Got Security?

John Preskill points me to Workshop on Classical and Quantum Information Security to be held December 15-18, 2005 at Caltech:

This workshop will bring together researchers from a variety of backgrounds who work on different aspects of classical and quantum information security. Participants will strive to identify issues and problems of common interest that can be effectively addressed by pooling their expertise.

Looks like a very broad spectrum of researchers will be attending, which should make for a very interesting workshop. Overcoming communication barriers is often the first step in interesting new research directions, and I’m sure one hope of the workshop is to aid this difficult process.
Which reminds me of a documentry I saw a few years ago which tried to identify why certain schools were much better at producing (not buying 😉 ) Nobel prizes. What the documentry argued was that the common theme for the top Nobel producing institutes was that they all had very loose definitions of what was considered “in a particular discipline.” Or, the way I like to think about it: the energy landscape on the space of different research programs had very shallow minimums at the top Nobel producing institutes, whereas at most other schools, these wells were very deep…sometimes being even infinite square wells! This is, of course, an oversimplification, but I’m glad to see that Caltech (highest Nobel prize count per member of the community that I know of ) is hosting a workshop which so well imbodies this spirit.

Marilyn >> Newton!

Hawking interview in the Guardian. Funny:

“If you could go back in time, who would you rather meet, Marilyn Monroe or Isaac Newton?” and after 10 minutes he says in that voice that makes the blandest statement sound profound: “Marilyn. Newton seems to have been an unpleasant character.”

The interview reminds me of attending a physics talk by Hawking at Caltech. What was awesome was that after the talk, there was a question from the audience. And then, because it takes Hawking a long time to compose his answer, the big wig professors at Caltech began to debate what exactly Hawking’s answer should be. Wow, very fun stuff to witness! Then after about 15 minutes came Hawkings answer. We all waited in antipication for that famous digitized voice. His answer to the question: “No.”

Dirac >> Feynman?

Information Processing points to a review by Freeman Dyson in the New York Review of Books of Perfectly Reasonable Deviations From The Beaten Track: The Letters Of Richard P. Feynman .
What I find interesting in the article is Dyson’s claim that Paul Dirac was a greater genius than Richard Feynman. Of course, judging “greater genius” seems about as silly as worrying about whether it is a “double backward doggy dare” or a “triple backwards over-the-top doggy dare.” With this caviot, I however, just have to say “huh?”
In my mind Dirac has three claims to genius: his derivation of the Dirac equation, his work on magnetic monopoles (showing that the existence of a single such monopole could explain the reason charge comes in discrete quantities), and his work unifying the early differing approaches to quantum theory. Feynman, in my mind, has four or more claims to genius: his derivation of the path integral formulation of quantum theory, his space-time approach to solving the problems in quantum electrodynamics, his work on the theory of superconductivity (showing the importance of quantum theory on a “macroscopic” level), and his model of weak decay (work with Gell-Mann which was also independently done by George Sudharsan and Robert Marshak.) So in my mind, I put Feynman just above Dirac (what, you mean you don’t have your own personal ordering of geniuses?)
And, after thinking about it for a while (too much time, perhaps!) I think I can guess why Dyson puts Dirac above Feynman (oh, to be a physicist known by your last name alone!) I believe the reason is that Dyson was originally a mathematician. Feynman’s work is filled with the sort of raw physical insight that physicists love and admire. Sure, making the path integral rigorous is a pain the rear, but it works! In Dirac’s work, we find, on the other hand, a clear mathematical beauty: the Dirac equation and the magnetic monopole are motivated more my arguments of symmetry than by any appeal to a physicist’s “calculate and run” methodologies (indeed the latter is not even known the correspond to experimental reality!)
So who is the greater genius? Well I “double dog dare you” to come up with reasons that Dirac is a greater genius than Feynman.
Update: See the comments for some fun back and forth. OK, in my head really I put Dirac and Feynman at the same level. What I find intersting is how one’s background influences this (silly) debate. If you are a particle theorist, I bet Dirac>Feynman. If you went to Caltech as an undergrad, I bet you have Feynman>Dirac. Ah, the ways theorists waste away their days.

