My last talk at Flash Brighton was (for me) a discussion about how other people used state machines in as3, basically because I was dissatisfied with the limited way that I was using mine.
It turned out they used theirs in pretty much the same way. As a backbone for simple staged processes such as bootstrapping, asset acquisition and shutdown.
And that was it.
I was a bit frustrated, my original intent for using a state machine was to completely model my controller, so I would have a map of processes through out the entire application. But I kept on hitting the same problems preventing me from doing this.
Coming out of the talk, I decided that I would set myself a challenge: to change the way I used my state machine, and map all my commands through it.
I succeeded (eventually).
In this session I shall highlight the problems involved and the very simple changes of perspective it took to overcome them.