Automatic tuning has been an elusive goal for database technology for a long time and is becoming a pressing issue for modern E-services. This paper reviews and assesses the advances that have been made on this important subject during the last ten years. A major conclusion is that self-tuning database technology should be based on the paradigm of a feedback control loop, but is also bound to build on mathematical models and their proper engineering into system components. In addition, the composition of information services into truly self-tuning, higherlevel E-services may require a radical departure towards simpler, highly componentized software architectures with narrow interfaces between RISC-style "autonomic" components.