Google Wave and the future of Science Communication

Cameron Neylon reports in Nature that Google is beta-testing a science communication tool using their recently announced technology, Google Wave.

Google Wave is a new technology being developed by the internet giant to be “what e-mail would look like if it were invented today”. Its online collaboration technologies promise to provide a new framework that could potentially help improve, well, almost everything we do online.

And that includes, of course, Science. In a somewhat ironic situation, scientists have been slow in realising anything close to the full potential of the internet in improving the way they collaborate and publish the results of their work. Journals have gone online, sure — no scientist gets their journal articles from the walk-in library these days. New scientific publications are published online first, and although I don’t have the data, it is a safe bet that the overwhelming majority of readers access their content through digital means. But that is not much beyond just a more efficient distribution for essentially the same content.

Although a large part of that, I believe, derives from the inertia of the current system of evaluation of scientific output, perhaps it is partly to blame on the lack of tools useful enough to become adopted by a critical mass of users. Perhaps some system based on Google Wave will be able to overcome that inertia. In his Nature article, Neylon points to a couple of expected benefits of this possible new tool: automated “robot” agents that could be used to perform tasks like update your wave content with new laboratory data in real-time, and a version control system that would make it easier for say, peers to check through analyses of scientific data, and potentially spot frauds or honest mistakes more easily.

What I found a little disappointing was that Neylon seemed to focus too much on the “error-correcting” capabilities of the new technology. The internet can do so much more than police scientists’ data and analyses. It can connect individual human beings in a way that makes the collective effort qualitatively different from anything that the individuals could achieve by themselves, or with just “nearest-neighbour” interactions. This is the goal we should aim at. All the knowledge and all the diverse ranges of intellectual abilities found in individual scientists (and ordinary people, why not?), directly connected with each other in a global brain with many more interconnections and forms of interaction than we currently dream of. I have no idea what will result from that, but I am extremely excited to find out.

The Ultimate Question of Life, the Universe and Everything?

“Yes,” he said at last in rather a strained drawl. “I did have a question. Or rather, what I actually have is an Answer. I wanted to know what the Question was.”

Prak nodded sympathetically, and Arthur relaxed a little.

“It’s… well, it’s a long story,” he said, “but the Question I would like to know is the Ultimate Question of Life, the Universe and Everything. All we know is that the Answer is Forty-Two, which is a little aggravating.”

Prak nodded again.

“Forty-Two,” he said. “Yes, that’s right.”

He paused. Shadows of thought and memory crossed his face like the shadows of clouds crossing the land.

“I’m afraid,” he said at last, “that the Question and the Answer are mutually exclusive. Knowledge of one logically precludes knowledge of the other. It is impossible that both can ever be known about the same universe.”

From Douglas Adams’ Life, The Universe and Everything

Few numbers have such a geek cult following than the number 42, thanks to Douglas Adams’ science-fiction series The Hitchhiker’s Guide to the Galaxy. I guess one of the reasons is that behind the sophisticated humour of Douglas Adams lies an interesting philosophical question. Is there an “ultimate” scientific or philosophical question? If so, what is it? (Perhaps this one?).

Of course, the humour of Douglas Adams sarcastically dismisses the idea of an ultimate question as silly. And in a way it is. But last month, in a meeting at the Perimeter Institute for Theoretical Physics called “Reconstructing Quantum Theory“, Bill Wootters presented the closest I’ve ever seen to a candidate. Bill found a formalism in which quantum mechanics can be represented in a real vector space, as opposed to the usual formulation in terms of complex vectors spaces.

The upshot is that he needs an universal rebit to be able to reconstruct quantum mechanics with that formalism. A rebit is just like a qubit but instead of a superposition of states with complex coefficients, you have one with real coefficients.

As an aside for those who don’t know what a qubit is. A quantum bit it is the quantum extension of the concept of a bit—the unit of information, the amount of information one obtains when finding the answer to a yes/no question. To represent a bit, all you need is one thing in one of two possible states, which are usually denoted ‘1′ or ‘0′. A coin, for example. Heads indicates ‘1′, say, and tails indicates ‘0′. Given a previously agreed code, you can transmit information with a sequence of coins.

