Below, Tulbure shares his thoughts on the current and future state of Jupyter.
1. How has Jupyter changed the way you work?
Our platform's back end is implemented using Python; therefore, we adopted Jupyter as a default tool for sharing executable snippets of code and for prototyping small parts of our system.
2. How does Jupyter change the way your team works? How does it alter the dynamics of collaboration?
As a team, we started using Jupyter almost everywhere where we needed an implementation technique. We started collaborating on these small pieces of code until we had the plan done; after that, the actual implementation was straight forward. We are also using the notebooks to make the demo and the documentation of our third-party API.
3. How do you expect Jupyter to be extended in the coming year?
Jupyter will definitely become one of the main technologies that will change the face of collaboration in big remote teams. Faster and dynamic kernels will become easier and easier to implement. Storage services will be selling cheap servers with preset kernels for Jupyter (or at least that's where I would start investing if I were them). In the coming year, besides more and more plugins for interacting with the UI part of the notebooks, I suspect services that host version control systems will integrate Jupyter as part of their UI and as part of team collaboration.
Another expectation would be the use of Jupyter in academic research. Having shareable, executable, and reproducible data is one of the key points in "pushing" science further.
4. What will you be talking about at JupyterCon?
Mark and I will be talking about the FAIR (findable, accessible, interoperable, and reusable) principles that every scientific paper should respect, this will be showcased using Jupyter Notebooks. We will explain what our users waned to see in our platform, how the process of enabling previewable
.ipynb files opened a new door for the world of academia—and how we've stepped trough that door by showcasing executable notebooks in what is soon to be a fully fledged digital lab for reproducible data that is built on top of Jupyter.
5. What sessions are you looking forward to seeing at JupyterCon ?
I would go to almost all of them, but I am very eager to check out the following:
- Citing the Jupyter Notebook in the scientific publication process
- Hosting Jupyter at scale
- Xeus: A framework for writing native Jupyter kernels
- Using Jupyter at the intersection of robots and industrial biology
- Cloud Datalab: Jupyter with the power of BigQuery and TensorFlow