Chapter 12. What's Next?

In this chapter, we will do the following:

• Highlight some of the proposed upcoming features of Jupyter that you may look forward to. (Most of these examples are screenshots and descriptions of plans, without any coding available. The ideas presented are in the early stages and are not generally available)
•  Give an example of an issue that occurs when multiple users access the same Notebook in a standard Jupyter installation.

JupyterHub

We saw JupyterHub in an earlier chapter.

JupyterHub is the multiuser instance of Jupyter. All directions where we need to have multiple users interact with a Jupyter implementation will be shown through Jupyter.

The first step was multiple users. This required Jupyter to allocate separate space for each user.

The next step is shared services that are accessed by the one Jupyter implementation, but many users. Services need to have authentication (is this really user xyzzy?) and Jupyter is moving toward using OAUTH, the standard for web service authentication.

 

 

JupyterLab

We have looked at JupyterLab in an earlier chapter.

JupyterLab is devoted to producing the next user interface design for Jupyter.

Scale

The current implementations of Jupyter have a scaling issue since, as the number of users increases, performance degrades significantly.

With the widespread adoption of Jupyter, the problem is becoming more apparent with additional use/users.

The Jupyter team is devoting a large part of their near-term efforts to enhancing the scalability of Jupyter to handle large numbers of users seamlessly. Some of the solutions being worked on include the following:

• A proxy API having an API between browsers and server services scales out the application
• MOAR servers, allowing multiple servers per user
• Further integration of OAUTH throughout the server/services
• Stress-testing work by the Jupyter team prior to market release

Custom frontends

With the extraction of the handling of user interface components, users can now have completely customized frontends for their implementation. All of the entry points are documented to perform the steps required.

Interactive computing standards

Jupyter is, by definition, an interactive computing platform. The Jupyter team is working with industry members to produce standards (that Jupyter will meet) in the areas of:

• Notebook file formats
• Interactive computing protocols
• Kernel interaction

For this chapter, we will use a simple Notebook that asks the user for some information, as demonstrated in the following screenshot:

 

Summary

In this chapter, we learned how to expose a Notebook so that multiple users can use a Notebook at the same time. We saw an example of the error occurring. We installed a Jupyter server that addressed the problem, and we used Docker to alleviate the issue as well.