2022-03-02 Thompson/Wagner Check-in meeting notes

Participants

  • @Matt Thompson

  • @Jeffrey Wagner

Discussion topics

Item

Notes

Item

Notes

General updates

  • JW – Still working on vsite tests, just need to keep my focus for a while and I’ll get them done

  • JW – ForceBalance - Getting LPW’s feedback

    • Should we just move smirnoff_hack into toolkit?

      • Unsure, possibly consider this later. Would need to synchronize releases to make this switch

    • JW pinged LPW on slack to see about coordinating a release, will follow up when we hear back

    • We may be able to future-proof smirnoff_hack a bit more by passing *args and **kwargs, like we did for partial charge caching in a previous PR

  • Would like to enlist JM’s time refreshing Interchange’s docs

    • Somewhat important, somewhat urgent/timely

    • No attachment to current docs - a ground-up rewrite would be fine

    • Hard (for MT) to assess what minimal set of docs users would provide users broad value

    • Should I write (rough) docs for units package?

    • JW – JM does well in “polish” mode, but more structure would be good if there are particular areas for feedback

      • MT – Maybe the protein-ligand notebook PR would be a good starting point. I’m open to anything from polish, to complete rewrite if the current state needs serious help

      • Also want info about the GAPS in docs/examples - What is missing that I haven’t thought about enough?

      • Finally, general polishing of any area of the docs

    • MT – Timeline: Starting movement in the next two weeks would be absolutely great.

    • Approval process: MT will be approver/guide for this PR/series of PRs to interchange.

    • Priority: JW will talk to JM, and communicate to MT what priority this task has relative to other JM tasks.

  • Need better branding (logos? DOIs/citation mechanisms?) for newer projects in our stack - basically everything after evaluator

    • Interchange

    • QCSubmit/bespokefit/other Josh + Simon stuff

    • Units + Utilities

      • JW – openff-units readme could be expanded with basically “starting with openmm”, “starting with pint”, “going to openmm”, “going to pint”, “arrays vs not-arrays”, and “serialization”. Large swaths of this could be copied near-verbatim into the 0.11.0 migration guide.

      • MT will write up the above, decide whether to merge directly or ask JM for review

  • https://github.com/openforcefield/openff-toolkit/pull/1198

  • ForceBalance CI - actually broken in mysterious ways or is Matt just in a perpetual state of brain fog?

  • If we WANTED to mark 0.10.1 and 0.10.2 as broken, here’s how to do it: https://github.com/conda-forge/admin-requests/pull/284

    • Decided with Simon’s input we do - PR opened





Action items

Decisions