From the debut of Reynolds's seminal paper on flocking in 1987, flocking has been a topic of great interest to the robotics and controls community. Flocking is the quintessential emergent behavior, where a large-scale pattern emerges from the local interactions between individuals. Harnessing the power of emergence is critical for the deployment of robotic swarms. Many studies have predicted that robotic swarms could outperform individual robots in a variety of application areas. Swarm systems can also provide more robustness to hardware failure at a lower overall cost. In this work, we discuss a brief history of robotic flocking, emphasizing systems where individuals are optimizers. We also present our biologically inspired robotic flocking taxonomy. We classify flocking systems based on the global objective that it is trying to achieve, and the environmental factors that the individuals exploit. Finally, we have several practical examples of optimal flocking in engineered systems.