pyOpenSci is now an independent community organization
I am thrilled to announce that pyOpenSci is now a fiscally sponsored project of Community Initiatives! This means that pyOpenSci now serves the community of scientific Python users and open source software maintainers. pyOpenSci is not affiliated with any specific organization. I (Leah) will now be devoting ALL of my (work ) time to the pyOpenSci project full time as the official Executive Director.
pyOpenSci’s current funding and structure
I am grateful for two years of funding from the Sloan Foundation to support this effort. Once I acquire additional funding, I will be able to hire more staff to grow the organization. I am working with a board of two right now: Tracy Teal (chair) and Karen Cranston. They both bring tremendous experience building and supporting The Carpentries, among other efforts; Tracy and Karen will guide me in propelling pyOpenSci forward. I also am working with two incredible editors that have been crucial in supporting peer review: David Nicholson and Ivan Ogasawara. In the upcoming months we will recruit more editors to support our open source software review process. Finally, I am also continuing to coordinate with and learn from the well-established and successful rOpenSci. rOpenSci has a strong precedent for open, community-driven peer review. Our community has a lot to learn from rOpenSci. I am grateful for their continued support.
pyOpenSci is not just peer review of open source software
Peer review is critical to the pyOpenSci mission. But I have pyOpenSci bigger and broader aspirations for pyOpenSci, too. There are several organizational goals supported by the Sloan Foundation grant that we will be working on over the next few years:
- Create and grow a diverse and supportive community around a dynamic and open peer review process and Python open science skills. Diversity is important and we plan to bake diversity focused programs into all of pyOpenSci activities from the beginning.
- Streamline and improve the quality of smaller Python Scientific Open Source software tools that support scientific workflows: think api wrappers, spatial data tools and domain specific tools rather than Numpy and Pandas.
- We do this through peer review which not only assess code quality, testing infrastructure and CI setup. Our peer review also evaluates tool usability, documentation clarity and ensures alignment with community accepted standards for Python packaging.
- One way our peer review process fosters usability of the tools is by ensuring documentation includes vignettes that showcase tool functionality and provide easy quickstarts for users.
- Once tools are accepted into our ecosystem, we will track tool maintenance and find new maintainers, or support sunsetting as it makes sense to ensure users aren’t left hanging.
- Through peer review we will also address redundancy in tool functionality across the scientific Python ecosystem.
- Create a model to support peer review within domain specific subcommunities. We will begin our work here with the Pangeo community. Here, we will support a peer review process with community-specific standards applied on top of ours to ensure tool consistency.
- Provide training and mentorship in skills needed to contribute to open source. These skills are not just for open source. They are also fundamental to good open science practices and will support career trajectories in open science too.
- Grow support for maintainers. A critical component of this effort will be a communication campaign around the importance of open source to science.
We will also continue to curate our partnership with the Journal of Open Source Software (JOSS).
Package maintainers can still submit a package to us if they wish to publish with the Journal of Open Source Software (JOSS). Our partnership supports a “fast track” review in JOSS if its functionality fits their scope and requirements. Once a package is accepted by pyOpenSci, if the package fits into the JOSS scope our review will be accepted by JOSS. A package author then only needs to write a paper for JOSS, and they will be “dual” published with us and JOSS. This, friends, is a win-win!
Who does pyOpenSci serve?
pyOpenSci supports software tools for open science and domain specific scientific projects. These tools and packages can act as glue that holds a research community together, but they haven’t always received as much support as the foundational frameworks that are widely used (e.g. Pandas, numpy, scipy). We also support scientists using these tools and will empower folks like you with the skills needed to both contribute to Open Source and improve your open science workflows! In the long term these skills will be invaluable in your data intensive career trajectory!
pyOpenSci is about collaboration - not reinventing the wheel
Collaborations with existing efforts are critical. I welcome any feedback from you regarding groups that we should be engaging with! I will also need help finding reviewers and editors, mentors to support training efforts and voices to spread the word about the project.
And… what’s the timeline?
Unless I figure out how to create numerous robot clones of myself, please know that all of the above won’t happen immediately. I am currently the sole employee of the organization and this is my second week. But never fear - we will accomplish a lot in the next two years and look forward to and will greatly value support from some of you along the way!
Stay in touch!
What can you do now?
There are a few ways to get involved with us now.
You can sign up to be a reviewer by filling out our reviewer form. If you do this, we will reach out to you if a package is submitted that suits your background and interests!
- We will also be growing our editorial board in the upcoming months. Once you have reviewed for us you are eligible to be considered as an editor. We will be posting more about this process in the upcoming months.
- Our open peer review process is running! If you have a package that you are working on, you can submit it to our peer review process.
- You can learn more about our review process here in our community guide (which will be updated in the upcoming months so that it is more clear).
- If you have a package that you wish to submit but need some guidance, you can also submit a presubmission inquiry or a request for help here.
- You can follow us on twitter to keep tabs on our progress.
pyOpenSci was built with support from numerous community members
I would be remiss if I didn’t take a moment to thank the huge suite of community members who participated in early meetings and discussions surrounding pyOpenSci. Please note that this list is not comprehensive and likely misses folks who attended discussions held at the SciPy Conference a few years prior to my involvement and those whose names I am missing on meeting notes. Feel free to reach out and ask for corrections as needed! Max Joseph, Jenny Palomino, Kylen Solvik, Chris Holdgraf, Paige Bailey, Luiz Irber, Arfon Smith, Leonardo Uieda, Filipe Fernandes, Carson Farmer, Neil Chue Hong, Joe Hamman, Ryan Abernathy, Lindsey Heagy, Mike Trizna, Noam Ross, Karthik Ram, Martin Skarzynski, Levi Wolf, Daniel Chen, Ivan Ogsawara, David Nicholson, Tania Allard, Kirstie Whitaker, Sander van Rijn, Philip Meier, Moritz Lurig … and so many more!
I am so excited to be leading this effort and look forward to keeping in touch with everyone and supporting the community as pyOpenSci grows.
More to come… soon!