...der PRL-Veröffentlichung, an der ich beteiligt war, über CP-Verletzung:
This is a triumph of presentation over content. I thought that Physical Review and Physical Review Letters were supposed to have some standards of novelty... Everyone who has worked on quark mass matrices for a couple of years - and Graham Ross has been doing it for decades - knows that theories with strongly unequal quark masses, as observed, almost always give small mixing angles, which place a strong limit on the amount of CP violation. The 'new' papers don't tell more than people knew already. In the 2006 paper Donoghue et al say: "We use statistical techniques to show that the observed masses appear to be representative of a scale-invariant distribution, rho(m)~1/m. If we extend this distribution to include all the Yukawa couplings, we show that the resulting Cabibbo-Kobayashi-Maskawa matrix elements typically show a hierarchical pattern similar to observations. The Jarlskog invariant measuring the amount of CP violation is also well reproduced in magnitude." This is virtually identical in content to the findings being publicized by Gibbons and colleagues. Why would they think an incremental improvement on previous work was worth writing to Phys.Rev.Lett. about? We have their statement "... [Our main] statistical observation seems to open up the possibility that the same mechanism that is responsible for the apparently unlikely hierarchy in quark masses might also explain why the observed value for |J| is so small." Anyone who had read the most basic paper on quark mass models - say Froggatt-Nielsen 1979 - and thought for half an hour knows this already. (Incidentally Froggatt-Nielsen also gives a very simple and clear understanding why quark masses could follow a power-law distribution.) The idea that Gibbons et al. can be 'opening up a possibility' that has been around for 30 years, and still works rather well today, is nonsensical.