Live data from Hacker News

A physicist argues that information is at the root of everything

economist.com

21–30 of 40 posts

Re: A physicist argues that information is at the root of everything

#21
Okay, I'm just going to come out and say it, since it comes to mind every time I see an article like this. Let me know if I'm off track.

Isn't it more accurate to say that "description," rather than "information," is the root of everything? Information itself is pretty abstract, so it seems far more useful to do it this way.

Re: A physicist argues that information is at the root of everything

#22

Earlier quoted context omitted.

"It's a book review -- doesn't really explain what the physicist is arguing in any meaningful sense." Indeed, the review ends with this: "Mr Vedral makes a persuasive argument for a third option: information can be created out of nothing." That's a remarkable claim, and the review says zip about how it is presented. It seems that it would also be the key concept of the book, making this a pretty lame review.

Well it's not really remarkable at all, the argument is basically that information and entropy are mathematically equivalent and so information like entropy spontaneously increases.

Information and entropy are absolutely not equivalent; entropy describes the upper bound on information content, but says little about the actual information contained in a coding. The statement that information is spontaeously created isn't based in any physics I know of.

Re: A physicist argues that information is at the root of everything

#23
post #21

Okay, I'm just going to come out and say it, since it comes to mind every time I see an article like this. Let me know if I'm off track. Isn't it more accurate to say that "description," rather than "information," is the root of everything? Information itself is pretty abstract, so it seems far more useful to do it this way.

They're using information in a technical sense, in the sense of "information theory" and maybe "Kolmogorov complexity".

Re: A physicist argues that information is at the root of everything

#24

This article seems to imply that the author equates Shannon entropy with information. Is that so? It has always struck me that Shannon entropy is a good measure of information, but is not a good definition of information itself. Like the difference between a vector and the magnitude of a vector, information and its measure are not the same. What is information itself? If this author actually has a definition of that,…

How about http://en.wikipedia.org/wiki/Kolmogorov_complexity ?

The big problem with Shannon entropy is that it's measured with respect to a probability space, so it's not an "absolute" measure of information. Kolmogorov complexity "solves" this problem by defining information content of a string as the length of the shortest program that generates the string. (You still have to pick a reference UTM, but this only affects the program lengths at worst by a constant.)

[Disclaimer: Hobbyist, not expert.]

Re: A physicist argues that information is at the root of everything

#25

Mistaking the deepest abstraction you are capable of for the fundament of all reality is hubris. Philosophy is one of those subjects that most intelligent persons think they can master pretty well. It all seems pretty simple and straightforward. Especially physicists fall into that trap again and again, up to great string theorists putting forward anthropocentric theories of the universe. These people usually make tr…

"He defines information based on the frequency of 'events'. This means that both time and 'events' are more primal than his notion of 'information'."

True, but what do you expect for a popular review of a book that itself is a popular review of some pretty abstract ideas.

The basic concept is not just that our universe can be described by statistical measures, but that the foundational laws seem to be governed by statistics. At the small scale (individual particles), quantum mechanics basically says that the universe runs itself by rolling dice in very particular ways. Events involving these particles seem to simply be the tangible embodiment of Shannon entropy. Even at the everyday scale of refrigerators and gasoline engines, entropy calculations are the bread and butter of thermodynamics engineers.

It is a major open question in physics whether entropy and statistics are truly foundational, or whether they are simply the large scale average result of a fine-grained deterministic foundation. The "Copenhagen interpretation" takes statistics to be fundamental, and by gosh the calculations sure work nice, but there is no evidence that it reflects a higher truth.

Re. the anthropic principle in physics, it is one approach to answering why certain laws of physics seem to be balanced on a knife edge between chaos on one side and stagnation on the other. If we posit that universes can be created with nearly arbitrary physical laws, then naturally we find ourselves in one that is neither too chaotic nor too stagnant for our existence. The creation of such a universe might be unlikely, but given enough opportunity it is inevitable. This allows for bizarre physical laws without requiring magic or divine intervention. (There are even evolutionary approaches, where events like black hole formation create a new universe similar to its parent. That way if you find yourself in a part of the multiverse conducive to your kind of life, neighboring universes are likely to be very rich in friendly conditions.)

"And how could it be, if we need minds and language to discuss 'information' in the first place?"

If the evolutionary approach were true, universes of pure thought could potentially be created, where the very atoms were symbols grounded in some reality. A mind there wouldn't pick up a pen to draw a symbol, it would just think it into existence. Or such a parent universe could create a universe like ours, to test whether symbol processing is a necessary law of physics, or if it can spontaneously arise from semi-chaotic fundamental laws.

