commenting on someone else's Jupyter Notebook

[Colin Rowat]

I am beginning to use Jupyter Notebook.

I want students to comment on each other's Notebooks as part of a validation exercise.

I know how to insert markdown, etc., but have not found any 'automatic' way to identify a comment's author. For example, comments that I insert in an MS Word document are prefaced with my initials.

Does anyone know if there is such a feature, or whether an author needs to manually enter their identify?

Thank you,

Colin

p.s. I've also posted this question at stackoverflow, at https://bit.ly/2QNZYeg

[Jason Grout]

No, there's no inherent way to identify the author of a markdown cell in a notebook (the notebook doesn't even understand the concept of multiple users).

Thanks,

Jason

--
You received this message because you are subscribed to the Google Groups "Project Jupyter" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jupyter+u...@googlegroups.com.
To post to this group, send email to jup...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jupyter/5f1e4454-6508-426f-ae26-336c1d2ccc89%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[elb...@jhu.edu]

There was a poster presented at JupyterCon 2018 last month that might interest you.  It was titled "Designing Comments" with several co-authors (Rose Chang, Meredith Granger, Alena Mueller, and Taka Shimokobe).  I believe their focus was on designing comments in JupyterLab in particular and I don't recall that they had much more than a prototype, but they are Project Jupyter Interns and designers so one might infer you are not the first to recognize how great it would be to have a commenting system.  

You might try searching to see if you can find more info or reach out to their team directly to see if there are plans for an actual implementation.

-David

[Colin Rowat]

Thank you Jason and David.

An abstract of the paper mentioned by David is available here: https://conferences.oreilly.com/jupyter/jup-ny/public/schedule/detail/71539.

[Brian Granger]

That poster is part of a new grant we have funded to work on commenting and annotation in JupyterLab. It is just getting started, but there will be activity related to this on GitHub as things begin to move forward.

--

You received this message because you are subscribed to the Google Groups "Project Jupyter" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jupyter+u...@googlegroups.com.
To post to this group, send email to jup...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/jupyter/08398caa-2146-48ca-baf7-467dd11192f4%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Brian E. Granger
Associate Professor of Physics and Data Science
Cal Poly State University, San Luis Obispo
@ellisonbg on Twitter and GitHub
bgra...@calpoly.edu and elli...@gmail.com

[Colin Rowat]

Thank you Brian.

Off-list, Rose Chang has sent me a link to the poster: https://drive.google.com/file/d/1j-k9WSQUGgxVYLIOwC0_2qCqoM8RmMjB/view

Best,

Colin

[Denis Akhiyarov]

Multi-user edit, version tracking and leaving comments are possible right now in colab notebooks powered by Google Drive.

I was not able to open your bitly link on stackoverflow.

[elb...@jhu.edu]

The Google Drive solution ends soon.  It depends on the Google Realtime API which is deprecated.  Google says it will work "normally" until December 11th and be turned off in January of 2019.  Discussed here: https://developers.google.com/realtime/deprecation

-David

[Nicholas Bollweg]

Saw the poster at JupyterCon: great progress!

As for nuts and bolts: here's my 2c. We would be remiss to not consider the W3C Web Annotation Data Model for the on-disk/on-wire format. Standards, interop, conformance testable, rich media, yadda, yadda. Definitely worth a look.

The most mature open source implementation in this space is hypothesis: for an example of hypothesis annotations, check out their documentation, which is hosted on readthedocs, but embeds an annotation layer, served by their flagship deployment API.

The hypothesis server itself is a bit heavyweight, as they (wisely) are using industrial-grade tools like elasticsearch, but should serve as at least a normative API. A Hub service would be pretty amazing.

The client is angular/jquery (and hammerjs for mobile), so likely not a great fit for Lab integration, but does have a lot of good insights.

[Tim Head]

On Fri, Sep 28, 2018 at 2:44 AM Nicholas Bollweg <nick.b...@gmail.com> wrote:

Saw the poster at JupyterCon: great progress!

As for nuts and bolts: here's my 2c. We would be remiss to not consider the W3C Web Annotation Data Model for the on-disk/on-wire format. Standards, interop, conformance testable, rich media, yadda, yadda. Definitely worth a look.

Yes please! I think this is a very important point so want to support it.

 

The client is angular/jquery (and hammerjs for mobile), so likely not a great fit for Lab integration, but does have a lot of good insights.

I think a cheap way to start experimenting with hypothesis (or other JS implementations of annotation tools that follow the W3C standards) is adding a  <link rel="canonical" href="..." /> (https://web.hypothes.is/for-publishers/#good-habits) to the rendered notebook HTML and storing it in the metadata.

Jon Udell experimented with this at the Binder workshop at UC Davis in autumn 2017. Might be worth quizzing him about what he found.

T

--

Let us be your hub hero https://hubhero.net

[Colin Rowat]

Apologies Denis - here's the full URL: https://stackoverflow.com/questions/52428149/commenting-on-someone-elses-jupyter-notebook