not the h-word!

In the morning, Sarah Ballard (Harvard) gave a nice talk about finding exoplanets with Kepler plus follow-up. She effectively noted that the candidates most in need of follow-up are Jupiter-mass and Earth-mass, the former because of confusion from blended eclipsing binaries (which look very similar in time history) and the latter because of confusion from more massive planets, which (either from blending or just bad luck) at low signal-to-noise can look like Earth-mass planets. She said that there are very few false positives among the Kepler KOIs at other masses, which is what I have been hearing from other sources (think Johnson and Kjeldsen, recently). She showed very nice models of some transit-timing variations that can be explained by multiple planets in resonance. She also said a lot about habitability while only barely mentioning that word, which was appropriate, given that what is called by the h-word is often just about mass, radius, and mean intercepted radiation from the parent star.

In the rest of the day I discussed probabilistic photometric distances with Rix and Foreman-Mackey, discussed the intergalactic medium and its detectability in different regimes with Hennawi's group, discussed Herschel data with Lang, and watched Foreman-Mackey have an epiphany about the relationships between hierarchical inference, graphical models, and the expectation-maximization algorithm.