Saturday, November 26, 2011 ... Deutsch/Español/Related posts from blogosphere

The Fabric of the Cosmos III

Quantum Leap

Just one link. If you have 53 spare minutes, you may watch the third episode of The Fabric of the Cosmos with Brian Greene on PBS.

The outline of the program below isn't supposed to be pretty or ordered...

It is about quantum mecbanics: quantum computing and bizarre "interpretations" of QM are included. Allan Adams nicely says that quantum mechanics is in a 100% agreement with everything we know. Great, thankfully, Lewin and Lloyd are saying the same thing.

Gross, Weinberg, and others appear. Many comments about the strangeness of QM. A quantum bar with a quantum billiard. Totally wrong comments by Brian that "what I do here may be an immediate effect far away". A relatively sober Brian wants to have the same drink as another Brian who is already clearly drunk while another Brian is moving around them in a fuzzy way.

Peter Galison and David Kaiser explain discrete spectra of atoms. Fathers of QM are shown. Bohr is shown as a ping pong player. Planetary model of the atom. Bohr's quantization on orbits. Galison described the guess as "unphysical". The transitions between the energy levels is why the episode was named "quantum leap".

Bohr-Einstein debates. Not too detailed. Bowling lanes rebuilt as a double-slit experiment. Incorrect statements that in the double-slit case, the electron may land at places that would be forbidden with either single slit. It's not true but the converse would be true: in the double-slit case, places that would be allowed with a single slit may become forbidden. Sylvester Jim Gates and Lenny Susskind say that waves and particles are different classically. But not in the quantum world. Weinberg says that Schródinger thought that the wave was a smeared electron. Finally, Max Born brings the probabilistic interpretation. Thankfully, it's said to be right.

Las Vegas is used to explain the probability. Casinos still now that they will win in the long run. Ed Farhi says that the world is fundamentally probabilistic and people have psychological problems with that. Max Tegmark says that without QM, the U.S. economy and all devices would stop. Well, yes, even atoms would stop. Weinberg: without QM, we would be back in dark ages or 19th century.

Bohr's view correctly presented: things are uncertain before the measurement. Einstein wanted a certainty at all times. He thought QM was incomplete. In the 1930s, Einstein thought he had found the Achilles' heel of QM. Entanglement. A typical misleading presentation of what it is: it is a damn ordinary correlation, Brian: it just uses all the possibilities for correlation allowed by the QM maths but the interpretation is exactly the same as that of an ordinary correlation. They present two distant random wheels of fortune which end up correlated. But that's exactly a wrong picture. The wheels of fortune in this setup aren't correlated but the two spins in QM are. In essence, the wheel analogy incorrectly assumes that the random generators are not allowed to obey any statistical distributions with correlations. Except that they are allowed to do so. This presentation of entanglement is normal for popular programs and books but it's fundamentally misleading.

Spooky action at a distance introduced. Alain Aspect talks about the "mystery". Weinberg uses a somewhat misleading language on "affecting", too. Many of these people simply still seem to share Einstein's misconceptions. Bertelmann's socks become gloves. Brian correctly says that there's nothing mysterious about the correlation of the gloves' chiralities. In 1967, John Clauser, an astrophysics PhD, didn't understand QM and wanted to settle the debates. He thought QM was wrong. Testing Bell's inequalities. Clauser didn't want to believe his eyes. Looking for ways to get rid of the inconvenient data. "Shocking" conclusion is that QM is right even if it disagrees with classical physics. Brian totally incorrectly concludes that a "spooky action at a distance" occurs.

Spock and teleportation. Anton Zeilinger talks about his quantum teleportation experiments in exotic islands. A blonde is teleported to NYC and doesn't even greet the female officer while leaving. Brian goes to Paris for a lunch and does greet the black officer, although not in French. The old Brian has to be destroyed, important and correct. Correct comments that it's still him because his identity is given by the information.

Lloyd is showing his weird lemonade made out of weird quantum lemons, namely a quantum computer. A misleading interpretation of a qubit that is "zero and one". The addition of state vectors is really basically "or", not "and". Quantum computers may find a path from a maze, "multitasking".

Final four minutes. Wrong comments that the "uncertainty goes away" for big objects. Wrong remarks that Niels Bohr didn't have an explanation. A wrong visual detail near escalators: Brian interacts (avoids a collision) with another Brian even though they're meant to be two terms in a wave function. Such an interaction isn't possible due to the linearity of the wave function. Proper QM is suddenly forgotten again and not presented as an alternative; the only alternative to a material collapse is a material interpretation of many worlds. Too bad. Correct comments that QM apply to everything, even the big things, follow.

It's fun to watch a document where one knows almost all the "actors" in person. Much like in the episode about time, I estimate about 1/2 of the statements in this program to be plain wrong.

The last, fourth episode, Universe or Multiverse, has already been aired as well but I don't see a full YouTube link yet. Here's a 30-minute trailer.

Add to Digg this Add to reddit

snail feedback (0) :

(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) })(window,document,'script','//','ga'); ga('create', 'UA-1828728-1', 'auto'); ga('send', 'pageview');