However, the article in Nature , goes on to say
OPERA expects the new result to rule out uncertainties due to the long timescale of the proton pulses. But concerns about the experiment’s use of the Global Positioning System to synchronize clocks at each end of the neutrino beam are unlikely to be as easily allayed, The use of GPS is novel in the field of high energy and particle physics and the same system was used for both the original experiment and the new run. Hagner also adds that she’d like to see the time measurement checked using another part of the detector, to increase confidence further.
So, apparently the Special Relativistic error in GPS may not have been accounted for. My question is simple ... why not just shoot a normal, everyday beam of light from a laser, measure that and get a baseline. If you have consistent error in your measurements, then you have something to work with. Then, try the 'magic' pulse and see if the result is the same. I like things to be simple and nearly foolproof. But, I'm just an engineer; not a physisist.
The fact that the calculated error on the pulse is 32ns, and the "faster than the speed of light" measurement is 64ns (2x the calculated error) is too close to the actual difference to be pure random chance.