2023-04-10 Hyrax Valkyrization Meeting

 Date

Apr 10, 2023 1pm EDT/10am PDT

Join Zoom Meeting:
https://us06web.zoom.us/j/89244297043?pwd=TjVsaWR0bGxXMjBVOXBaY0xIbVlhUT09

 Participants

  • @Heather Greer Klein

  • @Trey Pendragon

  • @Jon Dunn

  • @Juliet Hardesty

  • @Rebekah Kati

  • @Esmé Cowles

  • @Brian McBride

  • @Daniel Pierce

  • @Kirsten Leonard

  • @Rob Kaufman

  • @Alicia M. Morris

  • @david.schober

  • @Margaret Mellinger

  • @Beth Shields

  • @Carolyn Caizzi

  • @Jim Halliday

  • @Kevin Kochanski

  • @Nicholas Mark Homenda

  • @John Weise

  • @Bohyun Kim

  • Sebastien Korner (Michigan)

  • @Jon Cameron

  • @Emily Lynema

  • @Bradley Watson

 Goals

  • Agree on what “finished” looks like – work to be done to complete the integration of Valkyrie into Hyrax

  • Outline additional work needed after this point for other components, backend connectors, etc.

  • Determine who can participate in getting this work completed by contributing development time, funding service provider development time, and other expertise

  • Confirm a schedule for completing this work, and how additional institutions could contribute

 Discussion topics

Item

Presenter/facilitator

Notes

Item

Presenter/facilitator

Notes

Project Board review

Daniel Pierce

  • https://pg.nurax.samvera.org/

  • Main things that need work are generating a IIIF manifest, displaying a IIIF viewer, and derivative generation

  • Probably other things that need work too

  • Definitely work to help people migrate data and upgrade

  • Some developers have had issues getting tests to pass with all backends, which can impact how effective people are

  • UCSB/UCSD have fixed some of the issues that are on the board, haven’t been able to backport all of those fixes

  • In theory there should not be much work to implement Fedora 6 once postgres is working; suggestion to spin up a Fedora 6 nurax for parallel testing

  • as a community and administratively, once most of the work on the project board is completed, need to define what support means for a Valkyrie adapter in Hyrax. Helpful for figuring out how to or whether to support future new adapters after Fedora versions and postgres

Community-organized developer sprints: availability and planning

 

Heather Greer Klein

  • Surfliner team (USCSB and UCSD) offered dates from their workcycle to contribute to the work; this would be mid-July to August (July 31 to August 25th). Their sprints run Wednesday to Tuesday, but are flexible.

  • Is there a strategy that extends the worktime by expanding developers over a greater period of time?

  • Surfliner team and Princeton team are fairly large. Would more work get done all together, or with overlap on the front and back of the sprint?

  • Having a team sparting earlier, with some overlap to handoff work to another team. July-August team time could be the middle or the end of a series of sprints.

  • Other institutions may have developers to contribute to this effort.

  • Teams able to participate in development

    • Surfliner (6 developers)

    • Princeton (4 developers)

    • Indiana (possible, in addition to Daniel as Tech Lead and Julie for QA)

    • A few other institutions may be able to participate

  • QA testing would also be a vital contribution for supporting this work

Financially supporting contractors for Valkyrization project board sprints sprints

Heather Greer Klein

  • Hyku for consortia would contribute financially, and Indiana University could contribute financially as well

  • 4-5 developers on 2-3 sprints are what SoftServ has been looking at, and that would include work on QA. This would be what we would be raising funds to support.

  • We are clear on what is missing from nurax-pg to get it to functional. We also know that the migration planning work is important and high on the list of needs for SoftServ to address, but the unknowns are murkier there

  • Abstract work chunk around file versioning behavior that needs to be addressed for the adapters

  • What would be the work for the 8 weeks, what would be the definition of success?

  • concern about the definition of a working fedora 6 implementation would be scope creep

  • Proposed success criteria is Hyrax working with postgres as the backend. The project board reflects this as the target.

  • the Samvera Board would like to use repo.samvera.org as a migration test case

  • a clear and easy migration path is important and requires further experimentation to scope.

  • Want to build a parallel Fedora 6 testing suite, and a Fedora 6 advocate to engage in testing the migration and adapter process.

Other items, wrap up, plan for further communication

 

  • Need to have further discussion about the migration path.

  • Heather will communicate to the community about this meeting and ways to participate in either development and QA sprints, or contributing financially to support the sprints

 Action items

 Decisions