Designing Modular Software: a Case Study in Introductory Statistics

• Activity Number: 217 - Computing Making Impact: The Best of JCGS
• Type: Invited
• Date/Time: Monday, July 29, 2019 : 2:00 PM to 3:50 PM
• Sponsor: JCGS-Journal of Computational and Graphical Statistics
• Author(s): Andrea Kaplan* and Eric Hare
• Companies: Duke University and Omni Analytics
• Keywords: Interactivity; Modularity; Programming Paradigms; Reactive Programming; Reproducibility; Statistical Software

Abstract

Modular programming is a development paradigm that emphasizes self-contained, flexible, and independent pieces of functionality. This practice allows new features to be seamlessly added when desired, and unwanted features to be removed, thus simplifying the software's user interface. The recent rise of web-based software applications has presented new challenges for designing an extensible, modular software system. In this talk, we outline a framework for designing such a system, with a focus on reproducibility of the results. We present as a case study a Shiny-based web application called intRo, that allows the user to perform basic data analyses and statistical routines. Finally, we highlight some challenges we encountered, and how to address them, when combining modular programming concepts with reactive programming as used by Shiny.