We prove that previous claims of observing a direct superfluid-Mott insulator transition in the disordered J-current model are in error because numerical simulations were done for too small system sizes and the authors ignored the rigorous theorem.
@article{arxiv.cond-mat/0310251,
title = {Comment on "Phase Diagram of a Disordered Boson Hubbard Model in Two Dimensions"},
author = {Nikolay Prokof'ev and Boris Svistunov},
journal= {arXiv preprint arXiv:cond-mat/0310251},
year = {2007}
}