Tuesday, May 27, 2014 ... Français/Deutsch/Español/Česky/Japanese/Related posts from blogosphere

Constructor theory: Deutsch and Marletto are just vacuously bullšiting

I can't stand pompous fools, people who are completely dumb but who like to pretend how wonderfully smart they are. So it is not hard for you to guess that I was rather upset when I was forced to read a new preprint by David Deutsch and Chiara Marletto and the associated hype in Scientific American:

Constructor Theory of Information (arXiv)

A New “Theory of Everything”: Reality Emerges from Cosmic Copyright Law (SciAm)
If you happen to forget similar factoids, Deutsch is one of the philosophical babblers who likes to say ludicrous things about the allegedly unavoidable naive many-worlds interpretations of quantum mechanics, and so on. What is this constructor theory? It's a sequence of worthless would-be smart sentences sold as a "theory of everything" and a "unifying theory of classical and quantum physics" and "all information in them" which also "defines all forms of information" and transforms all of our knowledge to "claims that some tasks are impossible".



"Where is the beef?" the ladies would surely ask in this case, too. If you try to find any content inside these texts, you will inevitably fail. There is no content. It's just a stupid game with words and a couple of mathematical symbols.




For example, we're told to change our wording used to describe the energy conservation law. Ms Marletto tells us about her wonderful new idea that clarifies what the conservation law is:
“You simply say that the task of creating energy from nothing is impossible.”
How simple, deep, and useful. More seriously, first of all, this statement doesn't accurately express the conservation law – which also says that destroying the energy into nothing is impossible. Second of all, the contrived language above tells us nothing about the validity of the energy conservation law. Third of all, the law is exactly valid whenever the background or the effective laws of physics are time-translational symmetric; however, the law is violated in cosmology which helpfully allowed cosmic inflation to create the whole visible Universe out of (almost) nothing.




At any rate, their musings have nothing to do with the actual technical questions or the truth value of the energy conservation law – or any other scientific question, for that matter.

Their "work" is all about imposing a meaninglessly contrived language upon people. Moreover, Ms Marletto's understanding of energy – dominated by the question whether it can be obtained for free – is exactly the same childish misunderstanding of the concept that Feynman would humiliate in his text Judging Books By Their Covers. "Energy makes it go" was written as an answer to all questions in the book. Feynman pointed out that the children wouldn't learn a damn thing. They could have written "Wakalixes make it go" just as well. Moreover, the statements aren't really true because energy isn't what makes things go. It's just being converted from one form to another etc.

Some of the most experienced readers already know that
a kettle may heat water.
Fortunately, the authors allow us to formulate even such statements in a more "natural" and more "profound" way:
For instance, a kettle with a power supply can serve as a constructor that can perform the task of heating water.
LOL, it's a constructor. Who would have thought? One would think that a kettle isn't constructing anything – in the same sense as cranes do. But if you want to become so high that you confuse construction workers with cooks, it may be a great idea to unify cranes with kettles. And raindrops on roses and whiskers on kittens. Bright copper kettles and warm woolen mittens. Brown paper packages tied up with strings. These are a few of my favorite things. Engineers and cooks are the same thing, after all. They are constructors.

OK, what is a constructor? You study the paper and you find out that it is a meaningless word that may represent anything and that may be inserted into any sentence for pompous fools to look even "smarter". Instead of saying that X does Y, you must say that X is a constructor that is prohibited to do things different than Y. It isn't really quite the same thing and the whole expansion of the sentence is bringing you nothing of value but you don't care. I can't resist to compare these awkward sentences to yet another quote by Feynman who was reading some "smart" texts by participants of an interdisciplinary conference.
The individual member of the social community often receives his information via visual, symbolic channels.
He didn't know what the first sentence of a text could have possibly meant. He must have been stupid! However, he didn't give up and finally figured out what the sentence meant.
People read.
There was a longer paragraph which could have been translated as
Sometimes people read, sometimes they listen to the radio.
And so on. The Deutsch-Marletto texts are made uselessly contrived in the very same sense as the sentence involving visual, symbolic channels. There is no point of talking about a "constructor". This redundant concept doesn't help your understanding of anything in Nature or mathematics.

But equally importantly, the comments they are making about things like quantum information are just totally wrong.

They want to "unify classical physics and quantum mechanics" and the concepts of information in them. But this is complete nonsense. The frameworks can't be unified because they contradict one another. More clearly, quantum mechanics is right in Nature around us while classical physics is wrong. Classical physics is a limit of quantum mechanics – classical physics is approximately valid in some extreme enough situations accurately described by quantum mechanics – but not the other way around. One should never try to unify a thing that is right with a thing that is wrong. It's like pouring gasoline on the fire – and I won't even tell you which of these two elements is right and which of them is wrong.

Moreover, the information has always meant the same thing "in" classical physics or quantum mechanics. Information is independent. The concept of information exists independently of particular theories in physics. However, particular theories in physics differ in what they say about the form that information can take and how the information about the past that we already know is linked to the information about the future that we may only predict.

