Discussion:
Q: Given the duality of the canonical forms for systems that are both reachable and observable, what does it mean for a system to be one and not the other? In other words, if a system is - for example - observable, why can't the determined coordinate transformation not be used to formulate the system in controllable canonical form?