In quantum mechanics, for every two possible states of a system (say, ‘1′ and ‘0′), there are an infinity of possible states related to these two states by what are called complex superpositions. Those are mathematical structures that can be represented in the form c1 ‘1′ + c2 ‘2′. Except that in quantum mechanics a state like ‘1′ is represented by the symbol |1>, a notation introduced by Dirac. So for example we could define states like |+> = |1> + |2>, |-> = |1> – |2> or |R> = |1> + i |0>, etc. However, in any complete measurement of a qubit, only two outcomes are possible; a qubit can give you one bit of information, but one bit of information about an infinitude of mutually exclusive (or what Bohr called complementary) questions.

A rebit, or real bit, is an intermediate case where the coefficients c1 and c2 are allowed to be only real numbers.

This rebit is called universal because—and this is where things get interesting—it is in some sense shared by all other systems in the universe, according to the model of Wootters. It interacts nonlocally to all other rebits in the theory. In the model it is also necessary (so as to reconstruct quantum theory) that we are unable to determine the state of this rebit, even though it interacts with everything in the universe. This raises some interesting questions. Is the state of the rebit unknowable in practice or in principle? Wootters showed some interesting models for mechanisms responsible for this epistemic censorship. In one of these, the rebit is randomised by a very rapid rotation of its direction, much faster than we can experimentally detect.

But thinking about it, the rebit itself represents the answer to one question. One binary question, the answer of which is relevant to all systems in the universe. An ultimate universal question.

After a moment of suspense, Bill advanced his hunch: the ultimate question associated to the rebit relates to the direction of time. (Chris Fuchs, who was also attending the meeting, exclaimed disappointedly: “Direction of time? I was thinking of something more like the triumph of good over evil!”) However, this interpretation would seem to pose a problem to his model of a rotating rebit. In which “time” would the rebit be rotating, if it itself “encodes the direction of time”? Furthermore, by the principle of superposition there wouldn’t be just one question, but a continuum of complementary binary questions, the answers of which would be all the orthogonal pairs of real superpositions of |1> and |0>. Would those correspond to other possible pairs of directions for time?

I am very curious to read Bill’s paper about this work to understand his model in more detail. But whatever the question associated to the universal bit is, the answer can’t be 42 after all. It can only be yes or no. Or any real superposition thereof.

Google Scholar – hidden features

A quick tip: did you know that in Google Scholar you can add your local institution’s library (and the nearby institutions, and perhaps some national library you may have access to. Just type the name of the institution and they’ll find it for you), and show links to export citations to BibTeX? Just click on “Scholar preferences” right next to the search bar.

That’s gonna be a huge hand!

New prepint on the arxiv: “Experimental criteria for steering and the EPR paradox”

I have submitted a new paper to the physics arxiv, entitled “Experimental criteria for steering and the EPR paradox”. It is joint work with my collaborators Steve Jones, Howard Wiseman and Margaret Reid.

I’m particularly excited about finally getting this paper off, as it was holding back a few other works — and because I think is a nice paper too! Click here for the arxiv version.

Here’s the abstract:

We formally link the concept of steering (a concept created by Schrodinger but only recently formalised by Wiseman, Jones and Doherty [Phys. Rev. Lett. 98, 140402 (2007)] and the criteria for demonstrations of Einstein-Podolsky-Rosen (EPR) paradox introduced by Reid [Phys. Rev. A, 40, 913 (1989)]. We develop a general theory of experimental EPR-steering criteria, derive a number of criteria applicable to discrete as well as continuous-variables observables, and study their efficacy in detecting that form of nonlocality in some classes of quantum states. We show that previous versions of EPR-type criteria can be rederived within this formalism, thus unifying these efforts from a modern quantum-information perspective and clarifying their conceptual and formal origin. The theory follows in close analogy with criteria for other forms of quantum nonlocality (Bell-nonlocality, entanglement), and because it is a hybrid of those two, it may lead to insights into the relationship between the different forms of nonlocality and the criteria that are able to detect them.

