# PFHub Working Group April Meet 2024

[](https://hackmd.io/PDGscIYUR5uLzCAuVHarGg?edit)

Meeting agenda, notes and actions for 2024-04-02 at 12 noon ET

Organizer
: Daniel Wheeler

Attendees
: - Daniel Wheeler (he/him)
- Mythreyi Ramesh (she/her)
- Phil Staublin (he/him)
- Xander ()

## Links

- [Google Meet][meet]

## Agenda and Notes

1. Any questions or items to raise for discussion (please add)
2. What are all your thoughts on PFHub (honest criticism)?
- How can it be much better?
- Myti: Easier to find things
- Phil: Make the CLI work with documentation!!!
- Worked examples of implementations
- Upload documentation
- Web design:
- Xander: The front page is totally confusing and making stuff hard to find
- Xander: The about section needs to be on the front page
- Xander: README becomes ABOUT section and that is the front page
- Daniel: Contents in side-bar on front page
- Vision
- Xander: holistic phase field hub rather than a registry of phase field results
- What's the point?
- phil: Didactic value
- not promoting any particular code
- How can we expand the use and community?
- Myti: collection of PF codes called [MicroSim](https://microsim.co.in) (India)
- online course for phase field by [MP Gururajan](https://archive.nptel.ac.in/courses/113/101/113101072/) (NPTEL)
- What about the vision outlined below?
3. What are the core issues to address from above questions
- Improve website
- JupyterBook?
- Phil: let's do some research on tools like JupyterBook
- Phil: what's the objective?
- Improve the website so that content is more manageable and accessible
- Reduce friction and work required to maintain
- Myti: maybe some Fenics examples in notebooks
- Improve data infrastructure
- Moving parts
- Schema
- CLI Tool?
- Phil, Myti, Xander: test CLI
- Other tools and platforms?
- Improve outreach
3. What can you all do to help? How much time and effort?
- Maybe just giving ideas or more?
- What parts would you like to work on?
- Should we have meetings going forward?
- Monthly meeting

- What to do?
- Myti: Research tools / become familiar with JupyterBook
- Create own repo to mess with JupyterBook
- Maybe integrate README into JupyterBook mock up + a few benchmarks
- Phil: Mock up some markdown file for front page
- Make the README better
- Investigate tools
- Xander: Mock up how the front page should look and work and what information ti should have
- phase field splash (maybe?)
4. Make worthwhile for all of you?
- Possible publication in the future?

## Notes from Slides

Following are taken from slides for recent workshop.

### Current Vision

- Central registry of phase field curated data with CLI/Web/API tool
to view and query results
- Integrated with a phase field metadata standard

### Recent Updates

- Split out pfhub-cli into separate repository
- https://github.com/usnistgov/pfhub-cli
- Convert to using Poetry and Poetry2Nix for builds
- Implement bones of MkDocs documentation
- https://pages.nist.gov/pfhub-cli
- Implement Zenodo uploads/downloads using pfhub-cli
- Implement notebook rendering using pfhub-cli
- Convert between old and new schemas
- Validate old and new schemas

### Next steps

- Ensure all benchmarks work with new CLI tool
- Implement `pfhub submit` with CLI and integrate with website
- Update `usnistgov/pfhub` to work with CLI tool
- Improve documentation with a series of tutorials / how-tos
- Implement `pfhub.yaml` inference with `pfhub generate-yaml`
- Improve validation
- Eventually integrate with new phase field schema

### Phase Field Schema Goals

- Generate a metadata standard for phase field data
- Use current semantic web technologies
- Adopt metadata standard for some published use cases
- *Possibly* provide a template for FAIR metadata standards for
materials simulation

## Action Items

<!-- links -->

[meet]: https://meet.google.com/thf-ygzh-zjx