We may have some information about the spin of the electron a second ago and predict some measurement of the spin we perform the next second. This is true both in classical physics and quantum mechanics. The two frameworks radically differ by the "spectrum" of possible values that the spin can take (and similarly other observables). And they also fundamentally differ in the way how we make predictions about the future measurements – about the information that the measurements will yield (it's the very same thing) – from the information that the past experiments have already yielded. The laws of quantum mechanics are probabilistic and involve all the linear Hermitian operators acting on a Hilbert space (or density matrices which are also operators on the Hilbert space).

The engine that connects some pieces of information (e.g. ones about the initial state) with some others (e.g. those about the final state) is completely different in classical theories and in quantum theories. But the information is ultimately the very same thing. We make a measurement and we learn some information from that procedure. Information is whatever distinguishes one's state of ignorance (or knowing less) from the state of knowledge (or knowing more). In general classes of situations, information may be quantified, too. But there's no point in trying to decompose the words "information" or "know" to words that are more elementary. There are no words that would be more elementary. One must understand at least some words if she wants to be able to say anything about anything. Attempts to "define" words like "know" inevitably end as a ludicrous worse-than-circular game that "decomposes" simple enough words into words that are much more complex, contrived, or subtle – or several of these adjectives at the same moment.

Claude Shannon developed some theory of classical information based on logarithms that is useful especially whenever one is quantifying how many bits may be transmitted through one communication channel or another. Ask Shannon in the comment section if you want to know details. ;-) Do we have corresponding things for a Universe governed by a quantum mechanical theory? You bet. After all, our world is a quantum world so if Shannon's classical theory of information works, we must have an explanation for that. And we have one, of course. Shannon's information is linked to the entropy – and this logarithm-based formula has a completely natural quantum counterpart, the so-called von Neumann entropy.\[

S = -{\rm tr}(\rho \ln\rho)

\] Here, \(\rho\) is the density matrix. In a basis where \(\rho\) is diagonal, the formula directly reduces to the classical one. In this sense, the new aspect of information in quantum mechanics is that the state vector (or density matrix) may be sliced in many inequivalent ways – by measuring one set of mutually commuting observables; or another set of mutually commuting observables that however refuse to commute with those in the first set.

But once we measure something in a quantum world, you obtain the same "classical" information that you obtain by a measurement in a classical world that happens to have the same number of potential outcomes (the same spectrum). You may know something or not to know something; you may extract \(n\) bits from the measurements or another number. You may use more natural units of information, nats. One may also measure continuous observables. Such measurements naively yield an infinite amount of information (because there were infinitely many potential outcomes a priori) but due to some unavoidable errors, the effective information is actually finite. All these things are true in classical physics as well as quantum mechanics.

We have this extra term "qubit" or "quantum bit" to describe a unit of information in quantum mechanics. Does it prove that the information means something completely different in quantum physics? Well, yes and no. To have the information \(n\) qubits means to describe the physical system as a vector in a \(2^n\)-dimensional complex linear Hilbert space \(\HH\). Or a density matrix in the corresponding space of density matrices – pretty much in \(\HH\otimes \HH\). But you must understand that this state vector isn't classical information yet. It's just an auxiliary tool to predict information – to predict information that you may call classical.

At the end, you measure some observables such as the spin \(s_z\) or anything else and the state vector tells you the probabilities of individual outcomes. Once you get the outcome, like \(s_z=-1/2\), you have the same information – you may call it classical information – that you have in classical physics. There is no difference. What is completely different is "just" the way how physics predicts what will (or may) actually happen.

Before a measurement, the state vector or the density matrix describe the state of your subjective knowledge about the physical system. Whether you call it "information" or not is up to you. But whatever you do, the meaning of the density matrix in particular is the same as the meaning of its classical counterpart, namely the probability distribution on the phase space. It's straightforward to see that the probability distribution on the classical phase space is what one may get out of the density matrix in the classical, \(\hbar\to 0\) limit, assuming that we plan to measure the observables that are coordinates on the phase space. (In the classical limit, it doesn't matter much what we plan to measure because the commutators of the observables we care about go to zero: they are proportional to \(\hbar\).)

Would you describe the probability distribution on a classical phase space by the word "information"? Well, it's up to you. I would say that the distribution encodes one's subjective knowledge about a physical system that is only useful for predicting the actual information – the results of some measurements. I would personally only use the word "information" for something that changes your subjective knowledge – either measurements or someone's telephone call etc. about something he or she has measured (i.e. found out). One may also talk about the "amount of information" and there are formulae that allow us to quantify this concept as a function of the probability distribution in classical physics; or as a function of the density matrix in quantum mechanics. The essence of the formulae is the same in classical physics and quantum mechanics. It is no coincidence. The information is the same thing, something that changes our knowledge about something. Physical theories and frameworks differ in the "a priori spectrum" of the options that may be distinguished by the information; and in the actual methods to calculate some information about physical objects from other information (and they differ in the results of these methods, of course, i.e. in the actual predictions).

