By using the AdS/CFT duality, the computation of MSYM scattering amplitudes at strong coupling boils down to the computation of minimal areas on AdS_5 with certain boundary conditions. Unfortunately, this seems to be a hard problem. In this talk we show how one can make progress by restricting to AdS_3.
I review the status of (open covariant) cubic superstring field theories, their successes and their problems. I then propose a new superstring field theory, which avoids previous problems. The picture number is not restricted in this theory and the NS and Ramond sectors are naturally unified. Constructing the BV master action is straightforwards and leads to a theory which is defined in the whole Hilbert space, i.e., including all ghost and picture numbers and all the relevant sectors. When (partially) gauge fixed and restricted to the NS sector, this new theory reduces to the old one. Hence, all the good known properties of the old one are shared by the new theory.
