Slashdot Mirror
Physicists Theorize Out How To Retrieve Information From a Black Hole (sciencemag.org)
sciencehabit writes: Black holes earn their name because their gravity is so strong not even light can escape from them. Oddly, though, physicists have come up with a bit of theoretical sleight of hand to retrieve a speck of information that's been dropped into a black hole. The calculation touches on one of the biggest mysteries in physics: how all of the information trapped in a black hole leaks out as the black hole 'evaporates.' Many theorists think that must happen, but they don't know how.
Ask a politician or CEO or salesman. They routinely pull information out of a "black hole".
Don't waste your vote! Vote for whoever you want, unless you live in a swing state it won't matter anyways
The radiation coming out of black hole will have lots of particles, normalising any information into practical randomness, hence increasing the entropy of universe. it'd take a hell of an effort to find out which photon will carry the information of electron.
for data compression.
Sheesh, evil *and* a jerk. -- Jade
Why do they expect information to be conserved in the first place? Information loss is common and you don't even need black holes. A simple example is matter-antimatter collision which turns into two photons, so you lose information about the identify of the original particles.
No information is lost in that scenario.
Ignorant amateur here. ISTM that if a virtual pair appears straddling the event horizion, the one that gets away never was inside the black hole to begin with, and thus would not carry away any matter or energy. Isn't the black hole just working as an engine to extract matter/energy from the vacuum near the event horizon? Half of which goes in, making the BH bigger, and half of which escapes to the external universe.
In the unlikely event that that conception is correct, it would be interesting to think about what happens to the vacuum near the event horizion. Does it get depleted of its vacuum energy, or is it an infinite source? If depleted, does vacuum energy flow in from other nearby vacuum to replenish it?
Is the vacuum inside a black hole anything like the external vacuum?
Sheesh, evil *and* a jerk. -- Jade
just go to blackhole.com
Table-ized A.I.
It's not like there is a lot of experimental evidence here, one way or the other.
Physicists Theorize Out
Did they theorize the shit out of this thing?
systemd is Roko's Basilisk.
Why don't they use water-boarding? It works so well, as we all know.
/dev/null earns its name because the device is so small nothing can be read from it. Oddly, though, physicists have come up with a bit of theoretical sleight of hand to retrieve a speck of information that's been sent to /dev/null. The calculation touches on one of the biggest mysteries in physics: how all of the information written to /dev/null hole leaks out as heat from the CPU and gets 'dispersed' by the heat sink. Many theorists think that happens, but they don't know how to put humpty dumpty together again.
dd if=/dev/blackhole of=/tmp/data ?
Call me when there is an experiment to back it up. Otherwise it is just speculation.
putting the 'B' in LGBTQ+
fox news emits no information.
Some drink at the fountain of knowledge. Others just gargle.
how you'd retrieve the velocity, direction and spin state of the original particles
This is irrelevant to the physical information content of the system. Physical information is not about how many numbers you can or can't know for every given time. It is basically just the log of the number of possible states the system is in. If you measured everything possible of a given system, it would have no physical information because it would be in a single, specific state.
Try looking up one of many articles that actually explains what physical information is before making assumptions about the word meaning the same thing as the vague everyday use. Good Such articles typically make use of examples, like variations on Maxwell's demons (Physics Today had a decent article a couple months ago).
One such example can be quickly and crudely summed up by looking at a box with a particle bouncing around in it, where the particle can be in the left half or right half of the box at any given time. If the box was perfect, and had two opposing walls that could act like ideal pistons, then knowing what side the particle was in would let you compress the opposite side's piston nu-impeded, and then wait for the particle to push back. You can convert information the operator has (a restriction on the number of states) into energy. Similarly, although with a little bit more effort, you can show that getting that information out of the system to the operator, the restriction of the states of the system from two possible to one, costs energy.
It doesn't matter that the exact values of the original particles can't be measured, because afterwards you are left with two particles with the same total momentum, angular momentum, and energy, and they have numerous states they could be in. In a finite, closed system, you can even potentially have the two photons interact again to recreate the two new particles again. The interaction doesn't restrict the number of accessible states. However, a one way process like, the possibly naive, interpretation of how an event horizon works like a one way surface, does cause the number of possible states in a closed system to decrease.
Of the information loss paradox and various fixes. The "holographic universe" is another fix. That is copy of information inside the BH exists imprinted on the event horizon.
It doesn't seem to match the definition of information that programmers use routinely in their work. I'm fairly certain that they is a mathematical identity between the two uses of the word, as they both tie back to Shannon, but the use seems to have developed extremely differently.
To a programmer every feature that is used to describe an object represents a certain number of bits of information. Clearly it is being asserted that physicists use a very different meaning. It sounds is if it's something like "the log of the number of bits required to fully describe what is knowable about the state". This may actually be closer to Shannon's original meaning, as he was concerned about the amount of information that could be transmitted through a channel of a given bandwidth in a given amount of time, but that's a bit removed from the standard meaning used in computer science, programming, etc.
I think we've pushed this "anyone can grow up to be president" thing too far.
The connection between statistical mechanics and information theory was made by von Neumann right when talking with Shannon about his now famous work. If you want to complain the defition is bad, you can start by complaining about their work. Or you can be like Viol8 and equivocate in response to every article about physical theory (it has happened at least several times before) and pretend that refusing to learn about a topic is some form of insight.
I'm not saying the definition is bas, as it clearly isn't. In either field. I'm saying the two fields apparently use the term quite differently. This doesn't make either wrong. And I think that both would agree on the basics, e.g. that a bit is a unit of information.
P.S.: I tend to think of this as an oversimplification, but it's one we've built all our digital hardware around. The problem is it seems to make any relatively prime chunk of information require an infinite number of bits to express accurately. Try, e.g., to accurately express 1/3 except as a binary number. But practically the higher primes rarely occur, and usually an approximate value is good enough that we don't worry about the finer details. And when we do, there are ways to refer to it...e.g. ratios between two integers. Of course, that doesn't work for the irrational numbers... So I don't think we have an exact theory of information, but only a good approximation. But it's quite a good approximation.
P.P.S.: Perhaps a good theory of information would easily solve the three body problem. Currently we rely on infinite approximations, that we necessarily cut off at some point. Now chaos theory implies that even knowing the exact solution wouldn't help us, because we couldn't specify the initial conditions exactly enough, so that might not be a real benefit, but it would be a good theoretical benefit. There are a few other cases where the two body case is easy (well, relatively easy) and the three body (or five body, or...) can only be approximated. I have a suspicion that in many of these cases it's because somewhere in the fundamental assumptions there is something where a rule is used that works well for pairs of items, but not for relatively prime groupings. (Of course, here *I'm* assuming that a case with 4 elements can be handled as a pair of pairs, and this isn't always true.)
I think we've pushed this "anyone can grow up to be president" thing too far.
Your password can have a lot of bits of data but very little information. You notice where they say your password is weak? The complaint is about entrophy. And if you are cool you can measure that and then say something like "my password has 256 bits of data and 3 bits of entrophy."
The Akashic record has to be somewhere!
While correct, a password usually only has a very little information it is because of the larger embedded context. I.e., if your language is English you find it more common for certain letters to follow other letters, and for letters to follow letters, etc. So the intrinsic information is the number of bits used to express it, but the information in context is the number of bits required to represent it using an optimal compression algorithm using all the contextual information that you know. Both are valid measures of information.
I think we've pushed this "anyone can grow up to be president" thing too far.
Or "guess". But not "theorize". Ah, the pain.