There is really nothing else that can be said about these matters and that is both true and fundamental. These concepts require one to think a bit and to be smart enough but all these conceptual insights are compact and once you learn these principles, you know them. The amount of stuff one may learn about the meaning of information in physics is finite. It is limited. Once you learn it, and you should learn it before you get your undergraduate degree, you know everything about it. Extra babbling about the information is guaranteed to be meaningless or vacuous. Or untrue. Or some combination of these adjectives.

Deutsch and Marletto also write kilotons of bullšit about the meaning of entanglement (they want to say that it's not mysterious but their babbling is only meaningful if you are ready to think that it is mysterious, so the goals of their "explanations" are internally logically inconsistent) and the "reasons" why quantum mechanics prohibits cloning (founders of quantum mechanics surely knew that the evolution in quantum mechanics is given by a linear evolution operator and because a copying is a tensor square, a quadratic operation, it clearly cannot result from the linear evolution). Why would they write lots of confusing šit about these matters that are in principle trivial?

So we're also exposed to the standard anti-quantum propaganda that entanglement is a borderline crime and it shouldn't happen unless – and the rest of the sentence is their "contribution" – you have some powerful constructors inside kettles to heat water and to beat other constructors, or whatever. It's just batšit crazy.



A Deutsch-Marletto constructor.

There is nothing mysterious about entanglement. Entaglement is just the most general description of correlations between two subsystems of a physical system – roughly speaking, between two physical objects – that is allowed by the laws of quantum mechanics. Conceptually, the "mutual relationships" between the two objects that are summarized by entanglement is the "same thing" as the usual correlation in classical physics. However, the degree of correlation or the number of ways in which it may get manifested is different in quantum mechanics than it is in classical physics simply because they're two inequivalent frameworks. Every quantum mechanical theory is inequivalent to every classical theory. So of course that they should be expected to give different predictions about the degree of correlation between two objects or the number of ways in which correlations may get manifested. But the interpretation of the result is exactly the same. Entanglement implies that the probability distribution for pairs of quantities measured on the systems A,B isn't the simple product of two distributions,\[

\rho_{AB}(\vec x_A, \vec x_B) \neq \rho_A (\vec x_A) \cdot \rho_B(\vec x_B).

\] Whenever this inequality holds, there is some correlation. Entanglement means that there is a correlation, using the language of quantum mechanics. An entangled state in combination with the laws of quantum mechanics produces or predicts correlations in various quantities that couldn't be extracted from any classical model. But that shouldn't be shocking. Classical physics is a fundamentally different kind of a theory so it gives fundamentally different predictions than quantum mechanics does. So there is absolutely no reason to expect that their predictions should agree with the other side, that one framework should be easily "emulated" by the other. Quantum mechanics cannot be emulated by classical physics; classical physics may be just approximately emulated by quantum mechanics in a classical limit.

Their wording – e.g. the constant repetition of the meme that entanglement is "spooky" – makes it very clear that they're just two more people who have decided to never understand quantum mechanics – because they have vowed in front of the other members of the Satanist Cult to never give up the efforts to look for a classical model beneath quantum mechanics. But there ain't a valid classical model. Classical physics has been falsified. It's wrong. It's dead. It's kaput. Falsification is irreversible. You can never resuscitate it again. Quantum laws are the correct ones and the linearity of evolution and the existence and basic properties of entanglement are no mysteries or contradictions within quantum mechanics. They're elementary building blocks of all quantum mechanical theories. If you have trouble to comprehend that entanglement is completely natural and doesn't need any constructors or unification with classical physics or kettles to boil water, then you are exactly in the same situation as a person who is trying to read English but can't possibly swallow E,T,A,I,N,O,S.

Just like these 7 letters are the most frequent letters in English texts, entangled states and linear operators are virtually omnipresent in quantum mechanics and anything and everything linked to it. If you haven't learned E,T,A,I,N,O,S, then you are illiterate. If you feel uncertain about the existence of entanglement or linearity of all operators, you are illiterate in quantum mechanics. If you are waiting for a kettle or for Sherlock Holmes to reduce E,T,A,I,N,O,S or entanglement or linearity of operators to something that you already know, e.g. how to make poops, then you are a baby who has a long way to go before you learn the basic things. This conclusion is addressed to Mr Deutsch and Ms Marletto, too. She is attractive and he is... sorry, I can't think of a positive adjective now..., but they are totally deluded about modern physics.

I will kindly avoid the discussion of some additional concepts they use such as the "superinformation medium". Those things are bullšit on steroids. Reality does in no way emerge from a "cosmic copyright law" and their mixing of buzzwords from classical and modern physics with the words from everyday life such as "kettles" and "copyright laws" and "Sherlock Holmes" or with sensationally sounding redundant new clichés such as "constructors" or "superinformation medium" or "subsidiary theories" or "output attributes of substrates" (no kidding) is beyond the pale. All these things are used by them not only to mask that they're complete morons but even to look as smart folks in the eyes of people who are even more hopeless morons than themselves. I just can't stand such games that are all about people's pretense.

Add to del.icio.us Digg this Add to reddit

snail feedback (74) :


reader John Archer said...

I see you're in fine form again today, Luboš. You do like to engage! Great fun! :)

Constructors?

For me the immediate association is with object-oriented programming languages such as C++.