Angel Dust, Ozone, Wack, and Rocket Fuel

Course I’m sitting in on this term (will probably have to miss two lectures, drats!):

CSE 533: The PCP Theorem and Hardness of Approximation, Autumn 2005
Instructors:
Venkatesan Guruswami and Ryan O’Donnell
Meeting times: MW 10:30-11:50 at Loew Hall, Room 222
Course Description
The PCP Theorem states that there is a certain format for writing mathematical proofs, with the following property: a verifier can randomly spot-check only a *constant* number of bits of the proof and yet be convinced of its correctness or fallaciousness with extremely high probability. Further, proofs in this format need only be polynomially longer than they are in “classical” format, and can be produced in polynomial time from the classical proof.
The proof of this theorem in the early 1990’s was a great triumph of theoretical computer science, and it became even more important in the mid-1990’s when it was shown to be extremely powerful in proving NP-hardness results for *approximate* solutions to optimization problems.
Unfortunately, the known proof of the PCP theorem was extremely complex and also decentralized; even complete courses devoted to it rarely finished all of the proof details. However in April 2005, Irit Dinur completely changed the PCP landscape by giving a self-contained proof taking only a dozen or so pages. (See http://eccc.uni-trier.de/eccc-reports/2005/TR05-046/index.html)
In this course we will prove the PCP Theorem and also gives some of its consequences for hardness of approximation. Topics will be a mix of older work from the 1990’s, as well as very recent results on the cutting edge of research, such as:
Interactive proofs
The Parallel Repetition Theorem
Hardness of approximation for linear equations, clique size, maximum cut, set cover, and lattice problems
Fourier analysis of boolean functions
The Unique Games conjecture

Hopefully I will have time to blog about some of this, as it is probably some of the most beautiful stuff in computer science. Even if it does share a name with a drug.

QIP 2006

The Ninth Workshop on Quantum Information Processing, i.e. QIP 2006, has its announcement out. Although, strangely the three emails they sent me all crash pine! QIP will be in Paris this year. They’ve switched to making the process of selecting who will talk competitive (computer scientists, I say!) In addition to longer invited talks, there will be long (?!) 30 minute talks and short (I’ll say!) 10 minute talks from contributors. Unfortunately I won’t be able to make it this year due to other obligations. Drats, I’ve always wanted to spend time in Paris!

Subject: QIP’06 Call for Communications
This is the first announcement for QIP 2006, the Ninth Workshop on
Quantum Information Processing, to be held in Paris, France, from
Monday January 16th until Friday January 20th, 2006. You can find more
information at the conference website at http://www.lri.fr/qip06 .
The program of QIP 2006 will be organized somewhat differently from
the previous QIP workshops. Besides a reduced number of 45 minutes
invited talks there is a Call for Communications at
http://www.lri.fr/qip06/call.html for long (30 minutes) and short (10
minutes) contributed talks. Contributed talks will constitute the
major part of the scientific program.
We solicit your submissions to QIP 2006. Note that the deadline for
submissions is November 3, 2005.
During the workshop there will also be a poster session, a business
meeting, a conference dinner and additional social activities. There
will be a nominal conference fee.
See you in Paris!
The organizers
Miklos Santha, Christoph Durr, Julia Kempe, Sophie Laplante and Frederic Magniez
(http://www.lri.fr/quantum/)

Mach-Zehnder Gone Bad

Posted, without comment, in order to protect the identity of certain “experimentalists” aiding a “theorist” to take pictures of a theorist’s conception of a Mach-Zehnder interferometer:
Mach Whah?

Singapore Airport

I just couldn’t resist free internet access at 5am in the morning in the Sinapore Airport.
Update: Nor could I resist internet access at the Tokyo airport.

Gamma Watch

Just what every physicists has dreamt of: a watch with a Geiger counter in it! Of course, you can tell it is for physicists as well by the lack of style!

Science Fiction Clouds My Judgement

Strange stars:

Stars race around a black hole at the center of the Andromeda galaxy so fast that they could go the distance from Earth to the Moon in six minutes.
The finding, announced today, solves a mystery over the source of strange blue light coming from Andromeda’s center. But it generates a new puzzle: The stars’ phenomenal orbital velocity suggests they should never have formed in the first place.
Astronomers first spotted the blue light near Andromeda’s core in 1995. Three years later, another group determined that the light emanated from a cluster of hot, young stars. Nobody knew how many were involved.
New data from the Hubble Space Telescope reveal more than 400 blue stars that formed in a burst of activity roughly 200 million years ago, astronomers said.
The stars are packed into a disk that is just 1 light-year across.
That’s amazingly compact by cosmic standards. A light-year is the distance light travels in a year, about 6 trillion miles (10 trillion kilometers). The nearest star to our Sun is about 4.3 light-years away.
Unlikely setup
“The blue stars in the disk are so short-lived that it is unlikely in the long 12-billion-year history of Andromeda that such a short-lived disk would appear now,” said Tod Lauer of the National Optical Astronomy Observatory in Tucson, Arizona. “We think that the mechanism that formed this disk of stars probably formed other stellar disks in the past and will trigger them again in the future. We still don’t know, however, how such a disk could form in the first place. It still remains an enigma.”

The problem with science fiction is that whenever you read articles like this, and you see some strange configuration of stars in a cool locale, you immediately think…”Aliens!”