This chapter looked at Haskell data types from the perspective of data analysis. You are responsible for the conversions of data within Haskell. This is both a blessing and a curse. Haskell will never manipulate your data without your consent, even in those moments where some data manipulation can be forgiven. We also explored the task of navigating through three common data sources—the Haskell command line where you enter data yourself, csv files, and SQLite3 files. The command line is limited since we can only type so much data ourselves without getting tired. CSV files are the most common source of datasets found on the web. We also explored the difficulties in working with csv files; everything in a csv file is a String type and you, as an analyst, have to be dependent on the metadata that accompanies a dataset. We also explored SQLite3, which allows us to leverage two powerful worlds into an environment—functional programming and SQL. Using SQL will give us wonderful versatility over csv files, as we will demonstrate in forthcoming future chapters.

The next chapter will look at the necessary task of cleaning and organizing our datasets. We will look at Haskell's regular expression library in order to filter data based on specific properties of fields.