Did they manage fit in the obverse, namely destructors?

How about greentard recycling and garbage collection? But maybe they're saving that for "the general theory". :)


reader lukelea said...

Speaking of pompous fools, that's why I can't stand Thomas Friedman.

I wonder if this guy Thomas Piketty is in the same category? Somehow the notion that r > g (the rate of return on investment exceeds the rate of growth of the economy) is the great undiscovered key to economic inequality, strikes me as ludicrous.


reader Anon said...

A good call, Lubos. Deutsch has an ego the size of a small exoplanet. He has Wolfram-itis.


reader mtc7 said...

A little off topic here. This guy is making some pretty fantastic claims: http://ptep-online.com/index_files/2014/PP-38-05.PDF

I don't have the expertise to critique that properly but my first thought is that this guy is off his rocker although I can't pinpoint exactly why.


reader Alex said...

The term is 'fine fettle'.
I wouldn't like to be on the receiving end of that.


reader Marko said...

It seems the authors are trying to play the old game of trying to make sayings as complicated as possible. For example, instead of saying "An apple a day keeps the doctor away" you can say something like "A single pyrus malus per diem restrains the arrival of the Hippocratic apostle", or instead of "Not all that glitters is gold" you can say "All articles that coruscate with resplendence are not, ipso facto, auriferous".


reader John Archer said...

Form, I says.

No, wait!

Duck, I says Duck.


reader ThinkingGuest said...

Best post ever.
LOL, thanks!


reader Doug said...

We already have a theory that incorporates both classical and quantum kinds of "information," it's statistical mechanics.


reader Honza said...

They are late. ;-) Constructors have been already well defined in Stanislaw Lem's 1965 book "The Cyberiad"
http://en.wikipedia.org/wiki/The_Cyberiad
http://www.amazon.com/The-Cyberiad-Stanislaw-Lem/dp/0156027593
Quite possibly that book is their inspiration. ;-)


reader Fred said...

Piketty is also very good at hitting women. He was arrested for punching his girlfriend who is also a socialist minister in the french government in the face. Charges were dropped of course. Also the FT claims that his data is faulty. Also he extrapolates linearly from 2012 to 2100. He could almost be a climate scientist.


reader Fred said...

Lubos the first link is wrong. It does not go to the correct paper.


reader Fred said...

I think that they are planning to submit this paper to the postmodern social studies journal Social Text . But this time unlike Sokal they are serious.


reader Fred said...

Lubos why do you delete my comments ?


reader Peter said...

I actually hope that the general public will notice soon that the `research' of philosphers at universities produces pretty much zero content and that as a consequence there will be severe budget cuts for philosophy departments.


reader Fred said...

What's going on ? Am I banned now ? Why ?


reader Svik said...

Your net connection may be cracking up.


reader B Chin said...

Some years ago, I saw David Deutsch give a lecture about constructor theory telecast from England into MIT, he was blown up on a projection screen like the Emperor in the old star wars movies. At some point the internet connection crapped out. The guy sitting next to me immediately quipped "Steven Wolfram is jamming the transmission".


reader Rehbock said...

Some of my favorite things -this post


reader Gordon said...

Hmmm, someone other than me who appreciates that book :)


reader Gordon said...

Excellent post. Yes, invent some jazzy new terms and pretend you have a new, original idea...
I remember reading Deutsch saying that the Many Worlds Theory was "proven" by the double slit experiment with one photon producing an interference pattern by saying that "shadow photons" from other worlds interfered with the photon in this universe. First I thought,
"neat idea", then, quickly my bullshit alarm went crazy :)


reader Eugene S said...

Luboš just keeps getting better as a writer, witness the way he snuck in a reference to one of my, well, favorite showtunes.
I had long been hoping for a TRF guest article by David Deutsch, followed by a spirited discussion in the comments, but it was not to be. Well, one can still hope for Dr. Deutsch to perhaps show up in the comments, though I wouldn't hold it against him if he doesn't. (Deutsch, a true romantic in the Heisenbergian strain; Motl, an incorruptible judge, jury and executioner in the footsteps of Pauli.)

By now, you have to wonder what hurdle scientists looking to publish fear most: the anonymous reviewers hired by PRL... or the very public savage beating after publication by our host.

You and your co-authors have negotiated the obstacle course, first of drafting and re-drafting, many times, your paper, crossing your t's and dotting your i's, incorporating reviewers' suggestions, finally getting that coveted letter of acceptance. Only to be met, at the real end of the course, by the basilisk stare of you-know-who.

And in his science articles, our host is almost never wrong (his writings on culture and politics are another matter).


reader Eugene S said...

Perhaps this blog's motto should change from "Our stringy universe etc." to "Pain Train Is Comin'", and escaping from the Pain Train unscathed should rank as relief similar to this: https://www.youtube.com/watch?v=tjQH7V0O6Lo


reader Luboš Motl said...

Thanks for your kind words, Gordon. And these shadow photons are exactly what's so wrong about Deutsch's MWI.


