Introduction

Starting Point

A 10X engineer is one who can bring together ten other engineers and emerge with a shared understanding and rough consensus of the problem being solved and work that needs to be done.

— Lorin Hochstein

In this house we call them "project managers".

— Greg Wilson

Research Software

• Software that is created and run to answer specific questions
• Usual starting point:
  • The software is just a piece of laboratory equipment
  • Papers, theses, and other reports are the product
• But if the software becomes more popular:
  • Focus becomes building and sharing software that many colleagues can use
  • Like designing and building telescopes in astronomy

Meet Jess

• Doing their PhD in ecology
• Simulating changes in the ranges of several rodent species in the face of climate change
• Validating against dozens of datasets collected by previous students in their supervisor's lab
  • Which all need to be cleaned up and put in a consistent format

Three Years Ago…

• Jess started grad school
• Could write hundred-line Python programs with lists, loops, and functions using the Jupyter notebook
  • Learned it in their one and only undergrad programming course
• Knew enough HTML and CSS from a high school workshop to tweak their web page on the lab site

Their Goals

• Write and revise thesis and papers (collaboratively)
• Priority #1: get the data into a consistent format
• Also priority #1 (according to their supervisor): write the simulation program
  • Which will probably be 1000-1500 lines long
  • I.e., ten times larger than any program they've ever written

The Journey So Far

• Divide code into multiple files and use import
• Organize according to Noble's Rules Noble2009
  • Data in one folder
  • Shared code in another
  • One folder per paper/project with analysis scripts
• Shared Google Docs for papers
  • Because it's so much easier than the alternatives
  • And it gives them an excuse to talk to the lab's LaTeX expert, who's kind of cute

The Journey So Far

• MIT license for software
  • Lab standard: they just copied the file into their repo
• Using Git and GitHub with the GitKraken GUI
  • No branching and no collaborators: just a backup tool

• Managing other people isn't a problem yet
• Managing their own time is
  • Just started using GitHub issues as a to-do list

Me

• Don't remember much statistics…
• …but I've built a lot of complicated software…
• …and I've been lucky enough to hang out with some very smart people

Acknowledgments

• Daniel Standage for helping create the original workshop
• Karl Fogel for Producing Open Source Software
• Damien Irving, Kate Hertweck, Luke Johnston, Joel Ostblom, and Charlotte Wickham for Research Software Engineering with Python
• Alexandra Elbakyan for SciHub
• And everyone cited in the bibliography