Exploring, experimenting, and building software and trust at rOpenSci’s unconf18

  Stefanie Butland


June 5, 2018

We held our 5th annual unconference in Seattle, May 21-22, 2018 at Microsoft’s Reactor space. Researchers, students, postdocs and faculty, R software users and developers, and open data enthusiasts from academia, industry, government, and non-profits came together for two days to hack on projects they dreamed up and for an opportunity to meet and work together in person. We brought together 60 people from 11 countries in 5 continents - North and South America, Europe, Australia and Africa. Nearly one third of US states were represented!


What is rOpenSci’s unconf?

The rOpenSci unconference is our key annual event to share our vision of a culture that values open and reproducible research using shared data and reusable software. They have a rich history that began with rOpenHack in 2014. It’s called an “unconference” because there is no schedule set before the event – participants discuss project ideas online in advance and people self-select into project groups. By design, one third of participants had attended a previous unconf, and two thirds were first-timers; some were pre-invited and the majority were selected from over 250 applicants; half of the participants were women or other underrepresented genders. Next year we need to do a better job being proactive on diversity beyond gender.

It was truly an interdisciplinary crowd with people working or studying in ecology, public health, epidemiology, biomedical science, clinical research, cognitive science, health psychology, bioinformatics, geospatial science, statistics, data science, education, computer science, software development, software as a service, social sciences, and cyber.


Building a trust network

Great code and documentation has far less value without a strong community of users and contributors around it. At unconf we aim to build on our trust network where the work we do together and the relationships we build in person implicitly help us model good behavior when we’re living online, distributed around the world. This is supported by a Code of Conduct that is embraced by the leadership and the community. In many cases, an unconf participant is the sole practitioner in R or in promoting reproducible workflows in their home environment so the connections created here are invaluable. On the evening of Day 1 we invited people from the Seattle R and open science community to join us for dinner and we built out our trust network even further.

With only two days together, people needed to quickly self-sort into project groups and get working. To help first time participants know what to expect, prepare to be fully engaged and to understand that they really do belong in a group that includes some high profile contributors, I had short (and illuminating!) pre-unconf video chats with 34 people using an approach I piloted and described last year. This enabled me to make pre-unconf connections between some participants, like social scientists, people developing software for flow cytometry, or people who were visiting the same city after unconf.


Exploring and experimenting

Making unconf non-competitive gives people an opportunity to freely explore ideas and try things, from developing their first R package to making some critical SAS macros for health sciences available in R, to trying a bold thing that’s “so crazy it just might work”. The code and documentation for all projects is available at ropenscilabs, our “in-development” home on GitHub.


Sponsors

Thank you to unconf18 funders and sponsors, The Helmsley Charitable Trust, Alfred P. Sloan Foundation, Google, Microsoft, and RStudio, and our fiscal sponsor NumFOCUS for help with organization. Unconf sponsors play a unique role in that participants from Microsoft, Google and RStudio acted as helpers and participated in projects, sharing their expertise and insights.


Read more about unconf18

Projects posts

Let the developers know if you like what they did. Let them know if you’ve worked on something similar. Let them know if you’d like to contribute.

Projects recaps: mchtoolbox, pkginspector, dataspice, rOpenSciEd, rOpenInterviews - middlechild, defender, ropsec, keybase - jobstatus, motifator, QcodeR, opencv, trackmd - umapr, greta, roomba, proxy-bias-vignette, http caching

Community posts

After unconf I feel more confident in my role and more confident in being involved and contributing to the parts of the community I belong to … (Auriel Fournier)