We will publish our Jupyter notebook roadmap in the Collaboratory before the 2016 Summit in Florence. Are you going to the summit?
The one thing I would say that we're aware of Julia and the addition of non-Python kernels to the Jupyter system is interesting to us, but we have some higher priority items on our list.
Roughly speaking, in order of priority:
- More flexible kernel container management - Currently a user can run only one type of kernel at a time and it can be only docker image.
- More scalable infrastructure on the backend - we're looking at options to provide significantly more computation capacity for the notebooks. This will mean larger memory and more CPUs available to running kernels.
- Additional kernel types for Spark and other languages. Julia would be a candidate but would be based on a user survey.
In the meantime if you can describe specifically what sort of work you're doing with the Jupyter notebooks maybe we can suggest some alternatives to the current Collaboratory system.
I hope that makes sense and helps you to understand what is coming next.