My impressions on the Vienna Symposium

Finally have some time to blog. I’ve been travelling around going to conferences and visits, and I’ll try to leave some of my impressions here. The first stop was the Vienna Symposium on the Foundations of Modern Physics, from  June 11 to June 14. Yes, I know, that is so last month. But hey, I’m travelling, ok? Give me a break.

It was organised by the Institute of Quantum Optics and Quantum Information (IQOQI) at the Austrian Academy of Sciences and by the Faculty of Physics at the University of Vienna. Daniel Greenberger and Helmut Rauch were honoured guests, in the occasion of their 75th and 70th birthdays, respectively.

The list of speakers was impressive, and the talks delivered were on par. Some of the highlights, in my opinion:

Bill Wootters opened the conference with an interesting model of Quantum Mechanics on real vector spaces. His motivation was to give an answer to the commonly held misconception that complex numbers are essential for quantum theory. It is to be expected that any theory that reproduces quantum mechanics has got to be weird, and Wooters’ model isn’t an exception. He models a qubit as being composed of two parts, both represented in real vector spaces. It has to have two components if he is to have sufficient parameters to compensate for the lack of complex numbers. Besides the actual qubit there is an “ubit” — an “universal bit”. There’s a rule that one cannot learn the state of the ubit (so that you have something of the taste of an uncertainty principle), and furthermore, all particles in the universe share the same ubit (and here enters quantum nonlocality). I asked (yay! the first question of the workshop!) if he thought there was an analogy between his postulate that the ubit is unknowable and the epistemic restriction in Robert Spekkens’ toy model (hey, just found there is a Wikipedia entry for it!).  He said it was a very interesting question (thank you!) but he didn’t think about it. In hindsight, there is an analogy but also an important difference. In Spekkens’ model the epistemic restriction is about the ontic state of an individual system, whereas in Wootters model it is about an entity shared between all systems. This is why Wootters model can actually reproduce quantum mechanics, whereas Spekkens’ toy model can’t reproduce some features like violations of Bell inequalities.

Simon Kochen (from Kochen-Specker fame) followed with a talk titled “A Reconstruction of Quantum Mechanics”. He started by essentially assuming a Hilbert space structure for the experimental outcomes (although phrased in a quantum logic parlance: lattices and so on) and derived the rest of the formalism from it (with some extra assumptions he tried to justify with an appeal to experimental facts). However, it is well-known that if you assume the Hilbert space structure you can derive the quantum probabilities via Gleason’s theorem. When I asked him about how he justified that assumption (yay! the first question on the second talk of the conference!), he said something like “I am not trying to reconstruct quantum mechanics”.

After the coffee break, it was Stig Steinholm’s talk “Quantum Theory and reality”. He tried to find a way to ground the information encoded by a quantum state in an objective observer-independent manner. Anton Zeilinger added a comment near the end of the question time: “The concept of a reality beyond empirical evidence is not part of Science”. Steinholm’s answer was brilliant: “What’s the empirical evidence for that statement?”

Basil Hiley’s talk on how to derive the basic equations of Bohmian Mechanics for the Dirac equation using Clifford Algebras was very entertaining. Although he seemed to want to distant himself from being called a “Bohmian”. “All I’m saying is that it’s all there in the mathematics, you can interpret it as you want” he replied to a skeptical Zeilinger.

On the second day, Markus Arndt detailed the state-of-the-art on some of the very interesting experiments on matter-wave interferometry. They can interfere objects of up to 2934 atomic mass units and 5600 vibrational modes! One of these is called the “Vienna quantum man”, as the molecule has a shape that looks a bit like a stick man.

Raymond Chiao wants to test Heinsenberg’s Uncertainty Principle against Einstein’s equivalence. He claimed that the clash between the two could be tested with accelerated superconductors. His analysis seemed highly controversial, but hey, there was a clear experimental test proposed, so we can just leave it to experiment to decide.

Nicolas Gisin gave an interesting talk detailing some very nice Bell nonlocality experiments, including of a concept called “Bi-locality”, where the correlations between three particles are modeled by independent local correlations between a central particle and each of the other two. This being a weaker assumption than full locality, the experiments are easier to make, and can be more rigorous as far as loopholes are concerned. Needless to say, quantum mechanics was still upheld.

