The issues of how scale, aggregation and generalization affect how we model our world and then take measurements of those models that we often accept as truth do come up more often than you might think.
The length of the coastlines thing was just referencing an old academic paper on the subject.
A different application of the same subject would be how voting precinct boundaries (scale and aggregation) can affect the outcome of an election.
Related to that is the example is what Livecoug references above regarding what defines the population of a "city."