is false. In a similar manner, the closed-world assumption forces us to conclude
that the statement
Donald is a duck.
is false. Thus, the closed-world assumption has led us to the contradictory conclusion
that although at least one of the statements must be true, both are false.
Understanding the consequences of such innocent-looking meta-reasoning techniques
is a goal of research in the fields of both artificial intelligence and database,
and it also underlines the complexities involved in the development of
intelligent systems.
Finally, there is the problem, known as the frame problem, of keeping stored
knowledge up to date in a changing environment. If an intelligent agent is going to
use its knowledge to determine its behavior, then that knowledge must be current.
But the amount of knowledge required to support intelligent behavior can be enormous,
and maintaining that knowledge in a changing environment can be a massive
undertaking. A complicating factor is that changes in an environment often
alter other items of information indirectly and accounting for such indirect consequences
is difficult. For example, if a flower vase is knocked over and broken, your
knowledge of the situation no longer contains the fact that water is in the vase,
even though spilling the water was only indirectly involved with breaking the vase.
Thus, to solve the frame problem not only requires the ability to store and retrieve
massive amounts of information in an efficient manner, but it also demands that
the storage system properly react to indirect consequences.