I would like to stimulate a discussion on imagining us already in a world where we'll
have MC@NLO/POWHEG for all interesting processes (SM and BSM)
and a few (key) parton-level NNLO calculations (Z,W,ttbar inclusive jet,...) at the LHC.
How the exp results should be given in the paper for "standard candle measurements"
(with cuts or extrapolated to the total x-sec)? How the TH information should be
used by the exp? How the theorists should calculate
total rates (this is less trivial than one could think, as we have recently experienced
in the Higgs-NLO x-sec working group) and compare with data?
Is there, in principle, an optimal way to convey information out
of an experiment which will be maximally useful for the all community
(and PDF groups) ?