Erik's AGU talk planning

[consideRatio]

Notes sketched during meetings about planning an ACU talk / poster submission.

• [x] Gather info about the talks' format

  • Title
  • Abstract: Max 2000 characters
  • Virtual live presentation is 10 to 15 minute presentation
  • A full session takes an hour or two and involves ~6-8 talks and Q&A after each talk. If a session receives more than that, the other will be presented in a poster format.
  • The abstract submission will be used to decide if they can accept the submission as a presentation or as a poster.
  • Submission process: submit, get accepted for a certain format (talk, poster), then you have ~3 months to prepare the concrete content.
  • Deadline for submission: 4th August

• [x] Find reasonable session to submit to:

  • List of sessions
  • IN004 - Adopting Principles to Enhance Capabilities for Using Earth Science Data

    Other principles, standards, and practices also contribute to improving capabilities for reusing data. Session presentations describe how adoption of such capabilities enable, facilitate and/or enhance use of Earth science data.

  • IN009 - A Moonshot Ambition: Creating a Globally Integrated infrastructure to Find, Access, Interoperate and Reuse Earth and Environmental Science Data to address Global Challenges.

    This session solicits papers from researchers working at the programme, project, or group level that demonstrate results in achieving the ambition of a global Earth and environmental research infrastructure commons, maximising the potential to contribute to known (and unknown) global challenges.

  • IN016 - Cloud Computing's Support for Earth and Space Informatics has Become a Sophisticated Solution to Computing Needs

    This session seeks use cases and scenarios to highlight the reasons for adoption and barriers to the adoption of scientific cloud computing. Abstracts should describe use cases, benchmarks, and real-world examples of the scientific use of cloud computing.

  • IN020 - Data and Information Services for Interdisciplinary Research and Applications in Earth Science

    Presentations addressing the specific industry and societal needs, best practices, learned lessons and new challenges in cloud computing, big data, AI/ML (artificial intelligent/machine learning), data provenance, and scientific reproducibility, regarding information access, visualization, and analysis, are highly encouraged.

  • IN030 - Open-Source Software, Notebooks and FAIR Software

    This session will look at the role of FOSS, cloud-based architecture solutions, metadata, online notebooks and other user interfaces to support information sharing, scientific collaboration, scientific reproducibility, QA/QC and analytics at scale solutions.

  • IN034 - Science In Application Containers

    This session will encompass topics such as building and running containers, open source packages and environment setups that can be used by the earth science community, container usage in CI/CD, and availability of containers to be used for scientific experiments. We encourage submissions from anyone who is using containers to run scientific code.

• Scope a narrative for talk

  • [x] Brainstorm pieces
  • [x] Structure a narrative
  • [x] Flesh it out
  • [ ] Add speaker notes

Talk ideas

Initial brainstorm on topics by Fernando

• Deployment in the cloud, working concretely with domain scientists with needs that are flexible and evolving. How to make use of standard things and put it together into a tailored development for the needs of the users.
• What have I learned and what is possible?
• RTC, welcoming students, auth with github, SSH, magic_castle...

Long time goal of a deployment for researchers, they can choose and combine

• Hardware options
• Environment options: JupyterLab, VSCode, VNC, SSH

What are possibilities now in JupyterHub deployments?

• SSH to server
  • rsync
  • sftp
• VNC to server
• JupyterLab in server
• VS Code

The evolving role of a cloud based JupyterHub

• JupyterHub provides easy access to an Jupyter based environment.

• Evolution on several levels

  • JupyterHub itself is evolving (RBAC)
  • JupyterHub distributions are evolving (TLJH, Z2JH)
  • JupyterHub aware services are evolving (jupyterhub-ssh, jupyterhub-sftp, jupyter-idle-culler, daskhub, binderhub)
  • JupyterHub user environments are evolving (JupyterLab, VS CODE, RStudio, Remote desktops)

• Flexibility on several levels

  • JupyterHub deployments can be tailored
  • JupyterHub user environments can be tailored
  • JupyterHub's cloud environment can be tailored

Abstract proposal draft

https://hackmd.io/@29EjZnuAQx-ME7Olgi2JXw/ryTguWByK

References

• Abstract proposal draft - Comments very welcome!
• Submission portal
• (Private) My submission draft in AGU

[whyjz]

One quick thought about the session selection: IN004 (Adopting Principles to Enhance Capabilities for Using Earth Science Data) might be the other possibility. IN030 emphasizes the support of collaboration. For us that'd be the use of JupyterHub. IN004 focuses on the ways we use/share Earth Science data, which is also one of the key points of using JupyterHub. For me, both topics would be a good fit for the proposed abstract.

[whyjz]

Here are some questions we can attack in the abstract, which might interest earth science people:

• Why do we need JupyterHub? Isn't the traditional way (e.g. ssh login to a server) just fine?
• If my team is all using Matlab (which is true for a great portion of the ES community), How do we switch to JupyterHub collaboration, and what are the benefits for that?
• How steep is the learning curve of JupyterHub? Do we need a tech-savvy person for my team to manage the deployment?
• How does JupyterHub make the earth science data (or code) more FAIR?
• Do only the people who have to work with big data need JupyterHub? A more generic form of this question would be: who needs JupyterHub? Maybe geophysicists/oceanographers/atmosphere people for sure, but what about geochemists, structural geologists, and the people with expertise of the other sub-domains? This is heavily domain-specific so I think maybe we can revisit this several times as we flesh out the abstract and the presentation content.

[kmpaul]

To avoid confusion and aid in searchability, I think the typo in the issue title should be corrected to say "Erik's AGU talk planning".

[consideRatio]

@whyjz thank you so much for helping me scout for sessions of relevance, and onboarding me personally into this, and brainstorming of relevant content to talk about! I'm more thankful than you will know! Thank you!!

I'm feeling quite confident about the format at this point and think that I can start considering a more concrete topic fitting of a 10-15 minute talk / poster of relevance to the IN004 or IN030 sessions - this is my next step. I look to have something to outline by Monday, 2nd Aug.

[consideRatio]

@andersy005 i'm trying to reach you, are you able to check Pérez group slack?

[consideRatio]

Abstract submitted!!

[whyjz]

@consideRatio I am closing this issue as its major goal (abstract submission) is done. We can reopen this issue or set up a new one for the upcoming presentation if needed.