I must offer apologies for not posting per diem as I tried to do after reading a doomsday piece about the power law associated with blog readership, but I have two separate presentations about computer security and the relationship between magmatism and crustal extension this week, a possibly fantastic statistical association. This never normally happens but I manage to inadvertently cause it, so I can't complain.
Now---
Nova aired last week "The Ghost Particle", simply the best episode of Nova in the past five years. This was a science story that explained neutrinos and Pauli's proposal for them in the excess energy from beta particle (aka electron) emission, the first detection from a nuclear reactor from Reines and Cowan, and the fantastic Homestake Mine experiment from Ray Davis Jr., and John Bahcall's theoretical calculations that pushed the experiment foreward into the modern three flavored neutrino theory. Davis' son, Andrew, is a scientist here at Chicago who cofounded the Chicago Center for Cosmochemistry.
P.S. The link to the Ghost Particle goes to the producer's page rather than the main page because he lists the rules in making a compelling story, although he believed in possibly keeping it too simple--keeping the idea of a neutron out of the atom's nucleus in the animations for the nuclear reactions was in my opinion unnecessary. But still, bravo to you, a fantastic Nova.
I'm not sure I like Kottke's plot. I don't think that plot actually shows the power law directly. Two complaints: (1) he should plot size on the x-axis, and number of blogs on the y-axis; (2)he should plot number of blogs in a given range rather than the rank number of a given blog.
ReplyDeleteThinking about it, though, I've convinced myself that his plot is consistent with my understanding of a power law. flipping the axes won't change the power law, and using blog rank number is equivalent to taking an integral. And the integral of a power law is still a power law (though with a different value of the power). And he should have plotted it on a log-log scale so it came out to a straight line.