search

executablebooks / meta

A community dedicated to supporting tools for technical and scientific communication and interactive computing
https://executablebooks.org
129 stars 165 forks source link

Apply for an EOSS grant #287

Closed choldgraf closed 3 years ago

choldgraf commented 3 years ago

Background

CZI has another round of their Essential Open Source Support grant series open. Letters of intent are due March 30th (full proposals due several weeks later).

I think there is much work that could be done in the Jupyter Book / Executable Books ecosystem, much more than we could carry on our Sloan grant alone. A few ideas for things that we could pitch:

Questions

cc @pradyunsg and @sgibson91 and @cooperrc who I'm not sure get pinged by this repo but may be interested, and the @executablebooks/ebpteam who I imagine will be interested in this! :-)

chrisjsewell commented 3 years ago

Development that focused around bridging Jupyter Book and Jupyter Lab / other interfaces (e.g., MyST support in those interfaces etc)

Definitely interested in this, and also tying it in with my work on https://github.com/executablebooks/myst-language-support and https://github.com/chrisjsewell/rst-language-server I guess maybe I could be coerced into a bit of writing 😆

cooperrc commented 3 years ago

If so, any particular idea jump out at you as interesting to pursue? (either from the list above, or other ideas)

A big focus for CZI is on the role of FOSS in biomedical applications. Jupyter Book could play a huge role in this field, but its not widely established yet. I think exploring the publishing/bridging side of Jupyter Book might have a strong proposal:

Is anyone interested in helping wrangle and write a bit? (the application process isn't super hard, but will take some coordination and writing!)

count me in, let me know how I can help :+1:

KirstieJane commented 3 years ago

Heya! I would LOVE for JupyterBook to put in an EOSS grant. And The Turing Way would be extremely keen to support the application. Whatever you choose I'm sure we'll be happy with but let me ping out my personal dream and see if it can be integrated into any of the other plans!

Existing resources / early plans

Firstly, let me make sure that you know about two potential connections that are created / ongoing but need sustaining / maintaining / promoting.

Kirstie's dream :rocket: 🌟

We find that The Turing Way book is super intimidating for folks. There are loads of chapters, lots of content and I often hear "I'll read the book and get back to you" and .... well..... they don't. Because there's too much there to digest!

BUT - what if we could host subsections of the Jupyter Book in a "choose your own adventure" or "curate the reading list" format?

Then we could have views that said "Are you new to reproducibility? Would you like to learn about data management? How much time do you have (5 mins, 15 mins, 1 hour)?" and then someone could click through to answer those questions and see BOOM - their personally curated book!

All the content would be the same but we'd serve just a subsection of the content. I think it makes the book more accessible AND it really showcases the power of WHY you'd want to host a book online rather than just print out a PDF etc.

There's another small point which could fit in which is to customise the views - for example making a cover page - would be useful! We've found that seeing all the possibilities right at the start is what makes people anxious 🤗

choldgraf commented 3 years ago

Thanks everybody for this super helpful set of ideas! I will digest a bit and think! In the meantime I also wanted to share a conversation around publishing pipelines with Jupyter that many may also be interested in...not sure if/how it relates to this, but maybe it sparks some ideas?

https://discourse.jupyter.org/t/feature-idea-jupyterhub-binderhub-jupyter-book-as-a-publishing-platform/8359/8

ericholscher commented 3 years ago

Connections with the ReadTheDocs community to get JB support and documentation on there (either dev in Jupyter Book, or dev in RTD, or both)

We're a strong 👍 on this. We'd love to support this work, but we definitely would need more resources in a sustained manner to do this work. Having it be a CZI grant-funded role to at least handle part-time support, implementation, and maintenance of the integration between Jupyterbook & RTD would definitely allow us to support Jupyterbook.

I think there's already been a decent bit of research into what is required here in https://github.com/executablebooks/jupyter-book/issues/1091 and from folks on our team. We'd be :+1: to support an EBP grant on this work, or that included this work.

Let me know what we can do to more this forward, since I know the application deadline is coming up soon.

choldgraf commented 3 years ago

OK I thought a bit and whipped together this Letter of Intent for a proposal around the theme of Improving scientific publishing in the Jupyter ecosystem. Here's a rough draft:

https://docs.google.com/document/d/1mHqUmGeu1Y6YTCZLxa7EZSLExlRXtUUJQXncN97Kb6U/edit?usp=sharing

Rather than focusing on just the authoring pieces, it focuses on doing focused development + integration across many tools in the Jupyter ecosystem in order to create an "end to end" publishing workflow that could be replicated for many communities.

