bossee-project/proposal

Proposal

Building Open Science Services on European E-Infrastructure (BOSSEE)

BOSSEE is a proposal that was submitted on January 29th to the H2020 European E-Infrastructure call INFRAEOSC-02-2019.

Its mission is to promote Open Science (Open Source, Open Publications, Open Data, and Open Proposal) by supporting the Jupyter ecosystem, deploying generic Jupyter based services (binder federation and more), integrating them tightly in the European Open Science Cloud, and demonstrating their value through applications in astronomy, education, fluid dynamics, geosciences, health sciences, mathematics, and photon science.

The project was seeded as a follow-up to OpenDreamKit (see the brainstorm notes), and grown organically in the same open spirit. The proposers include core Jupyter developers (lead: Benjamin Ragan-Kelley) and long time Jupyter activists from many disciplines, in France, Germany, Poland, Norway, and Switzerland.

Work Plan

Positions

If accepted (notifications around June) the project will start around September 2019, and will open about 11 postdoc or research software engineer positions in our various institutions.

Stay tuned

• Twitter

• Slack channel

Notes for authors

proposal's Issues

Study gryd.us to inform the task about education

See this announcement on the jupyter-education mailing list about https://gryd.us

Re-review abstract

I wrote the abstract a looooong time ago. It should be reviewed based on the actual proposal.

We may also want to remove the reference to PaNOSC - we have much more evidence of Jupyter use the case, so I don't think it necessarily strengthens the case here.

Resubmitting to INFRAEOSC-07-2020

We are working on resubmitting the proposal to INFRAEOSC-07-2020, specifically a3:

(a3) Services supporting scholarly communication and open access (4M)

based on existing initiatives across Europe (institutional and thematic repositories, aggregators, etc.), the services should empower researchers and research communities and initiatives with the necessary tools and functionalities for systematic publishing, analysing and re-using of scientific results beyond publications (data, software and other artefacts), as well as supporting long-term preservation and curation. The services should also enable scientific workflows with adequate metrics and monitoring mechanisms supporting career development and the monitoring of funding and research impact. Support to a catch-all repository for open research should be provided.

---

I've been informed that we may want to change our acronym for the new call, since BOSSEE doesn't describe the updated mission as well. What I have thought of so far, given the publication focus and Binder-related work:

OSPRE: Open Science Publication of Research Environments (Yay)

Consistent order of participants, resources, participants on title page

I don't know if there is any system to how the Participants are listed in section 4. I think it is random.

The same is true for the order of resources in section 3.4.2

On the first page, the participants seem to be listed in alphabetical order of the name (exception: coordinator Simula first).

Would be good if we had the same order in all places. (Editing members.tex and resources.tex should fix it.)

Show example of notebook somewhere

It would be good to have an example of a notebook somewhere.

Let's assume a reviewer doesn't know what it is, then this will be very useful.

Who has an example ready or can produce one? The challenge is to show key features on little space. We'd also need a paragraph or two of explanation.

Collect cost per PM for each site / position type

cc: @IzabelaFaguet

Standardize wp/task/deliverable labels

In ODK we have not been very good at choosing good latex labels for our tasks & such, which always made it a pain to remember/retrieve them. We could try to aim for some better consistency, starting with ensuring the following schemes:

• the latex label, filename, and reference in the spreadsheet for a wp/task/site should be identical
• ???

Should we try to do this now? Or is this to disruptive at this stage where we are all editing simultaneously? We could also do it after the proposal is accepted. If it is ...
cc: @fangohr, @minrk

Describe particpant type somewhere

There is no description of the participant type in the document

• leadPI: leader of a site
• PI : leader of a task
• PM: project manager
• R : researcher and engineer ?

Tidy up file before submission (comments, significant infrastructure)

• even "make final" does not remove all comments in the final pdf. Check and remove manually

• remove "Significant infrastructure" headline from Participants where there is nothing to report

Collect PIC numbers for each site

Add a direct link from the README to the latest pdf produced by continuous integration

For very quick access. But also because I actually myself failed to access the pdf :-); there was no "Artifact" tab when I went to CircleCI. Maybe because I was not logged in?

