2020-01-07 Sprint Transition
Date
Jan 7, 2020
Participants / Stakeholders
@Felimone Amone Junior
@Matthew Kumbuyo (Deactivated)
@Chongsun Ahn (Unlicensed)
@Christine Lenihan
@Ben Leibert
@Sebastian Brudziński
@Aleksandra Ciesińska
@Mateusz Kwiatkowski
Retrospective
It's been hard to focus on the SELV project due to the holidays and the need to focus both on an Angola release and Malawi support. Malawi has needed urgent support, and @Mateusz Kwiatkowski is allocated 50% to providing it.
@Josh Zamor has been really responsive regarding the volume management requirements - it’s great to have his guidance.
@Felimone Amone Junior has also been very fast to give @Aleksandra Ciesińska feedback regarding the CSV configuration files.
Although there’s more to do with regard to defining the system’s configuration via CSV files, it’s great that we finished setting up the system’s infrastructure (reporting stack, Jenkins, etc.)
Showcase
@Mateusz Kwiatkowski demoed https://openlmis.atlassian.net/browse/SELV3-19
@Aleksandra Ciesińska finished and reviewed:
Planning
@Mateusz Kwiatkowski will finish writing the volume management related design tickets. Implementing them is our highest priority and will take the team about 2+ weeks.
Batch/aggregate requisitions need more design/discussion - we’ll meet about them separately.
The system’s CSV configuration files potentially still need refinement. Ben summarized the discussion within a comment on the page.
Ben and Felimone will tag-team the creation of report-related mockups.
The “Current CCE Capacity by Province and District” and “Cold Chain Allocation reports” should both include Period filters.
Superset doesn’t support cascading filters nor allow dropdown selections to be limited to just a single choice. It’s therefore possible for users to make ambiguous selections, like choosing two provinces and a single district in cases wherein the both districts contain a district by the same name. The Superset reports can and should rely on users to make sensible decisions. Rather than showing no data in ambiguous cases, they should generally run their queries exactly as specified. Allowing users flexibility in terms of how they use the reports is generally more important than protecting them from making poor use of the reports.