August 2017 Team Meeting Agenda Topics

[ellisonbg]

To help us quickly collect topics for our August 2017 team meeting, please enter comments and upvote existing comments. We are short on time (only two days), so this will help us to surface the most pressing discussion topics.

https://paper.dropbox.com/doc/August-2017-Team-Meeting-Agenda-NvwrkTa47sUQJbRjDwfzN

[Carreau]

I think we should make a JupyterCon debrief while still fresh in our heads.

(feel free to edit add subitems)

• lightning talks,
• closing statement
• Diversity in programming language
• attendee feedback, can we access it?
• More visible Tshirt "Ask me anything about Jupyter"
• Predefined tracks

Update: Topic discussed ✅

[rgbkrk]

Notebook HTML and JavaScript mimetypes

• Acknowledging the legacy "API"
• What belongs
• Can frontends iframe it
• ...

[choldgraf]

Use-cases/ways to highlight tools in the Jupyter ecosystem. Ideas to build documentation and demos. (in particular for the newer tools like JupyterHub and JupyterLab)

[damianavila]

Roadmap updates/upgrades (and with better definition/accuracy) so everyone could know about our next steps and can plan accordingly.

[ellisonbg]

What to do with our roadmap:

https://github.com/jupyter/roadmap

[bollwyvl]

UI, Metadata, and Tools for Internationalization/Localization of Notebooks

building on the existing core JEP

• what can be internationalized?
  • markdown cells
  • code cells (e.g. chart labels, etc.)
  • metadata fields (title, etc)
• what does it get me?
  • metadata/SEO flagging
  • output formats (multiple ipynb, single html, multiple html)
• low-friction ways to propose/review/iterate/accept new translations (non-developers)

[rgbkrk]

Spark UI

Infrastructure related pieces:

• Diffable updates
• Cluster resources

[bollwyvl]

UI, Metadata and Tools for Richer Layouts

continuing this JEP, this discussion, RISE, nbpresent

• what to make?
  • posters
  • slides
  • dashboards
  • multi-page documents, a la bookbook, multirise, etc.
  • apps
• what to consume?
  • any mime you can render!
  • multiple notebooks, etc.
• what else?
  • layers
  • archival output (i.e. DOM-based PDF, e.g. chrome-headless
  • themes

[rgbkrk]

Resource Info Request

This started out as a little discussion in jupyter/jupyter#264 at the last meetings. Now to follow up!

---

When working in a jupyter environment, it would be helpful to see current resource usage

• CPU(s)
• Memory of the system
• Memory of the current kernel

Beyond that, people need information from libraries. Taking Spark as an example, they want to know several fields:

• Spark UI (string - URL)
• Memory on executors
• State of executors (?)
• Spark Version (string)
• Hadoop Version (string)
• Shared cluster mgr resources (containers, vcore-hours, memory-hours, ...)

[minrk]

display-data cell type: jupyter/notebook#1123

[ellisonbg]

Jupyter Markdown syntax:

At last team meeting we discussed doing some research to understand the precise markdown syntax that the classic notebook supports. We have started this work by creating notebooks with the CommonMark test suite. We currently don't pass much of it :(

How should we proceed?

[ivanov]

Kernel testing (validation, infrastructure, etc)

[ellisonbg]

Where do we want to be in 3 years? Vision, strategy, big picture.

[choldgraf]

re: @ellisonbg 's comment, specifically the topic of striking a balance between the enterprise/larger business community, and the more "open-sourcey", small-business, or academic community

[rgbkrk]

Maintainer roundup

• Archiving (new hidden feature on GitHub), rather than jupyter-attic
• Current workflows

(Feel free to edit the above if you think of other subtopics)