R Shiny {golem} – Development to Production – Overview

This blog series follows the development and creation of an R Shiny application. For the purposes of keeping this focused on software development, we’ll be concentrating on Shiny rather than the business use case. As a background story, we’ll be creating an app for the hit TV show, The Office.  We’ll assume the following scenario:

NBC hired us to build a tool that will help analysts dive into data. This includes viewer satisfaction, episode ratings, script text mining, personnel involved in filming and costs of advertising.

The major reasons we chose Shiny for this project are:

• Customization of the user interface

• Simplification of the data processing

• Interaction and drill-down capabilities

• Efficient algorithmic data analysis

• Publication and deployment simplicity

For this particular project, we will be using the golem package to build our Shiny app. The golem framework consists of clean and distinct code that makes an app strong and sound to create scalable development for the future.

The final product will be: 

• Tested thoroughly

• Well documented

• Organized

These qualities and features of golem will make our app more resilient than a basic shiny app. 

Are you convinced this may be the greatest app ever? Stay tuned for future posts in this series, where we highlight planning, production, testing, pitfalls and more! 

“I’m not superstitious, but I am a little stitious.” – Michael Scott

Overview of what we will cover:

1. Overview (this post)

2. Data Discovery and Designing the UI

3. Getting started with golem

4. Setting your project up for success

5. Development process and testing

6. Preparation for CI/CD

7. Customization of the app

8. Deployment to the web

We’ll be covering this in a future post, but if you would like to get started..

Open a terminal and run the following:

$ R
$ install.packages(“golem”)
$ golem::create_golem("theoffice")