Myths or red herrings ?
There are two key issues with IA: the map and the territory.
The map is the language used to represent knowledge: symbolic ones will enable explicit representation of knowledge to be implemented with rules, non symbolic ones will favor implicit knowledge to be implemented with neural networks.
The territory is the context within which decisions are to be made: closed and finite ones are equivalent to playing grounds, open and infinite ones are populated with human beings.
Whereas IA systems usually combine these options, the 2*2 matrix can be used to formally define their capabilities.