Control of turbulent boundary layers has been a subject of much interest because of its high potential payoffs. The control requires, however, a thorough understanding of the underlying physics of the turbulent flow and an efficient control algorithm, both of which have been less than satisfactory. Great strides on both fronts have been made through recent advancement in computational fluid dynamics and control theories. New approaches to controller design are significantly different from existing approaches, which were mostly based on the investigator's physical insight into the flow, in that modern control theories are incorporated into the controller design. Examples include systems control-theoretic approaches such as H2- and Hinfinity- control, system identification, and adjoint-based optimal/suboptimal controls. Some of these approaches explicitly exploit certain linear mechanisms present in the wall-bounded shear flow, and their success suggests the importance of linear mechanisms in the turbulent (and hence, nonlinear) flow. Although all of these new approaches have been thus far based on numerical experiments (and yet to be verified in laboratory experiments), they have shown great promise and present a new approach to flow control. The success and limitation of various new controllers and their implications will be discussed.