Jupyter extension presents all available Jupyter kernel specs, remote Jupyter server and Python environments to users when users open a notebook. The selected kernel doesn't necessarily have all the dependencies installed or might be slow to connect to, meaning when users click the run button on a cell, users are not likely to get execution results immediately. The delay might be seconds, or minutes if it requires downloading of ipykernel or Python extension, etc.

Currently, after users click the run button, we immediately show a pending icon to the cell status bar, indicating the cell is about to run once the extension/kernel is ready. The outline and run all button in the top toolbar will show a spinner too. These are the only two places where we show progress.

Jupyter extension might show a notification when it receives the request to start a kernel. For example, it shows a notification of the ipykernel installation progress for a new Python environment.

From the image shown above, inside the notebook editor, there is no clear indicator that we are connecting to the kernel, except the pending icon on the cell status bar.

Proposals

To make the connection state more intuitive, we can explore a few other UX affordances to help users understand what's going on after they click the run button

1. Editor Progress Bar

We can show a progress bar at the top of the notebook editor

2. Spinner on Kernel Status Bar

We could turn the icon in the kernel status bar to a spinner, which can give a strong indication that the kernel is running