Katie is interviewing Jessica at 8 pm EDT tonight. Go listen, and feel free to call in.
Peter Woit has a long thread about the Navier-Stokes equations, which turn out to be still unsolved. Penny Smith withdrew her preprint after discovering an egregious error.
Via 3QD: the Lancet has a new study about Iraqi civilian casualties, concluding that there have been 650,000 excess deaths since the invasion. Unlike the previous study, this one has a halfway sane 95% confidence interval, 390-640k (as opposed to 8-194k). This includes Iraqis killed due to poor security rather than just Iraqis killed by the coalition, but it subtracts the background number of Iraqis killed by Saddam and the sanctions.