My understanding of RobotRollCall's explanation is that nothing falls past the event horizon, once it has formed. An observer falls toward the horizon. Gravity distorts time in a dramatic fashion. Outside observers see the falling observer in normal time, appearing to fall through the horizon. The falling observer experiences normal time and scatters off the event horizon. After scattering, the falling observer finds that trillions of years have passed in the "outside world", due to the extreme time distortion. No information is lost, it is scattered eventually, albeit in normal time as seen by the falling observer. I can find his original posts if they are of interest. A similar, but seemingly newer, view is expressed in the article Henry posted: "Stephen Hawking proposed a potential solution earlier this year. His idea is that gravitational collapse can never continue beyond the so-called event horizon of a black hole beyond which information is lost. Gravitational collapse would approach the boundary but never go beyond it."" On Thu, Jul 24, 2014 at 6:17 PM, Eugene Salamin via math-fun <math-fun@mailman.xmission.com> wrote:
In this hypothesis, what happens to an observer who falls into a black hole? Standard general relativity says the observer continues to exist after entering the event horizon.
-- Gene
________________________________ From: Jeff Caldwell <jeffrey.d.caldwell@gmail.com> To: math-fun <math-fun@mailman.xmission.com> Sent: Thursday, July 24, 2014 2:48 PM Subject: Re: [math-fun] Black Holes Aren’t Black After All?
Very interesting, and I hadn't yet seen that result.
Around three years ago, a Reddit user, RobotRollCall, posted many excellent answers to /r/AskScience. He (or she) said this about black holes:
"...as long as there's no flux through the boundary, a volume of space can be fully described by treating it as just a surface with no interior.
You can then take the next step and say that all of the information — which has a specific meaning here, basically referring to everything that's conserved, like energy-momentum and charge and baryon number and so forth — exists not on the interior of the volume, but rather on its surface. The volume of space, taken as a whole and viewed from the outside, acts like it's just a surface, with conserved quantities "painted" on its surface. ...
A black hole is, in fact, just such a volume. It has a well-defined surface: the event horizon. It's spherically symmetric. Its interior is uniform. And there is no flux whatsoever through its boundary, because of exactly the reason you noted.
So it turns out that quirky mathematical abstraction really can be applied to black holes in a way that isn't just approximate. And in fact, when you work through it, it turns out to have profound physical significance.
You can think of black-hole formation this way: At the instant the event horizon forms — and it really is an instant; it happens all at once, over zero time in all reference frames — the interior of the black hole completely ceases to exist, and is replaced by a pure surface, which contains within it all the information — all the conserved quantities, in other words — that were present in the spherical volume of space the boundary of which became the event horizon.
From that point on, there's no place in your equations to put any of the stuff that went into making the black hole. It literally doesn't exist any more. All of its mass, all of its momentum, all of its charge, all of its entropy is now the substance of the black hole's event horizon. Forever after, only the event horizon itself interacts with the universe, and that interaction can be described by a unitary S-matrix. Everything that should be conserved is conserved, no magic happens, nothing vanishes from the universe or appears out of nothing. For the time that the black hole exists — which may or may not be finite, depending on what the scale factor of the universe does in the future — everything that happens around it is determined exclusively by the properties of the event horizon itself.
Black holes, in other words, have no insides. They're surfaces with no interiors."
and
"Because our understanding of black holes has evolved a lot over the past twenty years or so. Basically everything that was believed to be true about them last century turns out either to be false, or to be much more nuanced and interesting than anyone ever suspected.
The best way to think about black holes (and I keep revising this little speech as I test it out on undergraduates) is not so much as objects so much as processes.
The interaction of matter and energy with a black hole can be described, mathematically, as a scattering process. Absolutely no different than the type of scattering that happens when a photon meets an atomic nucleus. Stuff goes into the interaction, and stuff comes out of the interaction, and what predicts the outs from the ins is something called a scattering matrix.
What makes black holes special, though, is the extent to which time is distorted around them. From the point of view of a distant observer — that's you and me — time near the event horizon is dilated to such an extent that it very nearly stops altogether. Because of this, the scattering process that happens around a black hole takes, instead of an amount of time so short as to basically be zero, countless trillions of years.
So what's actually happening to the stuff that "falls in" to a black hole is that it's scattering. Right now, it's scattering. But time is so dilated around the black hole that in our frame of reference the process takes, almost literally, forever.
So long story short, everything that "falls in" to a black hole scatters off of the event horizon. But no one will be alive to see it."
It's interesting that those ideas appear to have stuck, evolving into the results of the paper Henry quoted.
Jeff
On Thu, Jul 24, 2014 at 3:00 PM, Henry Baker <hbaker1@pipeline.com> wrote:
...
Information is ultimately carried by quantum particles such as photons. So at issue here is the nature of quantum mechanics. A fundamental postulate of quantum theory is that all the information about a system is encoded in its wave function and this always evolves in a way that conserves quantum information.
But the quantum information associated with matter that enters a black hole seems to devolve into a single state. When that happens, the information must be lost.
This information paradox has triggered some intense soul-searching. And as a result of this, Stephen Hawking proposed a potential solution earlier this year. His idea is that gravitational collapse can never continue beyond the so-called event horizon of a black hole beyond which information is lost. Gravitational collapse would approach the boundary but never go beyond it.
Today, Cenalo Vaz at the University of Cincinnati takes this idea and runs with it. The question he tackles is this: if a dead star does not collapse beyond the event horizon and so does not produce a black hole, what does it end up like?
To find out, he models the quantum behaviour of dust as it collapses towards the event horizon. “In this simple model, the picture that emerges is very different from the traditional view of a black hole,” says Vaz.
In this new model, there are two important parameters. The first is the so-called Schwarzschild radius, which is the size of the black hole that would form in a conventional model. The second is the radius of the dust ball as measured by distant observer.
The difference between these radii is a region in which the universe is unlike anything physicists have imagined. The total energy here is negative and quantum fluctuations are likely to dominate. But exactly what goes on will depend on the exact model of quantum gravity that turns out to be correct, something that physicists are some way from determining.
_______________________________________________ math-fun mailing list math-fun@mailman.xmission.com https://mailman.xmission.com/cgi-bin/mailman/listinfo/math-fun
_______________________________________________ math-fun mailing list math-fun@mailman.xmission.com https://mailman.xmission.com/cgi-bin/mailman/listinfo/math-fun