The Berkeley Institute for Data Science (BIDS) welcomes Min Ragan-Kelley, Senior Open Infrastructure Architect, back to the UC Berkeley campus after spending a decade working in Norway. He recently sat down with Kirstie Whitaker, BIDS Executive Director, to discuss his history and his future at BIDS -- he plays a key role in realizing the vision for BIDS as a space for open scholarship, open source, and interdisciplinary collaboration on AI in science and society.
Kirstie Whitaker: Hi Min! We’re so delighted that you’re back! Can you tell us a little about your journey to BIDS in 2014 and where you’ve been working since then?
Min Ragan-Kelley: I started working on the IPython project with IPython Parallel with Brian Granger as an Engineering Physics undergrad at Santa Clara in 2006, even developing an early IPython Notebook during the summer of 2006 (that did not resemble the current iteration). I then started my PhD in Applied Science and Technology at Berkeley in 2007, working in the nuclear engineering department on plasma physics simulations. During that time, SciPy really became my home community, attending the SciPy conference every year, working with Brian, Fernando Pérez, and many others on IPython and other tools in the scientific Python ecosystem, and the local Berkeley py4science community. It was during that time, in 2011, that we launched the IPython Notebook that ultimately came to be the foundation of what would become Jupyter. I finished my PhD with incredible luck just as we got our first big grant for IPython, so I came straight to work with Fernando across campus, then at the Helen Wills Neuroscience Institute. When BIDS launched in 2013, we were immediately involved, and our team became BIDS members in 2014, launching Jupyter and co-creating JupyterHub to help enable courses and research groups to enable students and researchers to access hosted computing infrastructure.
I left Berkeley in 2015 to move to Norway and work at Simula Research Laboratory with Hans Petter Langtangen in the biomedical computing department (later the department of numerical analysis and scientific computing), and that’s where I was until I came back to Berkeley this Fall. I continued to work on JupyterHub while at Simula, and began contributing more to conda-forge, helping package the traditional scientific computing stack around mpi, petsc, and fenics used by my new colleagues and their students. At Simula, I supported researchers in scientific software development, and educators deploying things like JupyterHub, facilitating open and reproducible scientific software practices. After 10 great years at Simula, I’m excited to return to BIDS to keep working on open infrastructure for science and education with all the amazing exciting things happening on campus.
"As a connector between campus and Project Jupyter, I’ll translate the needs of students and researchers across UC, into upstream development and deployment, which can benefit communities within and far beyond UC."
Whitaker: Is it exciting to be back on campus and as part of the CDSS college? What are you looking forward to being back at Berkeley?
Ragan-Kelley: I am contributing to JupyterHealth as part of Open Platforms for Health, a collaboration with Computational Precision Health. We are working on open tools to enable medical researchers and clinicians to work more interactively with health data, including from medical devices like continuous glucose monitors, and all the appropriately strict rules and restrictions around that. I am also working more broadly on supporting open infrastructure for science and education, particularly via Jupyter and JupyterHub. I’ll be deploying some JupyterHub infrastructure for the AI Futures Lab, exploring collaboration models, and working closely with the DataHub, Research Teaching and Learning, and Data Science Undergraduate Studies teams to improve the Jupyter and JupyterHub ecosystem for today’s students and researchers. As a connector between campus and Project Jupyter, I’ll translate the needs of students and researchers across UC, into upstream development and deployment, which can benefit communities within and far beyond UC.
I would absolutely love to hear from people at UC what you love about using computers to teach, research or communicate with each other - and what might make those activities even easier!
Photo: Fernando Pérez welcomes Min Ragan-Kelley back to UC Berkeley at the Community Park Social
Whitaker: You’ve been contributing to open source for a long time! Which projects are you currently focused on?
Ragan-Kelley: I’ve been working on Jupyter since 2006, almost 20 years (and several years longer than it’s been called Jupyter), and I’m grateful to have been able to continue for so long. I’m currently most focused on maintaining JupyterHub, which I started when I was last at BIDS in 2014, which we position as a collection of tools to help connect humans to computers. I also contribute to Binder to enable reproducible environments and executable publications, and conda-forge to help people distribute scientific software. We recently worked with Organizational Mycology who produced the excellent Voices of JupyterHub report, and my immediate focus is on using what we have learned from that excellent project to make improvements in how we support a healthy JupyterHub community.
Whitaker: If you had one tip for a new contributor to scientific open source software, what would it be?
Ragan-Kelley: The most effective way I have seen for new contributors to get started in open source is to solve your own problems. If you have a tool you use that doesn’t work or needs something, start by exploring how it can be better for you - search issues and read the code - and then try to make it better for you. The odds are there are others like you, who would benefit from that work, and it’s far easier to stay motivated to solve problems when they are yours and you consistently see the benefits in a tool you use every day. That’s when you stick with it long enough to develop into more of a maintainer position, feeling both pride in and responsibility for the work that you have done.