Abner Shimony’s (from CHSH fame) talk “Quantum Mechanics and Mind” — where he proposed an experimental test for the hypothesis that a definite reduction of a superposition only occurs after a conscious observer learns the outcome — caused clear discomfort in the audience. He was followed by his colleague Michael Horne (the second H in CHSH), who talked about a neat result on the shifting of fringes in an interferometer due to an applied external force.

It was inspiring to see Daniel Greenberger, (from GHZ fame) detail his “Tic-tac-toe theory of gravity”. He starts from a simple assumption that there are three types of mass which can repeal or attract each other in different ways, and derives from very simple non-mathematical arguments things that look like dark matter and the accelerated expansion of the universe. Interestingly, he derived those results before the existence of these phenomena was known (!) but the manuscript was rejected by an angry referee who replied that “we don’t need any new ideas on this field”(!!!). This got to be one of the best referee quotes ever! Surely, there are many problems that would need to be worked in his theory, like why don’t we observe clusters of the other types of matter, to which he gives only tentative partial answers, but he was the first to say he doesn’t take this theory seriously. His main aim was to present it as a motivational talk for the students and postdocs in the audience, so that we not believe that our elders know everything. Thank you, Danny!

I can’t comment on Nobel-prize winner Gerard t’Hooft’s talk. All I can say in my defense is that they shouldn’t have put him first in the morning after all the wine they served at the conference dinner! (ahem!) Well, ok, it was about trying to reproduce quantum mechanics using cellular automata, and the little I saw of it sounded very very interesting! Damn it!

Next it was the turn of Reinhard Werner (from Werner state fame) to remind us that quantum states cannot be thought of as being attributed to each individual system, as this amounts to a local hidden variable theory (alluded to in his title “The most popular hidden variable theory ever”). Actually, that is a fact “well-known by those who know things well”, as a colleague likes to say: it amounts to a quantum separable model, the violation of which is a demonstration of entanglement. But some things need to be repeated until they sink in, I guess.

Bill Unruh (from the Unruh effect fame) cautioned us that looking at the reduced density matrix for signs of decoherence can lead to wrong conclusions. Sometimes it is possible to have large entropy in the reduced density matrices but almost perfect coherence in interference experiments. Thanks for the heads up Bill!

There were many other very interesting speakers and posters, but I wouldn’t have the space to comment on them all, and if you have read this far you will agree with me. As usual, Zeilinger gathered a very nice group of physicists! Looking forward to the Vienna Symposium of 2011!

Bill Hicks documentary

An awesome documentary about my stage hero:

http://www.guba.com/watch/2000950423

(I removed the link that opened in this window because it was annoyingly starting even if you didn’t press play.)

Newcomb’s Paradox at the Gates of Heaven and Hell

In Limbo, outside the Gates of Heaven and Hell, St Peter stands with a scroll. He opens it up and reads it: “God has predicted everything you have ever done, and ever will do. He is not a magical being, however, but an extremely large mind capable of processing the consequences of all the actions that have ensued since He set the world running. For most people whether they go to Heaven of Hell is based on their deeds in life, but for those who wasted too much time with philosophy and learned about Decision Theory, He created a paradox to simplify processing. Here are two boxes. In one of them — it is open — there’s a guarantee that you won’t be tortured eternally if you go to Hell. But it will still be painful and unpleasant. There’s also vip tickets for rock concerts in Heaven. You can watch the Rolling Stones playing Sympathy for the Devil in 1968’s Rock and Roll Circus. Live. God diggs that gig. In the other — it is closed — there is either an entry ticket to Heaven or nothing. You can pick both boxes, or just the closed one. If God predicted you are going to pick just the closed box, He has already put the ticket to Heaven inside. If He predicted you are going to pick both boxes, he has left it empty.”

Bayesian Decision Theorists reason that whatever they do now, God will have predicted it. So they pick the closed box and go to Heaven.

Causal Decision Theorists reason that whatever they do now won’t cause the contents of the box to change. That regardless of whether or not there’s a ticket to Heaven in the closed box, they are better off also picking the open one. They pick both boxes, find nothing in the closed one, and go to Hell. But at least they are not tortured. And they had a smoke in life.