For some physicists, the many worlds interpretation may be an innocent choice of words that doesn't spoil any physics at all. They just use the words "another world among many" for one term in the wave function that has decohered from other, similar terms. So it's an alternative history that looks like a classical one where one may neglect the risk of interference with the other terms again in the future. The risk of re-interference has irreversibly dissipated (in an approximation) just like energy dissipates in friction etc.


But that's exactly the opposite of allowing the photon from "other worlds" to interfere with ours! A picture of "many worlds" where the individual worlds may literally influence each other is a downright inconsistency.


reader Luboš Motl said...

A scary video.


Tx for kidness and the idea about Deutsch the guest may be intriguing to retry.


reader Leo Vuyk said...

Qoute "A picture of "many worlds" where the individual worlds may literally influence each other is a downright inconsistency" However it is still a neat idea if you simply imagine that the universe is symmetric from the start, which means that there has to be an entangled anti matter world far away!


reader Fred said...

No. But my healthy paranoia is playing up. I submitted comments which went into moderation and then when I refreshed they were gone completely. I kept coming back for about an hour to check. This comment will probably appear in about a month.


reader Fred said...

Praise the lord (or Lubos)! My comment went straight up without moderation.


reader Luboš Motl said...

Dear Fred, I found your comments in the moderation queue in the morning and if you have had the white list membership before, you must have changed something about the way you connect. At any rate, I added your new credentials to the white list again.


reader Joseph Melia said...

David Deutsch: Centre for Quantum Computation, The Clarendon Laboratory, University of Oxford; Chiara Marletto, Materials Department, University of Oxford.



So what's this got to do with philosophy departments? (Let alone `philosphy' departments as you put it?).


But--hey--why let the facts get in the way?


reader Eclectikus said...

"Sympathetic" input on this subject in the arXiv Blog, it ends with this preventive ad-hominem (or maybe not so preventive if the author had already read this entry):

"Only a fool would bet against the possibility that constructor theory could also become a mainstream idea in physics that will have profound consequences for our future understanding of the universe. [Claude] Shannon who died in 2001, would surely be impressed."

https://medium.com/the-physics-arxiv-blog/9b8281bc793a


reader vlad25 said...

off topic: you may want to look at Nicholas Wade's new book, I can see a witch hunt developing there


reader RMB said...

Patrick Hayden too is not writing it off too quickly: "“The appealing thing about this formalism is the way that common features of quantum mechanics fall out,” says Patrick Hayden, a quantum physicist at Stanford University, adding: “I have real respect for the creative thinking behind constructor theory and its ambitions.” He notes, however, that there are competing attempts by other researchers to develop a deeper understanding of quantum mechanics, including ideas based on copying, and as yet it is too early to say which, if any, will prove to be the best description


reader B Chin said...

Deutsch as been giving talks on his so-called "constructor theory" at least as far back as 2003.


reader B Chin said...

There's much more, a whole field of quantum information theory. It's not trivial by any means.


reader RAF III said...

Nothing actually 'falls out' of constructor theory that isn't first dragged in. 'Superinformation', in particular, is simply an attempt to incorporate complementary observables into this non-substantive formalism.


reader J. Robards said...

I am shocked that educated people in physics would write up what clearly amounts to bad philosophy and then publish it. At bottom, the theory is another attempt to ground QM in CA, which is never going to happen. This is a DOA project. Why are so many brilliant physicists wasting time on crankery side-projects?


reader RAF III said...

So what does constructor theory have to do with physics?
But - hey - why let facts get in the way?


reader Joseph Melia said...

So where did I say constructor theory had anything to do with physics? Point I'm making is this paper has nothing to do with philosophy departments.


But--hey--why let the relevance of your response get in the way of responding?


reader Dilaton said...

Hm, maybe the people who decide about the funding of such projects should be reblaced by folks that are competent ...?


reader scooby said...

In the "many worlds" interpretation, there's got to be at least one "world" where Deutsch's constructor theory is right :)


reader Dilaton said...

Link ...?


reader Dilaton said...

LOL ... :-D


reader Dilaton said...

Ah, now I understand: the army of the dark Emperor consists of constructors, a scary vision ... :-)


reader Dilaton said...

It is obvious that once serious physics departments are getting invested with pointless or vacuous at least philosophical bubbling. The best solution to fight this desease is to eliminate its source ...


reader John Archer said...

The only thing scary about it is the fact that the portrayal of a highly indulged, out-of-control, raging, black ape, endearingly threatening weako wimpy whitey with extreme violence, all in a very laudable cause (i.e. as a suitable corrective for an inadvertent and very minor transgression against the Great Holy Green God Gaia), is to be considered more than perfectly 'acceptable'.

"But that's because it's all just comedy, see? With a nice little moral message. Simple. No problem. Be cool, man! Relax! Besides, we're just getting on with our regular righteous self-hate whitey-conditioning programme and making it very entertaining at the same time. What's not to like?"

Of course, had the producers reversed the race roles .... well, how do you think that would have worked out?

Well, I doan tink de big black judge would be amused, bwana.


reader Dilaton said...

Isnt this claim against the second law of thermodynamics? Reminds me of some crazy perpetuum mobile ideas ... :-D


reader Eclectikus said...

