It seems that there is some amount of cognitive dissonance that occurs when respondents hear about both mischievous hackers and criminal hackers. There are two ways that respondents resolve this: the simplest way to resolve this is to believe that some hackers are mischievous and other hackers are criminals, and consequently keep the models separate; a more complicated way is to try to integrate the two models into one coherent belief about hackers. This latter option involves a lot of effort making sense of the new folk model that is not as clear or as commonly shared as the mischievous and criminal models. The `contractor' model of hackers is the result of this integration of the two types of hackers.