Traversant la Manche

Until now, I have been forced to resist the clamour brewing among both members of my extensive readership (hi, dad!) to post a bit more often: my excuse is that, in the little over a month between early September and mid-October, I have travelled back and forth from Paris to London five times, spent a weekend in the USA, started teaching a new course, and ran a half marathon.

Ten one-way trips in six weeks is too many; the Eurostar makes it about as pleasant as it could possibly be: 2 1/4 hours from central London to central Paris by train (a flight from Heathrow to de Gaulle is faster, but the airports are less convenient and much more stressful). Most of my time in Paris was for Planck Satellite meetings, mostly devoted to the first major release of Planck data and papers next year — of course, by The Planck rules, I can’t talk about what happened. At least I have no more trips to Paris until early December (and only four or so hours a week of Planck telecons).

But in addition to three Planck meetings, I also helped out in my minor role as a member of the Scientific Organizing Committee of the Big Bang, Big Data, Big Computing meeting at the APC, which was an excellent gathering of cosmologists with computer scientists and statisticians, all doing our best to talk over the fences of jargon and habit that often keep the different fields from having productive conversations. One of my favourite talks was the technical but entertaining From mean Euler characteristics to the Gaussian kinematic formula by Robert Adler, whose work in statistics more than thirty years ago taught many in cosmology how to treat the functions that we use to describe the distribution of density and temperature in the Universe as random fields; he discussed more recent updates to that early work for much more general circumstances, the cosmological repercussions of which have yet to be digested. Another highlight was from Imperial’s own Professor David Hand, Opportunities and Challenges in Modelling and Anomaly Detection, discussing how to pull small and possibly weird (“anomalous”) signals from large amounts of data— he didn’t highlight many specific instances in cosmology, but rather gave examples with other sorts of big data, such as the distribution of prices of credit card purchases (with some particularly good anecdotes culled from gas/petrol station data).

Finally, in addition to those many days of meetings — and yes, the occasional good Parisian meal — there were a couple of instances of the most satisfying of my professional duties: two examinations for newly-minted PhDs from the Institut d’Astrophysiques de Paris and the Laboratoire Astroparticule et Cosmologie — félicitations aux Docteurs Errard et Ducout.