Slashdot Mirror
String Theory Predicts Behavior of Superfluids
schrodingers_rabbit writes "Despite formidable odds, condensed matter physicists have made a breakthrough most thought impossible — finding a practical use for string theory. The initial breakthrough was made by physicist and cosmologist Juan Maldacena. His theory states that the known universe is only a 2D construct in anti-de-Sitter space, projected into 3 dimensions. This theory manages to model black holes and quantum theory congruently, a feat that has eluded scientists for decades; but it fails to correspond to the shape of space-time in the known universe. However, it does predict thermodynamic properties of black holes, including higher-dimensional viscosity — the equations for which elegantly and almost exactly calculate the behavior of quark-gluon plasma and other superfluids. According to Jan Zaanen at the University of Leiden, 'The theory is calculating precisely what we are seeing in experiments.' Unfortunately, the correspondence cannot prove or disprove string theory, although it is a positive step." Not an easy path to follow: one condensed matter theorist said, "It took two years and two 1000-page books of dense mathematics, but I learned string theory and got kind of enchanted by it. [When the string-theory related] thing began to... make predictions about high-temperature superconductors, my traditional mainstay, I was one of the few condensed matter physicists with the preparation to take it up."
The Maldacena duality can't be used to 'make predictions' with a string theory, its just a correspondence between a string theory and a conformal field theory. It's useful because sometimes calculations which are hard in a CFT can be made in the corresponding string theory which is sometimes easier (or vice versa). It cannot be used to support the physical validity of some string theory.
Stick to physics; evolution does make testable predictions. It usually takes a while to run the tests, though.
Evolution certainly makes concrete, testable predictions: about what we expect to see in the fossil record, about what we expect to see in the genetic makeup of various species, about what we expect to see in the phenotypic features and behaviors of modern species, and about how we expect species to change over time. As a simple example of the last, consider http://www.ted.com/index.php/talks/paul_ewald_asks_can_we_domesticate_germs.html
Parent post is insightful. If a model is flexible enough, it can fit any data.
This is what makes evolution ... neither makes concrete testable predictions.
Really?
http://en.wikipedia.org/wiki/Tiktaalik#Discovery
"It's one of those things you can point to and say, 'I told you this would exist,' and there it is."
"Also, I think sometimes they like to show off by writing things people can't understand."
Definitely. E.g. the intro for "dot product" says "It is the standard inner product of the orthonormal Euclidean space." If you're trying to work out what a dot product *is* then that is a completely useless and confusing statement. Mathworld is usually much better than Wikipedia in this respect.
Also, you were Educated Stupid
string theory is kinda unelegantly difficult, so a lot of people don't really want it to be true.
Because quantum mechanics is so elegantly easy?
I think we need to face facts, here: no GUT is gonna be simple. If it were, it probably would've been discovered already.
Because quantum mechanics is so elegantly easy?
well... actually it is relatively easy, and mathematically not extremely different from classcial physics. Basically you write a 'h' in some places where there used to be a '0', and that apparently has all this implications as wave/particle duality, uncertainty principle, observing = changing, etc...
A bit hard to imagine, and sometimes counterintuitive, and the calculations can be quite some work (although QED and Feynman diagrams etc has made a lot of the calculations a lot easier. Make the diagram, for every line and knot in that diagram substitute a term in your equation, et voila!). It's actually a very beautiful theory. While the superstring stuff still remains a bit vague and complicated.
Planck's and Einstein's explanations for blackbody radiation and the photoelectric effect are generally not considered to be quantum mechanics. Essentially, they were phenomenological explanations for strange experimental data, and were not any sort of coherent, all-encompassing idea. Together with Bohr's model for the hydrogen atom, they are collectively referred to as the old quantum theory. Actual quantum mechanics got its start with Heisenberg's matrix mechanics and Schrodinger's differential equation formalism. Their formalisms (which are equivalent) are what tied all of the disparate, baseless predictions of the old quantum theory together with a neat little bow we now call quantum mechanics.
But the GP should make no mistake: quantum mechanics began making useful predictions immediately. I suspect that they've simply mixed up QM and relativity, for it was relativity that was a "beautiful" theory without much experimental backing. At the time, it could do basically one thing: predict the anomalous precession of Mercury. That's why Einstein never won a Nobel Prize for his work on relativity, even though it was one of the biggest game-changers in the history of physics.