Hoi Henning, Ed will correct me if/when I am wrong ... > so if I understand Urs correctly, we could run with the current > IslBrecoilUser over run 3 data, writing out the usual "big ntuples"? Not exactly, as run 3 is release 12, whereas our old executable was analysis-14, i.e. a 10-series. But that change is manageable, Ed has done it already. An added benefit is the 700/fb of generic BB. :-) Cheers, --U.