Eric Gentry

Conceptually, I think expand/collapse are good ideas for these kind of multi-notebook/console/shared-kernel workflows. Some things to note during my testing: - Expand and collapse all are only valuable when the...

> > some of the code changes will require commits here AND in the examples repo, so it makes sense to do these separately to expedite the original PR >...

> @ericsnekbytes could you resolve conflicts here (and ideally cleanup the history) (if this is still needed). Will do, I'll comb over and itemize/list any remaining work in the top...

@krassowski On the surface, it seems like you resolved all of the open items in the previous PR, but I wanted to do a detailed review to ensure I didn't...

I'm assembling a list of items to address in this PR below: - Use sphinx references instead of fully qualified links as noted [here](https://github.com/jupyterlab/jupyterlab/pull/14838#discussion_r1462045877) and [here](https://github.com/jupyterlab/jupyterlab/pull/14838#discussion_r1462046221) - Check the examples...

@krassowski These are the remaining items I found while checking the original PR, if you have any to add feel free. I had problems using sphinx references last time, so...

@krassowski I tried to use the binder link but it's broken...chat me sometime if you're free and would like to test.

It gave me some kind of invalid token error I believe...

Mike's proposal, especially step 3 where kernels are grouped by type/spec, is excellent as it clearly communicates what kernels are available by type, and concisely shows the case when a...

For context, [here are the docs](https://ipython.readthedocs.io/en/stable/interactive/magics.html#magic-load) for the `%load` magic command. I think what you want goes beyond what `%load` is meant to do, which is just insert text into...