After writing it down I am not 100% sold on it, nor am I sure that it is quite the right grant for this EOSS series - so happy to chat about other ideas or venues if people have them.

Would love to hear people's thoughts - I also have a few specific questions:

  1. Is this the right scope for a proposal? It is very broad, but I think would have high impact.
  2. Is this grant series the right target for this proposal? Or do you think it is ambitious enough to warrant a different target for funding?
  3. Is the breakdown of roles correct? It is roughly 25% hub/cloud, 50% jupyter book / myst, and 50% JupyterLab
  4. What would be the organization that accepts this money? Should it be a collection of organizations? (I think that 2i2c could hire several of the folks on there, but perhaps others are interested in participating as well?)
choldgraf commented 3 years ago

Hey all - I haven't gotten a ton of feedback on this, which makes me wonder if

  1. I haven't gotten the scope of the proposal correct
  2. That folks are stretched too thin to work on this

Unless I hear a clear positive signal from others, I will plan to not submit an EOSS proposal for the ExecutableBooks project, and instead focus efforts on smaller more focused proposals as the opportunity arises. I am also trying to coordinate two EOSS proposals on the JupyterHub/Binder side of things, and this is stretching me quite thin, so I don't know that I have the cycles to re-work this one with a better focus etc 😞

cooperrc commented 3 years ago

Hey all - I haven't gotten a ton of feedback on this, which makes me wonder if

  1. I haven't gotten the scope of the proposal correct
  2. That folks are stretched too thin to work on this

Unless I hear a clear positive signal from others, I will plan to not submit an EOSS proposal for the ExecutableBooks project, and instead focus efforts on smaller more focused proposals as the opportunity arises. I am also trying to coordinate two EOSS proposals on the JupyterHub/Binder side of things, and this is stretching me quite thin, so I don't know that I have the cycles to re-work this one with a better focus etc

@choldgraf, I added some comments to shift focus to some combination of "transparency", "open science", and "reproducibility". I think we can get the work done, but the immediate benefit to biomed users would be kind of an "easy-share" button. They could generate reports, blogs, training docs, etc. The end-to-end publishing would be an awesome future application, but the problems in the proposal will have to be ironed out before there's a new "Jupyter Journal".

choldgraf commented 3 years ago

I think that is a great idea to improve the scope of narrative. Thanks for that!

I think where I’m a little bit more uncertain is the details of administration, who this work would fund, etc. I think 2i2c can probably take on a subset of this work (maybe all of it, I am not positive though) but we’d need to know relatively quickly. Are others interested in doing pieces of this work as well?

(I should also note, I don’t think that this particular round of funding is mission critical for the project, so it is not the end of the world if we decide to delay for a little bit and apply for the next round)

cooperrc commented 3 years ago

I would love to take on a position either on the Jupyter Lab or Jupyter Book side. Doing some development for either interface/extensions and/or MyST markdown/Sphinx and some use cases (I do think we can extend use cases to more than just publishing, be it training material, blog posts, etc.).

I'm happy to help, but I don't want to step on toes. Let me know what you need.

choldgraf commented 3 years ago

@cooperrc - I have a feeling that we are rushing things a little bit with this proposal. What do you say we skip this round, and brainstorm ideas for a subsequent round (or a different funding stream). If you have any thoughts or ideas I'd love to chat!

cooperrc commented 3 years ago

@cooperrc - I have a feeling that we are rushing things a little bit with this proposal. What do you say we skip this round, and brainstorm ideas for a subsequent round (or a different funding stream). If you have any thoughts or ideas I'd love to chat!

:+1: Sounds good. I know you're juggling a few proposals right now.

choldgraf commented 3 years ago

I'm gonna close this one, but I think we should keep an eye out for funding opportunities or another EOSS call. @cooperrc if you're interested in pitching in - what kinda work are you hoping to collaborate on? And do you imagine finding funding for your own time, or students/post-docs/etc time? (you're a professor right?)

cooperrc commented 3 years ago

I'm gonna close this one, but I think we should keep an eye out for funding opportunities or another EOSS call. @cooperrc if you're interested in pitching in - what kinda work are you hoping to collaborate on? And do you imagine finding funding for your own time, or students/post-docs/etc time? (you're a professor right?)

Yes, I'm a professor (covered 9-out-of-12 months). Ideally, I could cover some Summer salary. Student funding would be ideal, but it depends upon funding timeframe (<6 mos = undergrad, 0.5-2 yr = masters, >2 yr = Ph.D.).