SQL Starter Pack
Overview
SQL or “Structured Query Language”, and otherwise pronounced as “Sequel”, is a programing language meant to interact with databases. In simplified terms, the language allows users to retrieve, edit, and add to a database.
History
SQL’s beginnings can be traced all the way back to 1974 from Donald D. Chamberlin and Raymond F. Boyce, two programmers who worked at IBM. Initially, their work with relational database management had them using the Square programming language. This language was short lived however because it was too difficult to use! As a result, the two aimed to develop a language that was both functional and didn’t have a high skill ceiling. And while their version of SQL has been iterated upon countless times by several people and organizations, the foundations of this programming language have quite the long history.
Popularity
Among data analysts/scientists SQL remains one of the most popular programming languages for interacting with organized databases. Use cases can apply to almost every corner of the market that collects data. Whether it is a social media company managing millions of users interaction data, a utility company logging customer information, or a store managing transaction information, the management of this data needs to strike a balance between power and usability.
Luckily, SQL, in its many forms addresses these concerns. Let’s take a look at an example highlighting the strengths and weaknesses of SQL.
Extracting/filtering information from database table
As you can see, SQL can be particularly useful in the extraction and light editing of data. Furthermore the syntax used is much more legible than Python. However, while SQL can be used for higher level analyses and transformations, this paradigm does a complete 180. Python, with its large catalog of flexible libraries makes higher level functions (regression, feature engineering, etc.) much easier to implement. And furthermore in regard to which language is faster, it depends. While there are an extreme amount of variable that can change the answer one way or the other, generally speaking the closer you are to the data the more likely SQL will be faster (Filtering, extracting, etc.). The more complex your manipulations become, Python will be faster and easier! Do not google how to do a regression analysis in SQL…
“SQL is a powerful tool that can be used for almost anything. That doesn’t mean that you should.”
