Implement Realistic Usage Scenario

[jdscriven]

The primary use case is storing a large amount of fielded metadata in memory.

So, for example, if we wanted to load all the worlds countries into memory (https://github.com/twigkit/worldfactbook-dataset) we would define a POJO like:

class Country{
  String code;
  String name;
  Braid region;
  Braid religions;
  Braid languages;
 ...
}

And then all the 'language' fields, for example, would be compressed in a single Knot.

This is still a very small dataset; ideally we would have something in the millions of instances rather than hundreds.