You mean this book, I guess...

http://www.amazon.com/Troublesome-Inheritance-Genes-Human-History/dp/1594204462

A quick look on google confirms that certainly is making some noise...


reader Swine flu said...

Is the field of quantum information significant just from the standpoint of computer science, or from the standpoint of physics as well? And if so, why?


reader B Chin said...

Deutsch is famous for one of the very early quantum algorithms. The first people to study quantum algorithms were necessarily eccentric, since few were crazy enough to think about such radically fundamental things. (Indeed, most physicists didn't think that there were any interesting quantum mechanics questions left in finite-dimensional Hilbert spaces.)

After Peter Shor invented the polynomial factoring algorithm and co-invented fault tolerance/error correct then the field caught on and has more mainstream types. BTW, there's a very funny book called "How the hippies saved physics" about the early days. There were some California new-age wierdos who thought that the EPR paradox meant that faster-than-light communication was possible. They were mostly ignored until one of them submitted a paper describing a device purported to communicate faster than light. Of course it was wrong, but several iterations of referees couldn't find the mistake. The error was finally found to be a violation of the "no cloning theorem", which was discovered in one of the referee reports.


reader B Chin said...

I'd say that David Weinland's nobel prize is a pretty good indication that the answer is "yes".

At the risk of embarrassing myself, I'll mention some of the connections. (I am NOT on the physics side of the field.) On the theory side of the quantum computing/information theory field there are some connections with condensed matter, for example the theory of "localizable entanglement".


On the experiment side of quantum computing there are course there are lots of connections with physics proper. In particular, building a working quantum computer is one of the sexiest and most challenging lab experiments imaginable. There are also applications to designing sensitive measurement instruments, for example atomic clocks and entanglement-enhanced optical imaging. Indeed, John Preskill at Caltech got into the field after realizing that the experiment was so sensitive that the gigantic weights in the LIGO gravity wave detector had to be treated quantum mechanically.

The pure quantum information theory point-of-view is to prove "hardware-independent" theorems about quantum mechanics. The natural questions bring to light features of quantum mechanics that weren't apparent previously, but that I think have some physical interest. Even the most basic results, such as "no cloning", quantum cryptography, impossibility of quantum bit commitment, ect, seem to me to give deep insights into quantum mechanics.


reader B Chin said...

I replied earlier, but it appears to have disappeared into the aether.

Given Wineland's Nobel prize lecture http://bit.ly/1k2EzKV, I would say that quantum computing/information is certainly significant from the point-of-view of physics.

I'm very far away from the physics side of the field, but I'll stick my neck out with a few remarks. (Maybe an expert will comment further.)


On the theory side there are certainly connections to condensed matter physics, and a fair number of people go in between both fields. For example, one has the theory of "localizable entanglement".

As far as experimental physics goes, perhaps it suffices to say that construction of an actual quantum computer is one of the most sexy laboratory results that one can imagine, requiring precise control of composite quantum systems. Furthermore, one of the principle difficulties in building a quantum computer, extreme sensitivity to the environment, can be turned to an advantage in building accurate measuring devices. For example, one can make better atomic clocks and use entanglement to enhance low-light optical imaging. Indeed, John Preskill at Caltech first became interested in quantum information because the gigantic weights used in the LIGO gravity wave detector had to be treated quantum mechanically because of the extreme sensitivity of the experiment.

As far as "pure" quantum information theory goes, in spirit it is the study of quantum mechanics in a hardware independent fashion. Some of the early results, such as the "no cloning" theorem, quantum teleportation, quantum cryptography, impossibility of quantum bit commitment, ect, aren't things someone working at a particle accelerator would necessarily think about, but I think they say something deep about our quantum mechanical world. They illustrate the difference between quantum mechanics and classical mechanics in a much cleaner and deeper way than the standard "black body radiation" chapter at the beginning of a traditional introductory quantum mechanics book. (It's almost a crime to start such a book in the usual way.)


Then there are quantitative bounds on the information-disturbance tradeoff, the ability to recover information stored in a physical system, and the like, which all seem like "real physics" to me.

Of course, there are other parts of the field that don't have the flavor of physics, but it gives a framework for asking basic questions about quantum mechanics that wouldn't have been obvious otherwise.


To avoid embarrassing myself I'll stop trying to talk about physics, though.


reader Gene Day said...

Who wrote the garbage that your link brings up, Eclectikus? I cannot find an attribution.
Was it Patrick Hayden? If so, it would appear that Stanford’s physics department has gone to hell.


reader Eclectikus said...

No, Patrick Hayden provides some sanity to the article indeed, it has had to be written by someone from the staff at arXiv Blog.


reader Gene Day said...

Thank you.


reader Gene Day said...

What does it mean to develop a deeper understanding of QM? Even looking for such a thing implies that, somehow, QM is incomplete. It is not.


reader Swine flu said...

Thank you for your detailed reply. I guess the more general way I could have framed the question is by asking whether "information" really needs to be added to the small list of fundamental physics concepts.


reader B Chin said...

