JupyterCon Online: more than a conference
With glee, on February 7, we announced the return of the Jupyter community conference as a global event to be held in Berlin, Germany, in August. In a burst of enthusiasm, 174 people signed up to review for the technical program (you still can!).
Barely a month later (March 11), the World Health Organization declared COVID-19 a pandemic, tech companies had already made the call to cancel all travel, then Europe became the epicenter of the pandemic.
As we began working on contingency plans and negotiating with the contracted venue, we silently published the JupyterCon website, and the Twitter account fell silent. Now we’re ready to make public our plans for JupyterCon Online to be held October 5–17, replacing the in-person conference this year.
This is our vision for the conference of the future.
The key question is: in the absence of physical presence to celebrate our community, meet new people, and network for professional opportunities, what could be the incentives for attendees, speakers, sponsors, and volunteers to be part of the conference? A sharp focus emerged on career advancement and learning as the motivations.
We developed a vision in which a “conference” is now a learning platform, unconstrained by synchronous schedules or geographical location, coalescing a multitude of mini-events and rad new content, learning experiences, and online social interactions.
The project we developed comprises:
- an online learning platform to create courses and organize content, providing user profiles to track learning and earn micro-credentials,
- integrations with third-party tools for web conferencing and text-based threaded discussion,
- online labs with access to JupyterHub/Binder attached to the content,
- beyond just “talks,” content organized as mini-courses with permanent resources attached and ensuing conversation.
The conference of the future convergences with online learning and credentialing.
Conference Program
The technical program for JupyterCon Online will be selected by a team of topic track chairs, informed by volunteer reviewers, from submissions to the NOW OPEN Call for Proposals. Much of the program will be on-demand, but a few important pieces are planned to be live:
- Tutorials: consist of prepared written materials and exercises in Jupyter notebooks, pre-recorded video by the instructor, live office hours with participants each day, and text-based discussion. The conference team will create a MOOC-style mini-course from the author-prepared materials. Participants completing the tutorials will receive certificates.
- Keynotes: streamed live to YouTube each day, with private backchannel discussions in the JupyterCon Mattermost server, public backchannel on Twitter, and also live, moderated Q&A after the talk.
- Regular presentations: pre-recorded, with timed release on YouTube Premiere, backchannel discussions in the private text-based forum, and in public on YouTube and organically on Twitter.
- Panels of Speakers: since the regular presentations are pre-recorded, these are an opportunity for the audience to interact live with the speakers. We’ll cluster speakers by topic, for a live broadcast discussion with a moderator, after their pre-recorded presentations aired. Audience can submit questions ahead of time for moderators to choose, and can also ask live.
- Posters: these are digital artifacts that can be static or interactive (e.g., Voilà dashboards), plus a pre-recorded 2-min pitch on video.
- Live lightning talks: 5-min moderated live presentations, with a text-based backchannel discussion, but no Q&A with speakers.
- Birds-of-a-feather: open-forum video chats organized organically among attendees.
- Interviews with influencers in the community as an additional draw of activity and discussion. Other live panels not connected to pre-recorded talks (e.g., Q&A with JupyterHub developers).
Part of a larger initiative
I worked with NumFOCUS (where I’m co-chair of the Board of Directors) to develop a wider program that we named NumFOCUS Academy. It will boost the educational mission of the organization, whose signature program is the PyData conference series—also put at risk by the pandemic. We wrote a proposal for funding to the Sloan Foundation for development of a scalable ecosystem consisting of an online learning platform, a JupyterHub server, front-end websites for JupyterCon, PyData and NumFOCUS Academy, and services like e-commerce, single-sign-on, analytics, and so on.
Sloan approved the grant on May 20, making this ambitious project a reality!
The JupyterCon leadership also secured a Platinum sponsorship from OVH Cloud, who will host all infrastructure needs for the first year. (OVH already hosts nbviewer and a large part of MyBinder.)
Our technical partner is IBL Education: they will deploy and support the learning ecosystem, with Open edX as the centerpiece technology. (Read more: What’s Open edX?) IBL is behind the NVIDIA Deep Learning Institute, the Freshman Year for Free initiative of Modern States, and many other major digital learning initiatives. They also previously worked with me to develop Open edX extensions to create courses based on Jupyter notebooks.
JupyterCon Online will be much more than a conference. We conceived a long-term strategy with the key vision of magnifying career-advancement opportunities for all members of our community, and assembling a permanent library of learning resources.
— Lorena A Barba, JupyterCon 2020 General Chair
JupyterCon Online will be held October 5–17, 2020.
NOW OPEN: Call for Proposals
Want to sponsor JupyterCon? Email us at jupytercon-sponsor@numfocus.org to receive a prospectus.