Some people keep their minds trapped in a looping mental calculation and stay in Limbo forever. St Peter tries to remind them every million years that they are welcome to pick the closed box and come to Heaven. Most remain unconvinced.

Web democracy

This post will be an exercise in futurology. I predict that in the next few decades we will see some interesting debates about the legal rights of internet users over the content they create and their personal information, and their right to scrutiny and decision power over how their flux of information is handled by corporations.

Say you create a social network website that allows people to upload their music libraries, which will be available to other users of the website. With the information gathered from the network of users, the website offers me “recommendations” of music I may like. The recommendation value of a particular piece of music may be based on the likelihood that users with libraries similar to mine also have that piece of music in it.

Now say you are offered a lot of money from a music distribution company to tweak the results of the recommendations so that certain artists will be recommended more often than others. Should you accept the offer? Should you have the right to accept the offer?

This dilemma is no different from that faced by radio stations about the playlists they air (although radio is not nearly as influential a media today as it was in the past, I believe). But while the question has been legally ignored in the past (as far as I know. Which is not much. But this is just a blog post so stop being picky), the 21st century may bring a whole new light to it. And to many other similar problems which never actually existed before.

That’s because in the past it seemed to be virtually impossible to enforce anything about this. How are we going to know that they have changed the playlists? Even without any influence from commercial interests, it would come down to the expertise of a DJ in feeling what kind of music his audience would like to hear. How can we know the DJ hasn’t skewed his judgement? It’s not like there was an algorithm to decide about playlists in the past!

But now there is! The website last.fm (by the way, here is my Music Profile) uses some sort of algorithm  — which I don’t know precisely, but which no doubt uses data gathered from people with similar interests — to provide me with recommendations. Technically, we could now know (or be able to have a high degree of confidence on) whether websites like last.fm are skewing their recommendations to fit commercial interests.

In more important issues, we could know how Facebook or Google decide on the advertisement they provide. We could know how Google ranks the websites they display after a search. We should be able to have some degree of trust that Google isn’t reading our Gmail or using the information in our private content for their commercial purposes.

You could argue that once an user enters an agreement — you know, that interminable fine print in those “check the box to use our services” screens — with these companies, they accept those terms and are subsequently bound by it (Some people have analysed the fine print on Facebook’s Terms of Agreement really carefully, and it’s scary).

But when the user base reaches a certain point, we are not anymore talking only about a non-contextual transaction between an individual and a company, but between a population and a company. And here the issue becomes more complicated.

Users value not only the content generated by individual websites, but increasingly they value the relations they maintain and the content they share within the network: the friends they have added to Facebook, the music they can be recommended by other people with similar interests, their MSN contacts and their associated chat history, and many other similar information exchanges.

Some could say that open-source software could provide an alternative, and if people don’t change to open-source, they’ll have to accept the terms of agreement offered by the commercial providers. But commercial companies will likely always have more resources to make their software attractive, and so people will likely always have an inclination to use commercial software.

I don’t have an option to join an open-source social network. If I don’t like the terms of Facebook’s or Orkut’s agreement, I just can’t be on a social network. This is a very different situation from that of my agreement with Gmail: if I don’t like Gmail, I can with some ease switch to the similar service provided by a plethora of different email providers. But I’m stuck with Facebook an Orkut if I want a social network. And so are all the other users.

The network of users controlled by Facebook cannot be reconstructed without a coordinated effort of all users in the network. The amount of cooperation required to make all users on Facebook migrate (and add precisely the same content with precisely the same relations within the network) from Facebook to, say, Orkut, Google’s rival, would be comparable to a war effort. There is a huge inertia in that system. The communities of users represented on Facebook or Orkut are attached to those providers in much the same way that communities of people are attached to their country.And as the complexity of those networks increases, this fact will be even more evident.

Should those communities be able to change the terms with which they are bound to their “governing body”, i.e., the companies that serve, coordinate and control their interactions? How could that happen? Since some of those network are composed of people in different countries, how can their rights be handled by the present systems in power? Or should we just leave it to free market? Can the interplay of competition, offer and demand alone lead us to more democratic Terms of Agreement to rule over our increasingly complex virtual interactions?

