October Chat Meeting

Tuesday, 2017-10-03, 11am EDT

Location: https://gitter.im/chimad-phase-field/chat-meeting-2

Previous Meeting: https://github.com/usnistgov/chimad-phase-field/wiki/September-Chat-Meeting

Etherpad for live note taking: https://etherpad.net/p/chimad-phase-field-chat2

Agenda

• Meta issues
  • Can everyone edit the wiki?
    • Please let me know if you can't!
  • Is wiki a good place for the notes and agenda for the meeting?
    • More permanent and reliable than the Etherpad
  • Review agenda
  • Another Whenisgood poll for next meeting?
• Action items from last meeting
  • See https://github.com/usnistgov/chimad-phase-field/wiki/September-Chat-Meeting#action-items
  • Did anyone do anything?
    • Inaction confessions
    • LaTeX slide for CHiMaD Phase Field, presented by Trevor at the NIST Diffusion Workshop (11 Sept): https://gist.github.com/tkphd/797ce91f510f035662423c78f48a7cae
• Update on meeting with Ben Blaiszik (developer of Materials Data Facility)
  • Olle, Andrea, Ben, Daniel W., Trevor
  • Issues from the meeting:
    • https://github.com/usnistgov/chimad-phase-field/issues?utf8=%E2%9C%93&q=is%3Aissue%20is%3Aopen%20MDF
  • How close to integrate with MDF?
    • Should web form automatically dump data to the MDF? Trevor has views on this.
  • What about storing the simulations in an MDCS instance along with data?
• Engagement plan
  • https://github.com/usnistgov/chimad-phase-field/projects/6
  • We need ideas for presenting data on comparing simulation results
    • 2 point stats for example
  • Milestone 1.1 is where we get serious about engagement
    • https://github.com/usnistgov/chimad-phase-field/milestone/8
• Dev plan
  • Milestones: https://github.com/usnistgov/chimad-phase-field/milestones
• Website updates since last meeting
  • Merged all the upload-a-thon simulations
  • Andrea wording fixes: https://github.com/usnistgov/chimad-phase-field/pull/493
  • Update community pages: https://pages.nist.gov/chimad-phase-field/community/
    • Please send me a profile!
  • Some others:
    • https://github.com/usnistgov/chimad-phase-field/pull/412
    • https://github.com/usnistgov/chimad-phase-field/pull/413
    • https://github.com/usnistgov/chimad-phase-field/pull/409

No-action (stuff to think about) items for November

If you have an idea below then maybe flesh it out with an issue.

Daniel Wheeler

• Milestone 0.4
• Milestone 0.4.1
• Milestone 0.4.2

Daniel Schwen

• add tags to allow arbitrary simulation groupings, this will help us create a series of simulations so that we can do comparison of all moose simulations for different time steps - in the long run we can do convergence tests, see https://github.com/usnistgov/chimad-phase-field/issues/500
• display error estimations in some way once tags have been implemented
• get involved with website PRs

Trevor Keller

• best-practices focus
• host phase field / benchmark / upload workshop at meeting
• workflow demonstration would be really useful
• Come up with a user story or two representing our target demographics: grad students and devs
• Conduct a timestep and a mesh convergence study on spinodal decomposition; upload bundled simulations with "best" tag
• Use HiPerC
• Look into hosting a workshop at http://www.siam.org/meetings/ms18/

Stephen DeWitt

• Webinar: https://github.com/usnistgov/chimad-phase-field/projects/6#card-4956547
• examples of how the benchmark problems could be used, and subsequent building out of extensive documentation on implememting, uploading comparing

A. M. Jokisaari

• Give more invited talks to research groups about the benchmarks and website - get some invites: https://github.com/usnistgov/chimad-phase-field/projects/6#card-4956520
• seminar talks at universities (same as above)