Skip to content

Standard Names Meeting Minutes 2025 03 20

Jump to bottom
Courtney Peverley edited this page Mar 20, 2025 · 2 revisions

Agenda

Attendees: Michael Kavulich, Cheryl Craig, Jesse Nusbaumer, Dom Heinzeller, Jordan Powers, Soren Rasmussen, Courtney Peverley, Grant Firl

PRs and Issues for main branch

Action items from last time

Standard Names governance

We should stand up a group containing at least one member from each group using (or considering using) Standard Names

  • NCAR/CGD
    • CESM/SIMA
      • Many present
    • ESMF
      • Ann Tsay responded that she would inquire about other ESMF team members, but was willing to be POC in the meantime
  • NOAA/DTC
    • Many present
  • JCSDA
    • François Hebert did not respond, will follow up
    • Steve Vahl?
  • NRL
    • Dom
  • NORESM down the road

These members can meet regularly as needed to discuss updates to the standard names, rules, and other Standard Names topics.

Standard Names updates

Opened issues that were brought up in the last PR and not resolved:

  1. Decide what variables should and should not be standard names
  • "flag" and "index" variables (some boolean, some integer)?
    • From the notes, it sounds like we had previously considered not including "flag" variables (boolean)
    • For CCPP, could these variables be packaged into a constituents object?
    • For other use-cases,
    • Currently (with "index" variables) represent approximately half (!!) of all standard names, and many have questionable value for inter-component functionality
    • If flags are included, how to name them? We currently have "flag_for_", "control_for_", "identifier_for_", "do_", and some have no qualifier at all!
  • scheme-specific tuning coefficients/weights
  • scheme-specific bins (e.g. GOCART aerosols)
  1. Should Standard Names include constants, or should this be a separate standard?
  • Should we include definitions, or just the names? Should the names be base names? (probably)

To-do/Action items

Other issues needing discussion

Notes

  • Tag for JEDI stable version (#96)
    • Michael to create v0.2.0 tag (71d9a92 hash)
  • Changes after rename of repo (#98)
    • Waiting on Steve G re-review
    • Michael K to respond to Dom’s comments
  • Michael working in branch on: #91, #92, #93
    • Will start using “description” or “comment” instead of “long_name”
      • ONLY for standard names repo (won’t change long_name in metadata)
      • Don’t want confusion about there being a required “long_name” for metadata
  • Governance document
    • Michael seeking comments/suggestions
  • flag/index variables - should these be in the dictionary? How to name them?
    • Jesse - likes it when the name is ~ a statement (like “is_mpi_root” or “do_radiation”) - makes it easier to understand what “true” or “false” means
    • Keep flag variables in a different section? (not physical quantities)
      • Jesse agrees
      • Michael K to separate into different section and then we can discuss how to go about the names/rules
        • Cheryl - also would be nice to sub-categorize into the classes - “control_for”, “do”, etc
  • Courtney - could maybe do forks with full standard names for individual hosts
    • But enables divergence and is hard to maintain…
  • Scheme-specific tuning coefficients/weights and bin variables (e.g. GOCART aerosols)
    • Maybe another section(s)?
  • Constants
    • Cheryl - should be included; could be separate section
    • Jesse - should agree upon value as well?
  • Michael to put out feelers to establish new meeting time for standard name discussions

Clone this wiki locally