Introduction

Research Software Design by Example

The Problem

Most research software engineers are self-taught programmers
Often have gaps in their knowledge
One of those gaps is software design
A large program is more than just a dozen small programs
Most programming language features exist to manage complexity

Complexity and size — Figure 1.1: How complexity grows with size.

Comprehension

Experts understand lower and higher abstraction levels than novices
Preferred level shifts with experience
Code that is optimal for one reader may not be optimal for another

How to Learn Design

Best way to learn design is from examples
These lessons use small versions of common research problems
- Generally accessible to people who are programming
- Introduce some fundamental ideas in computer science

Audience

Maya has a master’s degree in genomics
- Knows enough Python to analyze data from her experiments
- Struggling to write code other people can use
These lessons will teach her how to design, build, and test large programs in less time and with less pain

Prerequisites

Write Python programs using lists, loops, conditionals, dictionaries, and functions
Puzzle through Python programs that use classes and decorators
Comfortable writing Unix shell scripts using for and wildcards
Read and write a little bit of HTML
Use Git to save and share files

Usage

Written material: Creative Commons - Attribution - NonCommercial 4.0 International license (CC-BY-NC-4.0)
Code: Hippocratic License
Source available in [our Git repository][config_repo]
Can all be read on [our website][config_website]

The Author

Greg Wilson has worked in industry and academia for 40 years, and is the author, co-author, or editor of over a dozen previous books. He was the co-founder and first Executive Director of Software Carpentry and received ACM SIGSOFT’s Influential Educator Award in 2020.