Nowadays Physics is a very big chunck of science, and although in our University courses we try to give our students a basic knowledge of all of it, it has become increasingly clear that it is very hard to keep up to date with the developments in such diverse sub-fields as quantum optics, material science, particle physics, astrophysics, quantum field theory, statistical physics, thermodynamics, etcetera.

Simply put, there is not enough time within the average life time of a human being to read and learn about everything that is being studied in dozens of different disciplines that form what one may generically call "Physics. 

Conferences such as ICNFP 2015 - where the acronym stands for "International Conference of New Frontiers in Physics" - are basically the only chance most physicists have to try and get exposed to the recent advances in the many different sub-fields, as during one week these events give participants a broad oversight of many different topics. 

The conference takes place in Kolimbari, a seaside town in the north-west coast of Crete, and precisely in the Orthodox Academy of Crete, a wonderful venue that hosts events of this kind since 1968. The academy overlooks the Aegean sea and the land on which it is built, which was donated by a nearby monastery, inspires visitors to get away from their email browsers and engage in meditation and productive thinking.

This year I will be presenting at ICNFP an overview of the CMS experiments and its most recent results. I am going to post here the slides of my presentation and some commentary, but I will also try to cover some of the many interesting results that are shown by my 200 or so colleagues.

As I write this post, Bernardo Adeva is discussing the measurements of Bs decays performed by the LHCb collaboration. The speaker showed some interesting results, among which stands out the discrepancy of the S_5 observable in B->K*μμ decays first reported in this paper.

The local significance of this discrepancy with the SM predictions is in the 3.7 sigma ballpark. As LHCb shows results for eight kinematical observables in this decay mode, and they divide the data in eight bins in q^2 (the squared energy scale of the process), it would be fair to expect that, for data coming from SM processes alone, one of these 60-or-so bins would depart from the predictions by two sigma or so. In other words, there is a "look-elsewhere effect" at work, and I was not able to understand from the speaker's answer to my question after the talk whether LHCb does evaluate a "global significance" for the agreement of the eight observables to SM predictions. If I understand correctly (please correct me if I'm overlooking something), the 3.7 sigma correspond to a rather smaller global discrepancy once you look at the big picture.

Shortly later, the same speaker also discussed the new Pentaquark-like signal found by LHCb in J/psi plus proton final states. And again I felt the urge to interject. The speaker quoted a significance of 12 standard deviations for the signal, "once systematics were accounted for". I argued that 12 standard deviations mean that one claims to know the tails of one's systematic uncertainties (and their shape) to a part in ten to the thirtieth or so. That is ridiculous of course! I pointed out that significances above 5 or 6 units should not be spelt out, as the quoted number has no real meaning - or if it is claimed to have one, it is definitely a wrong claim.

But alas, I see that the majority of my colleagues still do not get it - we are physicists, but we should know the language we speak when we report our results - while we do not: very few of my colleagues understand what a claim of "12 sigma" means and why it is ridiculous to make it.