JupyterLab, the next generation of Jupyter notebook software, is one step closer to becoming a fully collaborative tool for data science, machine learning, and computational research. The new Commenting extension can now be used by teams to ask questions, delegate tasks, address issues, request edits, make decisions, and ultimately to resolve problems as a team. Comments the team find particularly useful can be memorialized as annotations.
In the past, JupyterLab users would use third-party apps to coordinate their efforts around shared code and data. JupyterLab Commenting streamlines contextual coordination by providing a rich commenting interface within JupyterLab where everything in JupyterLab can be the ‘target’ of a comment-thread. While similar to the commenting experiences in Google Docs, Dropbox Paper, and other multi-user environments, JupyterLab Commenting is the first to bring a commenting system at full-scale to a data analytics platform where users can comment and resolve anything.
Really, I can comment on anything?
Yes! Not only can comment threads be created on a file, users are also able to comment on any and every entity in JupyterLab. From Notebooks, data grids, text files, and graphs, all the way down to notebook outputs, individual lines or characters in a text file, the contents of a CSV cell, row, or column, etc. This flexibility frees the user to focus on achieving their goal instead of focusing on how to coerce their tool.
All interactions are designed to be quick and concise. Only comments with visible on-screen targets show by default. A right-side panel is used to create and manage, and review comments inside JupyterLab on a per-file basis. While the commenting panel is opened, all targets that have comments will be indicated and interactive. The indicators, when clicked, will focus their associated comment thread. When the panel is closed, comment indicators will be hidden to declutter the workspace. This interface can also be used to memorialize a comment into the associated Knowledge Graph (JupyterLab Metadata!). With the seamless integration and clutterless design, this extension allows for collaboration and coordination within JupyterLab that was not previously possible!
When JupyterLab’s real-time collaborative editing feature is completed, this commenting system will be incorporated into it. In the meantime it will work independently (working example here). If users are connected to the same server, a new thread or comment created by one user will immediately appear to all other users connected to the same server, and comment target indicators will update as well. Thus, comments are created and resolved with a real-time feel for users.
This extension unlocks the exciting new ability to comment on code and data within JupyterLab. Teams can now more easily coordinate their efforts to resolve issues together, without leaving the JupyterLab application. Speed, efficiently, quality of work, and user satisfaction are positively impacted. Go download the JupyterLab Commenting extension today!