"Turbulence is the probably most complex and at the same time most relevant example of spatio-temporal disorder in nature. The transport of heat and mass in stars, the formation of planets, as well as flows in the earth atmosphere, oceans or around vehicles are all governed by turbulence. Despite its ubiquity our insights into this complex phenomenon are very limited. In contrast to many studies which are concerned with turbulent flows at high parameter values I will here use a different approach and investigate turbulence when it first arises and where it is the least complex. I will focus on canonical shear flows, comprising pipe, Couette and channel flows. I have recently determined the critical point for transition in pipe flow, which had posed a riddle for more than a century and inhibited further progress towards a fundamental understanding of turbulence close to onset. At first I will clarify if this transition generally applies to all canonical shear flows. Next I will explore links to non-equilibrium phase transitions in other areas of science by determining the critical exponents and the universality class of the onset of shear flow turbulence. I will investigate and identify further bifurcations the turbulent state experiences as it develops from a spatially intermittent to a space filling state. This will for the first time provide a complete picture of the onset of turbulence and establish links to turbulence studies at higher Reynolds numbers. Investigating the mechanisms leading to fully turbulent flow will not only give valuable insights into the nature of fluid turbulence but may also lead to new ways to control it. Finally I will exploit these insights and devise methods that completely relaminarize turbulent flows. Subduing turbulence is of great practical importance since frictional losses in turbulence are much larger than in the laminar state and hence relaminarization leads to substantial energy savings in transport problems."