At #astrohackny, Ben Weaver showed us more development he has done on fitting the individual-exposure radial velocities in the APOGEE data. It looks to me very likely that he will be able to improve the APOGEE data analysis with these, and maybe also discover some cool things.
After the hacking, I spent time with Price-Whelan and Johnston talking about chaos. I admitted that I have no idea what constitutes a chaotic orbit! There are differences, it appears, between diffusion in frequency space and Lyapunov times; these aren't in contradiction, exactly, but the differences are surprising given my (trivial, incomplete) intuition. Something is odd to me about the related but seemingly very different points that a chaotic orbit is 5-space-filling (not just 3-torus-filling) and that it has a positive Lyapunov exponent and that it diffuses in frequency space. These all seem like three very different manifestations of chaos.
Kopytova and I spoke about fitting brown-dwarf spectra. We identified a set of easy projects, one of which is to fit the spectral data with a flexible calibration (or, equivalently, continuum) model. For the longer term, we talked about making data-driven improvements to the models, which seem to be incomplete or in need of work in the H band.