By Julie Lowndes | January 28, 2019
Introducing the Champions program
This article was originally published on medium.com
By engaging, empowering, and amplifying scientists with cross-cutting open data science approaches, Openscapes aims to create a more efficient and collaborative open culture in environmental science, to enable better science in less time. Openscapes centers around the Champions program, which is a mentorship program that empowers environmental scientists with open data science tools and grows the community of practice.
There are many things to say about the Champions program, much more for a single post, and there will be more posts to come. Here, I’ve focused on:
- describing the operations of the program (modeled after Mozilla Open Leaders)
- describing the content of what we’re teaching (modeled after Lowndes et al. 2017)
- giving an update of how it is going so far since launching earlier in January (one month in)
“Champions” are early career environmental researchers in leadership roles – for example: faculty, lecturers, and program managers. Recognizing that early career scientific leaders a) often had no formal computing or data training b) do not have time to learn All Of The Things c) need ways to manage all the data and analyses within their labs in way that also streamlines onboarding and offboarding lab members, the program’s aim is to provide guidance, paths, community, and agency for Champions and their labs to incrementally engage.
Openscapes Champions is a five-month mentorship program that is led remotely. Each month, there is individual mentorship for each Champion, and two “Cohort Calls” where we convene all Champions together, along with their lab members. Throughout, all participants learn about open practices and data science; strategize how to engage; meet other labs; set, review, and meet goals; and stay accountable. For each 1.5-hour Cohort Call, I teach for a maximum of 40 minutes, so that we spend most time discussing these topics in the context of their labs.
So how do you lead a remote program like this? You model it after Mozilla Open Leaders, which has developed a great program that they have repeatedly tested and improved over seven rounds. It takes intense coordination, deliberate focus on creating a positive and inclusive culture, design to make calls engaging and interactive, Zoom, and Google Docs. (I think this is really cool how the intentions of the program are realized with the software tools available — you will see this as a theme throughout). I will focus on the software here, and have a whole separate upcoming post on culture.
Zoom and Google Docs
Participants all join Cohort Calls through Zoom, a video conferencing software that we use to interact in real time and also record the whole call as an archived resource for participants. In addition to accommodating large groups for discussions, it also has a great breakout room feature. Breakout rooms allows all participants to not only listen to lessons and discuss as a full group, but also lets them break into small groups to discuss topics. Following breakout groups, participants rejoin and summarize to the group both verbally and in writing, using Google Docs.
Each call has an agenda written as a Google Doc to enable real-time participation. During the call, all participants have the agenda open, and follow along, writing their thoughts in specific places and participating in active shared note-taking. This is an example pre-call agenda. Following the real-time writing and engagement during the Cohort Call, the agenda serves as a record of the discussions had that participants can revisit and anyone who missed can review.
The program architecture (both conceptually and physically with software, agendas, and planning) that Mozilla has developed is repeatable and scalable to serve future cohorts – as well as different disciplines and communities. The way the program can grow is clever: the coordination and content can be led by a small team that welcomes others from the open community to donate 1-2 hours of time per month to provide individual mentorship. This makes it possible to support more participants: the program could not scale if it required that I provide individual mentorship for all Champions; I could not support more than 7-10 Champions at a time. Mozilla Open Leaders itself started off with one person — Abby Cabunoc Mayes — mentoring ten participants, and now in its seventh round, serves six concurrent cohorts of participants with a community of mentors (many of whom were mentees in previous rounds). With conservative growth assumptions, I have modeled that within three years we could reach over 7500 scientists through Openscapes.
For future cohorts, Openscapes candidates will nominate themselves and there will be a formal selection and review process. In this inaugural pilot cohort, I selected Champions with whom I had already been having informal discussions on these topics for months or years.
So what is the meat of the Cohort Calls? I am developing a lessons series that is framed around a scientific publication that I led with my team: Our path to better science in less time using open data science tools. The paper is a retrospective of how the Ocean Health Index workflow changed over four years; the Openscapes Series aims to help guide others to incrementally make progress themselves, no matter where they are starting from and where they hope to get. The first half of the series focuses on efficiency and open culture within the lab, and the second half is about sustained learning and bringing these practices to the broader campus community.
I have been writing my plans down in the Champions Series, which serves as a free, open-source curriculum. It is currently in heavy development (and I am of course behind) but it is openly available as I progress. It is a resource for the Champions program, for stand-alone workshops that are currently in design, and available for anyone to discover, use, and remix. I don’t scroll through the Series as I teach, but I instead create slides from it; the Series serves as my lecture notes.
And again, software really helps me be coordinated and share my work as I go along.
Bookdown and blogdown
I’m writing the Series with bookdown, which is created by Yihui Xie at RStudio and is a way to organize and format RMarkdown documents into books you can read on your computer or device. Using bookdown lets me share my progress as soon as I’ve written it. I’ve written other bookdown books for open data science trainings and really advocate for others to use it to write lab protocols, lessons, whole theses — I certainly wish I could have written my whole PhD thesis with bookdown!
The openscapes.org website is another place where the Champions program has a home online. I developed this website with blogdown, which is also which is created by Yihui Xie at RStudio and is a way to organize and format RMarkdown documents into websites with easy blogging capabilities. (Note: there are less involved ways to create websites with RMarkdown, I’ve made a training about basic RMarkdown websites too).
It is really empowering that I, who am not a web designer or publisher, am able to make and maintain beautiful website and books. It really clarifies and provides structure for my thinking and communication. And, that I am able to do so without a lot of additional software, since both blogdown and bookdown are based in R, RStudio, RMarkdown, git, and GitHub, which are already the tools for my analyses and scientific workflows. Thank you Yihui and the RStudio team!
One month update
The Openscapes Champions program began in January and has already had our first one-on-one mentorship calls and two Cohort Calls. They seem to be going well! On the second call we had 15 lab members in addition to the 7 Champions and there has been good discussions and feedback. Some good questions have come up, like “how do you help people learn from each other in the lab, particularly with lots of turnover?”, “how do you reduce the time it takes to get back into a project from months ago?”, and “what’s a good way to set up GitHub for the lab?” Each call we end with some “efficiency tips”, which so far have been keyboard shortcuts that are incredibly satisfying time-savers.
I’ve learned that I need to leave even more time for discussion, both as a group and in the breakout groups. Having the time and place to think and talk about these issues is valuable, and I need to balance it and become more compact with what I want to teach. While I have a plan for the Series, I am constantly trying to incorporate needs and questions into it so that this experience is as valuable as possible. Everyone has been really patient and supportive and I will be busy improving everything as we go along.
We have started cataloging the labs’ approaches to reproducibility, collaboration, and communication, which we will use to help identify places for incremental improvement to prioritize next steps. More on this in another post! And I will also introduce the Champions and their labs, who have been really great and working on exciting and critical environmental questions.
By mentoring early career researchers in leadership roles, Openscapes helps create more champions for open, collaborative data practices. Openscapes complements existing efforts in this space (e.g. rOpenSci, RLadies, RStudio, The Carpentries) by working with scientists who have not yet engaged in open data practices, meeting them where they are, and facilitating their engagement with existing open tools and communities.
We are excited to help grow the community of practice by mentoring and supporting scientists to champion open practices in their labs and departments!
Photo by Elliot Lowndes