My talk at “Causation and Decision”

The Centre for Time at The University of Sydney has made available the audio and slides for a talk I have given at the “Causation and Decision” conference in Sydney in January. They can be found here, under the tab “Causation and Decision”.

This talk was essentially the same that I have given at the “The Clock and the Quantum” conference at Perimeter Institute — “Newcomb’s problem in the light of Bell’s theorem”. Professor Huw Price, Head of the Centre for Time, saw that talk and invited me to presented it in this conference in Sydney. Huw talked after me building on my argument against Causal Decision Theory and presenting his proposal for an unifying view.

Actually, there’s also another set of slides/audio from a talk I gave at the first workshop organised by the Perimeter Institute – Australia Foundations (PIAF) collaboration. That’s under “PIAFWorkshop2008″. That talk was titled “Steering, predictability and other concepts of ‘experimental metaphysics’”.

The Centre for Time webpage says those files will eventually move to the Sydney eScholarship Repository, so try this link if the former are dead.

Darwinian Love

In honour of Darwin’s bicentennial anniversary and Valentine’s day:

———-

Darwinian Love – Valentine’s-day version:

Way back in the primordial periods of evolution, the two unicellular organisms meet:

1 – “You complete me in so many ways!”

2 – “I know!! Feels like we are made for each other!!”

1 – “I’ve been through so many mutations in this life. Only you can fix my broken genes. I’ll never let you go! I’ll always be by your side!”

2 – “I love you so much I wish we could just merge into each other…”

1 – “Let’s just be together… Lay quietly with me. Just listen to my mitochondrias beating…”

2 – “Wow! It’s like a big warm star coming from inside your cellular walls…”

1 – “Just be with me…”

Her nucleus lights up, and she exhales a fragrant mixture of molecules. He rotates his flagellum excitedly, propelling himself towards her. They merge as one, interlacing the most intimate parts of their genetic material,  in a fulgurant dance of chemical bliss.

3 – “Wow… I never knew…”

———-

Darwinian Love – Darwin-day version:

Way back in the primordial periods of evolution, the two unicellular organisms meet:

1 – “You complete me in so many ways!”

2 – “I know!! feels like we are made for each other!!”

1 – “I’ve been through so many mutations in this life. Only you can fix my broken genes. I’ll never let you go! I’ll always be by your side!”

2 – “I love you so much I wish we could just merge into each other…”

1 – “Let’s just be together… Lay quietly with me. Just listen to my mitochondrias beating…”

2 – “Wow! It’s like a big warm star coming from inside your cellular walls…”

1 – “Just be with me…”

2 – “But what are we gonna say to everyone? Our twin sisters, aunts, grand-aunts, first- and second-cousins? You know how eukaryotes can be too nucleated about these things…”

1 – “We’ll say that we l.. ll… love each other! That we want to be together forever. That from now on, we don’t want to be seen as two individual cells but as a… an… organism!”

2 – “Great!!!… But what are we gonna do about babies? I mean, if we just reproduce separately, then when I become my daughters they… I mean, I…  I always get confused with personal identity in the after-mitotic life… anyway, we may not be together anymore!

1 – “Hmmm… yes… aha! Yes, great idea!” — and her flagella are aroused in excitation — “we’ll just reproduce at the same time!! Let’s call it ’sex’!”

2 – “Hmmm… you naughty boy…! I see where you wanna get with this sweet talk…!”

1 – “Boy?… hmmm… I like the name… so anyway, we get together, and… and… when we are reproducing, we… we… we synchronize it so that… our genes interlace with each other and, and… and we will fuse as one! And we will be fitter than ever, and our children will conquer the world with their creativity and diversity, and…!”

2 – “Shh… Come here, let’s just do it…” — her nucleus lights up, and she exhales a fragrant mixture of molecules. He rotates his flagellum excitedly, propelling himself towards her. They merge as one, interlacing the most intimate parts of their genetic material,  in a fulgurant dance of chemical bliss.

3 – “Wow… I never knew…”