## Saturday, October 16, 2004

### Talks at Harvard this week

Let me say a couple of words about three talks we've seen this week:
• On Wednesday night (postdocs' journal club), Toby Wiseman was describing a 1994 paper by Iyer and Wald about black hole thermodynamics in generally covariant theories. It was a very interesting talk. Wald and Iyer start with a theory described by the Einstein-Hilbert action, corrected by arbitrary higher powers of the Riemann tensor and their derivatives (as well as other fields) and use a rather convoluted mathematical formalism to define the Noether current corresponding to the diffeomorphisms associated with a general vector field. Then they argue that the first law of thermodynamics must involve the dynamical entropy whose dependence on the form of the action is well-defined. You take the derivative of the Lagrangian with respect to the rt-rt component of the Riemann tensor, and integrate it over the horizon. This generalizes the simple formula A/4G to theories with more complicated actions than just the Einstein-Hilbert action. By the way, the result seems to agree with my heuristic universal derivation of the entropy.

• On Thursday, Ari Pakman from the Hebrew University was speaking on the Duality Seminar about holography and D-branes in N=2 Liouville theory (and the corresponding cigar geometry). As far as I can say, he understands these enterprises in the d=2 cripplification of string theory quite well, even though he is still a grad student. He had to study the implementation of the ZZ and FZZ D-branes in this background, and so on. For everyone who did not know the person with the cigar on a photograph: it was Che Guevara. Ari believed that everyone had to be familiar with the photographs of this Latin American communist asshole, and this assumption was proved incorrect.
• On Friday, the family lunch seminar featured James Sparks who described a countable family of Sasaki-Einstein five-dimensional manifolds that are generalizations of the base of the conifold, and can therefore be used as factors multiplying the five-dimensional anti de Sitter space in the AdS/CFT correspondence.