Thomas Kluyver

I don't think the schema includes keys in metadata, because the ones we use don't have to be there, and arbitrary ones we don't use are allowed there. It's essentially...

Oh, you're right. Thanks! I don't look at the schemas very often...

Min: > I'd like to iron out what "explicitly specifies a kernel" would mean, if not the kernelspec name in the metadata. In the context of (i), I mean that...

To frame the issue a bit differently: when we devised kernelspecs, we envisaged them basically representing languages - you have a Python kernelspec, an R kernelspec and so on. Those...

> Selecting a KernelSpecManager (or KernelPicker) implementation should probably be a global choice, not application-specific. Actually, my thought is that the notebook application may have a different KernelPicker from command...

Good catch, thanks. It was probably written by someone thinking of Python rather than JSON. Do you want to make a PR to fix it?

We're no longer supporting reading and writing notebooks in py files, and the functions to do so are [not documented](http://nbformat.readthedocs.io/en/latest/api.html). Feel free to fork that code into your own project...

The structure of a notebook object in memory is part of the nbformat API, so the `as_version` parameter exists to separate that from the contents of a file on disk....

What version of `notebook` are you using? This is likely to be a bug of xterm.js (which actually does the terminal emulation) rather than terminado (which passes the data through...

So the version of xterm we bundled was 2.9.2 - you can see it here: https://github.com/jupyter/notebook/blob/5.2.2/bower.json#L27 That still appears to be the latest release, if you want to open an...