https://bluejeans.com/426716450

Agenda

  • Plans for WG wrap-up and making progress on DMTN-085.
  • Table round on ongoing activities.
    • Simon Krughoff & Hsin-Fang Chiang:
      • Discussions on Provenance for QA needs
      • Summary of discussion in the meeting:
        • The current middleware plan is that there really will only be one source of truth for provenance (thumbs up)
        • We are happy with the idea that row-level provenance to the dataset level (e.g. “tell me exactly which CCDs contributed to this object measurement”), but that it will be possible to track object measurements back to the coadds on which they were made, and then the educated user can use further tooling to investigate the composition of that coadd. The WG will likely recommend that this tooling be made available.
        • We propose that the WG should note the possibility of provenance aggregate measurements in its report, but are not immediately convinced that it's worth spending significant time investigating them.
    • Lauren MacArthur: Digging into lsstDebug.
    • Trey Roby: PR on DMTN-085.
    • Eric Bellm: Will be bringing text forward before vacation.
    • Unknown User (tmorton): Started sketching out more concretely some of the drill down ideas.
    • Angelo Fausti:
      • Reading BG3 registry design.
      • Is thinking about moving the SQuaSH metric tables into the BG3 registry.
      • Motivation would be to re-use their existing database structure.
      • Angelo is going to think about this a bit more; if it's something he feels is worth taking forward, first port of call would be a chat with Jim Bosch.