Previous month:
July 2005
Next month:
September 2005

Katrina’s effect on LIGO

I’d been wondering what was going on at LIGO Livingston since it’s right outside Baton Rouge, and would be affected by Hurricane (now Tropical Storm) Katrina (storm science, multimedia). In Google Maps, you can see the interferometer just north of the marker for Livingston. Anyway, there had been no word at all via email from anyone at the lab. I tried to connect to the lab website (linked above), but it was down and still is, as of this writing. Finally, the lab director sent email to all members of the LIGO Scientific Collaboration that the lab is closed, and off-limits, and its power cut due to Katrina.

I was just mentioning this lack of contact to Phil Torrone on IRC, and he mentioned that Brian Greene told him in an interview that physicists (and I would say scientists, in general) don’t do very much informal online communication, like IM, IRC, or blogs. That’s changing, though. Which would be a good thing, in my opinion. I’m very accustomed to having a rather constant “feed” of informal information, and feel very out of touch when I don’t get any contact.

Listening to New wave jacket by Polysics

Continue reading "Katrina’s effect on LIGO" »


So, the interview with Harvard Med’s Dept. of Radiation Oncology Physics Division went well. The group is doing some interesting things, research-wise. One thing I was struck by was that everyone in the group was happy to be working there, and they all got along well. I’m just waiting for their faculty to meet and decide if they want to make me an offer.

The suspense is killing me.

Probability and statistics

Probability has never been my strong point, much less when I’m running on little sleep. Anyway, a couple of questions which came up during my interview were just straightforward applications of Bayes’s Theorem. Of course, I didn’t remember the statement of the theorem, and was too flustered too figure it out with an interviewer looking on. Bleah. So, here it is:

P(A|B) P(B) = P(A, B) = P(B|A) P(A).
P(A|B) is the posterior (or marginal) probability: the probability of A occuring given that B has occured; P(A, B) is the joint probability, the probability that both A and B occur; P(A) is the prior probability, the probability that A will occur at all.

A simple example, a question which was asked of me: Suppose it is known that 20% of Americans suffer from lung cancer. Suppose, also, that 50% of Americans smoke. It is also known that of the sufferers of lung cancer, 80% are smokers. What is the probability that an American smoker will contract lung cancer? Let A be “lung cancer”, B be “smoker”, so by Bayes’s Theorem, the probability that an American smoker will contract lung cancer is

P(A|B) = P(B|A) P(A) / P(B)
            = 0.8 * 0.2 / 0.5
            = 0.32

Technorati Tags: ,

Thinking with my fingers

I just realized at least one thing coming out of a series of technical interviews in Chicago: I don’t think too well with words, especially when it comes to Unixy things. For instance, one of the interviewers asked me about awk, and then proceeded to ask me how to extract the 6th column of a comma-delimited text file. I wrote

  cat file | awk '{print $6}'
which is OK for whitespace-delimited files. To do comma-delimited, I should use
  cat file | cut -f6 -d,
This would have been completely intuitive if I were at a terminal typing rather than trying to answer a verbally posed question. I have long gone past verbalizing certain things which I have done many times. Being rather a cynic, this is what I think is “zen” at least as the concept appears in certain martial arts.

Listening to Estrella from the album “Kosi Comes Around” by DJ Koze

Technorati Tags: ,