Add images to WP 2 and 3

We start to have good coverage with images in WP4 (it's easy), and Thomas is working on something for section 1.

Can we show some images in the WP2 and WP3 for core and ecosystem? Just a screen shot of the Binder home page, and or a screenshot of binder building container? Or something from the LIGO publication that shows use of the system?

This should be relatively low effort to create and will provide visual candy.

Third parties: CNRS ?

Do we have to add the CNRS as a third party in section 4.2 ? If yes, here is a description text

LINK OF THE PARTICIPANT OT THE THIRD PARTY
The CNRS, “Centre National de la Recherche Scientifique” (National Center
for Scientific Research) will be a linked third party of École Polytechnique
in the framework of a Joint Research Unit agreement (Centre de Mathematiques
Appliquées, CMAP). The CMAP Joint Research Unit legally links the parties,
which have long-standing collaborations well beyond the BOSSEE project, and
therefore fulfil all requirements in Article 14 of the Model Grant Agreement
to be considered linked third parties.

LINKED THIRD PARTY DESCRIPTION
The CNRS is a public organization under the responsibility of the French
Ministry of Higher Education and Research. As the largest fundamental
research organization in Europe, CNRS carried out research in all fields of
knowledge, including the humanities and social sciences, biological
sciences, nuclear and particle physics, information sciences, engineering
and systems, physics, mathematical sciences, chemistry, earth sciences and
astronomy, ecology and the environment. CNRS encourages collaboration
between specialists from different disciplines in particular with the
university thus opening up new fields of enquiry to meet social and economic
needs. With 34,000 researchers, engineers and technicians, CNRS is organized
around 10 institutes which orchestrate its scientific policy. CNRS has a
budget of approximately €3.3 billion. Its 1,053 laboratories, including 95%
in partnership with universities, higher-education institutions and other
research organizations, are present throughout France.

Populate work packages with one paragraph each to outline purpose

Define and share the aims and objectives

The aims should give in just a few words an overview of the project: motivation, strategy, and how it all ties into the call.

Here is a first attempt:

proposal/objectives.tex

Lines 12 to 26 in 3554419

	\item \label{aim:jupyter}%
	Support (and steer?) the Jupyter ecosystem, as a general purpose
	toolbox for interactive computing and visualization that supports
	the entire life-cycle of open science, from initial exploration to
	publication, engineering (better word(s) about usage in industry
	here?), teaching, and outreach.
	\item \label{aim:services}%
	Leverage this technology for all scientists, in Europe and beyond,
	through generic services that are tightly integrated into the EOSC,
	\item \label{aim:demonstrators}%
	Demonstrate the value and versatility of such services through
	applications to mathematics, medicine, (... what's done at XFEL),
	...

"Open Science" or "open science"

Throughout much of the proposal the term "Open Science" is capitalized, I suppose, in order to give the term extra weight.

However, there are some places where it is not capitalized. For stylistic consistency perhaps most of those cases should be changed to "Open Science" as well?

Add Comic

This [1] is a nice starting point to explain where we are coming from:

the comic illustrates the current state of the art prototype workflow, partly achieved with the OpenDreamKit project, which has overlap with some of the partners in this project.

In this project, we want to make this process more reproducible for the future, more flexible and improve several other components of the ecosystem that allow this.

Should be mentioned somewhere in the first few pages? Could also go to the impact section - there we haven't got much material yet, and we could use the comic as an illustration for what the impact of the project could look like, and that it is broadly applicable across many domains. [Maybe that's better?].

[1] https://opendreamkit.org/public/images/use-cases-binder-logbook-solution.png

Project management and structure

cc: @minrk, @fangohr

As planned, I have imported the management structure section from ODK, and adapted it (including the picture). Please review!

There are some TODO's here and there. Major point to check:

• are end users sufficiently represented?
• I haven't touched the risk management section; are there things to update there w.r.t. ODK?
• Hans: I specified that you would be the Project Deputy (which in practice you have been in the preparation :-) ). Is that ok with you?

No leadPI for INSERM

All the types used in INSERM description are R (researcher).

Who will be leadPI (leader of the site)?

You probably have also PI (leader of a task) who are not the leadPI.