The lack of insight into business rules relating to job postings makes
it difficult to write utility code that may reduce code duplication
across CareerBuilder / Monster / etc.

For example: company X has four similar positions to fill.  The final CSV
output should not combine postings across sites in the same file because
there's no way to know definitively whether these are unique positions
(did company X post two positions to CareerBuilder and two to Monster?
Did they duplicate all four postings on both sites?)  Assumtions about
the ways an employer creates a posting will likely result in misleading
output.

Another example: the "distance from searched zipcode" query parameter
accepts different values on Monster than it does on CareerBuilder (as
well as a different query parameter key.)  This makes it difficult to
reduce code duplication in in the model objects, and difficult to write
mindful code for future additions to the list of sites we scrape.