traveled along the path. If we take all the possible paths, we get a random mixture of phases for all the possible distances, and an average of random selection of phases will add up to zero. So while light can be considered to take all possible paths, they all cancel each other out. Well, all except for a few special paths. This gets a touch mathy, and I don't know how to prove this in a straight forward way. The exact field of math is called Calculus of Variations. One of the results from it is that if you take a random path and slightly vary it, the length of the path will vary slightly, unless this path happens to be an extreme path. For light, in practice, it's always going to be the shortest path. When you take the shortest part between two points and make a slight deviation from it, the change in length is infinitesimal. What that means is that near the shortest path, there are infinitely many of tiny variations that all have the same phase. And that means that light traveling along the shortest path between any two points will not be canceled out. This is usually phrased differently: light always takes the shortest path between two points. The more general application of this principle to particle physics is known as Path Integral Formulation. In a more general representation of particles, instead of length of path the quantity we are interested in is called action, and if you ever wondered why Principle of Least Action exists, this is the root cause. A classical trajectory of a particle is a path along which action is stationary. And it's the reason why we have to consider particles as taking every possible path in space and time in QFT, and yet we arrive with a system where, when we take measurements, nothing goes at FTL speeds. There are a whole lot of caveats here, and I'm oversimplifying all over the place, but hopefully, it gives you some idea of why we can have QFT and GR and they say seemingly contradictory things, while being fundamentally derived from the same principles. QFT is the field theory over some fixed space-time structure, and GR is "averaged out" mean field theory of that space-time structure itself. Does that mean that General Relativity will fall short in predicting certain phenomena? Yes, absolutely! GR fails miserably when you try to describe what happens to particles as they cross event horizon, for example. We have no formalism that can handle that exact case. And it's also why I'm cautious about claims regarding time travel and FTL. There's just enough gray area there to leave uncertainty. But there aren't any inherent problems with either that we know of yet.

Would we? In Quantum Field Theory, when we are evaluating particle interactions and have to add up al possible exchanges, we include FTL ones to get the right answers. We also include ones going back in time, and simply count them as antiparticles. If microscopic causality violations are simply part of particle dynamics and macroscopic ones require astrnomic events, but are localized phenomena even then, how are we suposed to detect it?

If GR is broken, so is Standard Model. Which means absolutely ALL of physics. That's not the weak link here. And conditions for causality violation in GR are extraordinary. We had many decades between black holes being a purely math prediction of GR to actually discovering them, and in that time, you could have made the same argument you're making now that they're just a math artifact. Conditions for causality violations would have to at a minimum involve collisions of such objects, and we've only been able to detect these for a few years. Give it time.

Global causality is overrated and not strictly necessary in field theory. Ability to build FTL drive absolutely implies ability to build a time machine, but there is nothing fundamentally wrong with the later. Stating anything with certainty would require better understanding of quantum gravity than we possess, but taken individually, General Relativity allows time travel and Quantum Mechanics has mechanisms for resolving time travel paradoxes. I can't guarantee that something doesn't break spectacularly when you put the two together, but given that existing models don't have a problem with time travel, insisting that it's "physics-breaking" is at very least completely unfounded. Alcubierre metric is not it, however; that much I can agree on. But it was only ever meant as a simple toy model, and expecting it to be a practical drive is no more reasonable than trying to build a gasoline engine from schematic representation of one in thermodynamics textbook.

I always enjoy watching conspiracy nuts try to explain Skylab. I don't think we ever had anything besides Saturn V that could have put it in orbit. I'm actually pretty sure winglets are an even greater boost to efficiency in takeoff and landing than in cruise, but yeah, since fuel is a much smaller fraction of operating cost on short hops, it probably doesn't make as much difference. It's also worth noting that while some types of winglets can be added in retrofit, some of the fancier ones, with these swooping curves, are both a more recent invention and are much harder to retrofit a plane with, so they only show up on new planes.