Oh, well, certainly it's nice to be able to understand the von Neumann entropy in terms of quantum data compression. And I certainly believe John Preskill's montra "Information is physical"-- meaning that information doesn't exist unless it is stored in a physical device, and it can't be processed except by a physical device subject to all the laws of physics. But if you're asking the deeper "it from bit" question of whether some future theory of physics will be fundamentally about information, then I have absolutely no competence to form opinion. That's way above my pay grade.


reader Swine flu said...

The use of physical principles in designing novel computers leaves less room for crackpottery than trying to go in the opposite direction. On the other hand, I've been too lazy to read up on the attempts to use information ideas in fundamental physics, so I am trying not to jump to conclusions even though I feel like it at times. :)


reader vlad25 said...

After having read this book I have to say that Wade's arguments are often pretty weak. Yes, it's a very interesting topic which has been shunned because of political correctness, and yes, Wade gives some interesting sources, but he doesn't really add anything substantial, he uses terms way too loosely, sometimes he's just contradicting himself, really painful.
I'm also wondering whether he has been self-censoring on the intelligence topic.


reader deeeatthh said...

That was a bit harsh Motl. Nevertheless, I definitely want to have your baby hearing what you say there. What is happening in science at the moment is really scary. People are forgetting what it even is. People are never finding out in the first place. Science is really fragile in this way, because no one ever wrote down or made a list, what it takes to be science. Got passed down from one brilliant guy to the next. It worked...because successful threads in scientific discovery stayed successful for generations.
Science much more like an organism than a philosophy. It lives on. But it can die.


reader Luboš Motl said...

Your comment is intellectually vacuous anti-scientific populist rubbish.


reader deeatth said...

Don't you think his part in Quantum Computing is exaggerated? He gets called a pioneer, but for that to be meaningful his early paper would have to have gone on to be influential, and form the bedrock of later developments. None of this happened at all.
Was he the first? No. Does his algorithm yield? No. Have his ideas or his work produced or seriously prerequisite subsequent historically significant breakthroughs? No.
Nor is any of the above likely to transpire in the future.
So why exactly, how, does he come by the prestige sections of the media and scientific chattering classes bestow on him. And what gives these people the right to pick and choose absent due process?
Deutsch hasn't made a major contribution to science.


reader deaaath said...

To be fair on the reporters, the fact is they don't seem ever to critique newly announced work by whoever. Their style is, give a big cheer..whaateeever!


reader deeath said...

Well actually, Hayden gets it wrong. But that's not to say any stream of effort should be written off.
Leave the door ajar by all means, but we're talking there about the future and what may happen. But Motl's, so far as I can see; and FWIW the way I perceive things too, is not about writing the future off for Deutsch, but the ideas and claims attached to them, already presented.
Why is Hayden wrong. Deutsch has chosen a specific kind of seed proposition, that of ubiquity. In other words, the Constructor concept ubiquitously fits almost every coordinate in the universe.
I.e. any locale anywhere corresponds to a background dimensionality (the constructor) and the instantaneous intersections of particles, light, wind, etc (the task substrate)
This is a perfectly legitimate way to kick off an attempt at a big theory. the big category level there's only really two ways to go. This is one - that being the abstract, abstract enough to be consistent with just about anything at or near ubiquity.
Newton for an example of the other one: precise description of a universal or something important and world changing.
Both approaches are scientifically legitimate. But they are almost perfect mirror opposites, in that Newton presents the result of 20 or more years work done by a true genius. So there's not much on the to-do list for Newton. It's done.
Deutsch presents a generic..an abstract. As such it holds the status of a seed idea only. It's a seed idea. They aren't hard to do. For example, another possibility would be something like "things absorb things". It's got the ubiquity, but it's worthless pending the work to come, that builds something world changing onto that space.
The ubiquity start point, represents the beginning of a journey and the precision-of-the-specific represents the end of a journey. The complaint some have with Deutsch, is that, well he doesn't seem to get it. Hayden probably gets it in general, but only pretended to attend to Deutsch's idea.
But Deutsch doesn't seem to get it. If he has good friends one of them should clarify the matter for him, before he inflicts lasting damage to his credibility.
So what he doesn't get, is that if you define a generic that is ubiquitously true, then what you have is a truism, that you might as well call for what it is: vague. Not a problem...so long as you get it. So long as you know all the work is ahead. The genius, it it's there, is in the future, ahead.
A Corollary of this is that the ubiquitously true abstract, is obviously going to pair up to QM, energy, Information, locality. Whatever you want., It's ubiquitous truism.
Therefore simply defining shit based on the abstract, without having first done the hard work and put the meat on...is meaningless. Specious. Stupid.
He's not a bad guy IMHO. But I'm with Motl. And so is logic and reason and all good sense


reader deeathth said...

Motl...Even you still need to read what people say. I was agreeing with you.
The problem when scathing-ness ends up your default, is laziness. It's much easier than to discern.
Also, do you know you just told me the nature of science itself, in the same breath speaking of bullshitting?
You don't know the nature of science better than I. Even were should you, you've no means to say.
The scientific revolution remains unexplained. Philosophy is pre-science, so that's no good. You don't know. That you think you do, puts you with Deutsch and the others.
We don't know yet, and won't until Science itself extends upward over itself.


reader deeattth said...

dude you jumped me then deleted my reply. Don't make come and bust you up man (just kidding)


reader deeath said...

