[Boost-users] statechart comparison

Is it still the case that boost statecharts require allocation and destruction of a state object when entering and exiting that state? I'm curious, as I'm comparing boost statecharts with Samek's. Heinzmann provided some commentary on the differences: http://www.state-machine.com/resources/Heinzmann04.pdf Samek's QHSM implementation does not require allocations for entering and exiting states (specifically, QHSM states are not object instances, rather they are represented simply as a pointer to a member function). Of course Heinzmann was also promoting his own personal implementation...which I don't think he ever finally published or made available for download.