Re: A physicist argues that information is at the root of everything

#26
post #6

From the most useful Amazon.com review of the book ( http://www.amazon.com/review/R15E7HO19A2FQF/ref=cm_cr_rdp_pe... ): " The core of the argument (on page 201) I will not spoil here, but it's not wholly convincing on several grounds. First, although he does not further dwell on it, the author favors the 'Copenhagen' (or observer effect) interpretation of quantum physics, which is used to buttress & underpin his argu…

Whether a physicist (in modern times and below a certain age) espouses MWI is quickly becoming a great litmus test for whether you should bother listening to them on anything that isn't established textbook material. Copenhagen is pseudo-scientific nonsense, and I'm amazed anyone can work on developing quantum computers while simultaneously denying the existence of what they're designing. I'd cut him some slack if this were just some tangential side point, but if it's critical to his argument the whole of it isn't worth anyone's time.

Re: A physicist argues that information is at the root of everything

#27

" Unusually for a physicist, Mr Vedral spends a fair bit of time talking about religious views, such as how God created the universe. He asks whether something can come out of nothing . . . Mr Vedral makes a persuasive argument for a third option: information can be created out of nothing. " Somehow this reminded be about this verse in the Bible: "In the beginning was the Word, and the Word was with God, and the Word…

.. and the heavenly spirits decided to set the Word free with a GNU license, and released Open Office, and after fixing most of the bugs God saw that it was good (or at least usable).

Re: A physicist argues that information is at the root of everything

#28
post #6

From the most useful Amazon.com review of the book ( http://www.amazon.com/review/R15E7HO19A2FQF/ref=cm_cr_rdp_pe... ): " The core of the argument (on page 201) I will not spoil here, but it's not wholly convincing on several grounds. First, although he does not further dwell on it, the author favors the 'Copenhagen' (or observer effect) interpretation of quantum physics, which is used to buttress & underpin his argu…

Whether a physicist (in modern times and below a certain age) espouses MWI is quickly becoming a great litmus test for whether you should bother listening to them on anything that isn't established textbook material. Copenhagen is pseudo-scientific nonsense, and I'm amazed anyone can work on developing quantum computers while simultaneously denying the existence of what they're designing. I'd cut him some slack if th…

"Whether a physicist (in modern times and below a certain age) espouses MWI is quickly becoming a great litmus test for whether you should bother listening to them on anything that isn't established textbook material."

Errr, do you mean you should bother listening or not bother?

My understanding of MWI doesn't go much beyond what I could understand of an early Wheeler paper (I thought it was from the mid-40s, but Wikipedia says MWI starts with the thesis of a student of his with the first publication in 1957 ... and I read that paper (whenever it was published) 25 years ago so it's gotten pretty fuzzy).

Hmmm, Wikipedia says MWI "along with the other decoherence interpretations and the Copenhagen interpretation" are the mainstream interpretations, so I suspect that answers my question.

Re: A physicist argues that information is at the root of everything

#29
post #22

Earlier quoted context omitted.

Well it's not really remarkable at all, the argument is basically that information and entropy are mathematically equivalent and so information like entropy spontaneously increases.

Information and entropy are absolutely not equivalent; entropy describes the upper bound on information content, but says little about the actual information contained in a coding. The statement that information is spontaeously created isn't based in any physics I know of.

How does the (conventional) big bang theory fit into that? At first glance it seems like valid evidence for spontaneous creation of information.

Re: A physicist argues that information is at the root of everything

#30
post #6

From the most useful Amazon.com review of the book ( http://www.amazon.com/review/R15E7HO19A2FQF/ref=cm_cr_rdp_pe... ): " The core of the argument (on page 201) I will not spoil here, but it's not wholly convincing on several grounds. First, although he does not further dwell on it, the author favors the 'Copenhagen' (or observer effect) interpretation of quantum physics, which is used to buttress & underpin his argu…

Whether a physicist (in modern times and below a certain age) espouses MWI is quickly becoming a great litmus test for whether you should bother listening to them on anything that isn't established textbook material. Copenhagen is pseudo-scientific nonsense, and I'm amazed anyone can work on developing quantum computers while simultaneously denying the existence of what they're designing. I'd cut him some slack if th…

The reason people can get away with Copenhagen is that decoherence is absurdly effective in all realistic cases, causing there to very little uncertainty about whether and when a measurement happened. Insofar as we can all agree when a measurement happened, physicists operating under Copenhagen are going to be just as effective as those who understand MWI. And for most of the theoretical and especially computational aspects of quantum computers, this is the case. The only place where someone operating under the Copenhagen interpretation is going to falter is when they have to think hard and in novel ways about how to prevent decoherence.

So I really don't think interpretation is going to be correlated with advances in the computational aspects (i.e., what kind of problems are quantum computers good for once they are built, etc.) but I would predict that future insights in how to build computers which are robust against decoherence will be dominates by those who understand MWI.

Post reply on HN