I was going to show you my secret underground station and let you go in the hovercraft BUT NOW I WON'T


reader Luboš Motl said...

Dear Deeath, I blacklisted you because you have been using sockpuppets, most recently "Jon L", to fill my blog with rubbish. The amount of your idiotic and obnoxious comments on this blog was marginally unacceptable already before I noticed your sockpuppetry.


reader jonnydark said...

"Science is really fragile in this way, because no one ever wrote down or made a list, what it takes to be science."

This is patently false. Have you read ANY philosophy of science?

Try reading some Karl Popper, he claimed that falsification was the essential criterion for a scientific theory. In short, for a claim to be scientific, it needs to testable in a way that could prove it to be false. Scientific theories are never proven correct - the best ones are just the most corroborated by evidence. This was in part an attempt to distinguish the natural sciences from what Popper deemed to be pseudosciences such as Freudian psychoanalysis and astrology.

Prior to Popper there were the enlightenment ideas of the scientific method underpinned by inductivist logic. I was taught the scientific method when I was in high school!

And it didn't even stop at Popper, Thomas Kuhn went on to claim that science doesn't operate using the scientific method proposed shifting paradigms within science.

So while there is still debate about the nature of science, there is a core that all published and accepted science adheres to. To claim that "People are forgetting what it even is" and that


reader B Chin said...

He discovered the first example of a quantum algorithm outperforming a classical one. This encouraged people who were paying attention to search for other such algorithms, which helped get the field going. In the early days there was lots of low-hanging fruit, but 99.9% of the people studying quantum mechanics just couldn't see it. It didn't occur to them that simple theorems in quantum mechanics could have a profound impact on the notion of a physical computing machine. It took some far-out and wacky people to discover this research direction.


reader deeeeeeeaaasathhhhhh said...

Hi Jonny,
I hear what you say, but Karl Popper produced almost all his work relating to science, not from study of science at all, but attempts to explain science such as empiricism, inductivism and so on. It's worth noting that Deutsch is probably the world's foremost Popperian and that he uses that philosophy to produce his work and that philosophy is represented at every level in the work itself.
Popper knocked things like empiricism down on technicalities, as if they were completed rigorous philosophies. Which they weren't ...they were more like observations, mostly arising from pioneers in science. Who by the way, very openly acknowledged there was a large amount going on in terms of the basis of discovery that was mysterious to them. They were too busy discovering to bother with philosophy, so the idea behind their...fragments..of philosophy was to add some value to what they assumed would be a work eventually, of science. Not philosophy.
Popper then uses his criticisms to define philosophical errors. Like justificationalism or foundationism, inductivism and so.
He adds several more components based on those components.
Then he argues that all that is left is conjecture...and refutation. And he's right...but only in the world he just created.
Kuhn's basic idea was that discovery was ultimately fuelled by the build up of anomalies. However, he never clarified the kind of anomaly he meant. He probably meant anomalies that emerged from robust scientific discovery. I would say that is implicit. But because he left it unstated, his ideas are now used by many of the forces that for whatever reason seek to overturn robust science (i.e. weak evidence for telepathy like testimony are forwarded as anomalies).
Inductivism was an important early attempt to explain what was happening. But as mentioned it was much more akin to an observation than a finalized philosophy (hence poppers demolishment of it was not legitimate). Inductivism leaves many questions unanswered. It was very much a statement of what seemed to be happening. Testimony from the geniuses.
FWIW - the way you presented your post to me was unnecessary scathing. You should not follow the behaviour of others if that is what you are doing. If for no other reason than it makes you look weak and easily led.
You also presented yourself as someone with more extensive and more depth of reading and thinking than I think you probably. For example your characterization of Popper is fairly close to the hallmark set of traits Popperians associate with novices.
I wouldn't say this to someone publically normally but you appear to be anonymous as am I, so I presume you will not object.
Note to Lubos Motl: I wasn't sock puppeting which is linked to trolling. All I was doing was 'starting over'. You had already begun to block me or apparently so. I had wrote a couple of posts clearly intended for you alone since you were blocking me. You posted them. Perhaps it wasn't purposeful. I'll let it go.
But so long as you don't have just grounds to throw me out, I will come and go as I please.
If you want to put your time into vigilance over this, then that is even more reason for me to sign off here by first complementing what appears to be your amazing intelligence. But then. Why are you throwing your life away on this blog? You should not give up over what happened. And it isn't appropriate to your potential what you are contributing.


reader Matthew Lubic said...

Hmmm... well.... maybe.... perhaps it's because you don't understand something that they do, or maybe they understand something that you don't.

I guess you could call that either a trivial profundity or a profound triviality. You choose. Or choose not to choose.

Be well... even if it takes effort. And try to chill. It's much less expensive in terms of your survivability, takes little effort and can prove to be of much, much value... to you.


reader cel1ne said...

Addendum: Science also usually employs Occam's razor.