Current status

To view the (entirely anonymous) aggregated stats/datavis/data exploration for the study, visit: https://sustainable-open-science-and-software.github.io/survey-datavis/

The source code for the above analyses is here: https://github.com/yochannah/sustainable-communities-tracker

Note that some of the pages have had to be removed as they would too easily reveal which github repos were associated with study participants.

History

The thesis is in; preprint & publication will be forthcoming. The TL;DR of this study is: don’t use GitHub metrics without context, as similar looking projects may have vastly different reasons for the same github metrics.

Update May 2023

Thesis due in one month. All hands on decks!

Update Nov 2022

Data analysis in progress, survey dataset is pseudonymised and deposited on Zenodo.

Still todo:

  • Finalise Analyses
  • Pseudonymise GitHub metrics
  • Write up (as a thesis chapter…)
  • Deposit preprint
  • Send to publisher (who?!)

Update JUNE 2022:

Final surveys have been sent out to participants (40 in total), and write-up/data analysis is in progress.

Update NOV 2021:

THIS SURVEY IS NOW CLOSED** Interested in participating? Please visit this survey for more info and to get started. It takes around 15 minutes to complete. Alternatively, you can download a PDF copy of the participant information sheet (the sheet is also shown in the survey if you are an eligible participant).

About the study

Research software is often abandoned or shut down, for one reason or another. While some reasons may be straightforward, e.g. a sole maintainer has moved on, or grant funding has ceased - some projects are able to withstand these barriers and may remain active and maintained despite adversity.

I’m running a study where I plan to monitor open source projects over the period of a year, measuring common performance indicators, to see if any indicators are common to projects that remain sustainable and active.

During this period, I’ll be doing the following:

  1. Running an initial survey, where I gather info about the project age, leadership, and GitHub URLs. Participants will be asked to add a short notice to their readme.
  2. After the initial survey, I’ll also gather information about the GitHub projects - this is info such as number of contributors, number of PRs, time taken to close/merge these PRs, and issues closed. Some of this info will be gathered using scripts, such as tools from GrimoireLab and other parts will be gathered manually. An example of this is the Code of Conduct - while I can programmatically check for the existence of CodeOfConduct.md, I can’t easily check for enforcement contacts without manual checks.
  3. 6 months [Nov/Dec 2021] and 12 months after the initial survey [May/June 2022], I’ll send out a follow-up survey and run follow up GitHub checks, to see changes in projects over time.