Six tips for running a successful unconference

[This article was first published on rOpenSci - open tools for open science, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

Attendees at the May 2017 rOpenSci unconference. Photo credit: Nistara Randhawa Attendees at the May 2017 rOpenSci unconference. Photo credit: Nistara Randhawa

In May 2017, I helped run a wildly successful “unconference” that had a huge positive impact on the community I serve. rOpenSci is a non-profit initiative enabling open and reproducible research by creating technical infrastructure in the form of staff- and community-contributed software tools in the R programming language that lower barriers to working with scientific data sources on the web, and creating social infrastructure through a welcoming and diverse community of software users and developers. Our 4th annual unconference brought together 70 people to hack on projects they dreamed up and to give them opportunities to meet and work together in person. One third of the participants had attended before, and two thirds were first-timers, selected from an open call for applications. We paid all costs up front for anyone who requested this in order to lower barriers to participation.

It’s called an “unconference” because there is no schedule set before the event – participants discuss project ideas online in advance and projects are selected by participant-voting at the start. I’m sharing some tips here on how to do this well for this particular flavour of unconference.

1. Have a code of conduct

Having a code of conduct that the organizers promote in the welcome goes a long way to creating a welcoming and safe environment and preventing violations in the first place.

2. Host online discussion of project ideas before the unconference

Our unconference centered on teams working on programming projects, rather than discussions, so prior to the unconference, we asked all participants to suggest project ideas using an open online system, called GitHub, that allows everyone to see and comment on ideas or just share enthusiastic emoji to show support.

3. Have a pre-unconference video-chat with first-time participants

Our AAAS CEFP training emphasizes the importance of extending a personalized welcome to community members, so I was inspired to make the bold move of talking with more than 40 first-time participants prior to the unconference. I asked each person to complete a short questionnaire to get them to consider the roles they anticipated playing prior to our chat. Frequently, within an hour of a video-chat, without prompting, I would see the person post a new project idea or share their thoughts about someone else’s. Other times, our conversation gave the person an opportunity to say “this idea maybe isn’t relevant but…” and I would help them talk through it, inevitably leading to “oh my gosh this is such a cool idea”. I got a huge return on my investment. People’s questions like “how do you plan to have 20 different projects present their work?” led to better planning on my part. Specific ideas for improvements came from me responding with “well…how would YOU do it?”

Between the emails, slack channel, issues on GitHub, and personal video chats, I felt completely at ease going into the unconf (where I knew next to no one!).

-rOpenSci unconf17 participant

4. Run an effective ice breaker

I adapted the “Human Barometer” ice breaker to enable 70 people to share opinions across all perceived levels of who a participant is and introduce themselves to the entire group within a 1 hour period. Success depended on creating questions that were relevant to the unconference crowd, and on visually keeping track of who had spoken up in order to call on those who had not. Ice breakers and the rOpenSci version of the Human Barometer will be the subject of a future CEFP blog post.

5. Have a plan to capture content

So much work and money go into running a great unconference, you can’t afford to do it without a plan to capture and disseminate stories about the people and the products. Harness the brain work that went into the ideas! I used the concept of content pillars to come up with a plan. Every project group was given a public repository on GitHub to house their code and documentation. In a 2-day unconference with 70 people in 20 projects, how do people present their results?! We told everyone that they had just three minutes to present, and that the only presentation material they could use was their project README (the page of documentation in their code repository). No slides allowed! This kept their focus on great documentation for the project rather than an ephemeral set of pretty slides. Practically speaking, this meant that all presentations were accessible from a single laptop connected to the projector and that to access their presentation, a speaker had only to click on the link to their repo. Where did the essence of this great idea come from? From a pre-unconference chat of course!

In the week following the unconference, we used the projects’ README documentation to create a series of five posts released Monday through Friday, noting every one of the 20 projects with links to their code repositories. To get more in-depth stories about people and projects, I let participants know we were keen to host community-contributed blog posts and that accepted posts would be tweeted to rOpenSci’s >13,000 followers. Immediately after the unconference, we invited selected projects to contribute longer form narrative blog posts and posted these once a week. The series ended with Unconf 2017: The Roads Not Taken, about all the great ideas that were not yet pursued and inviting people to contribute to these.

All of this content is tied together in one blog post to summarize the unconference and link to all staff- and community-contributed posts in the unconference projects series as well as to posts of warm personal and career-focussed reflections by some participants.

6. Care about other people’s success

Community managers do a lot of “emotional work”. In all of this, my #1 key to running a successful unconference is to genuinely and deeply care that participants arrive feeling ready to hit the ground running and leave feeling like they got what they wanted out of the experience. Ultimately, the success of our unconference is more about the human legacy – building people’s capacity as software users and developers, and the in-person relationships they develop within our online community – than it is about the projects.

“I’ve steered clear of ‘hackathons’ because they feel intimidating and the ‘bad’ kind of competitive. But, this….this was something totally different.”

– rOpenSci unconf17 participant

Additional Resources

To leave a comment for the author, please follow the link and comment on their blog: rOpenSci - open tools for open science. offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

Never miss an update!
Subscribe to R-bloggers to receive
e-mails with the latest R posts.
(You will not see this message again.)

Click here to close (This popup will not appear again)