From http://arxiv.org/abs/0803.3944. The author postulates that a natural explanation for the observed metallicity gradient in DLAs can be explained via dense, metal poor disks and less dense metal rich outflows.