Time/Place
This meeting is a hybrid teleconference and slack chat. Anyone is welcome to join...here's the info:
Attendees
Part 1:
- Danny Bernstein
- Andrew Woods
- Peter Winckles
- Jared Whiklo
- Bethany Seeger
- Youn Noh
- Thomas Bernhart
- Daniel Lamb
- Aaron Birkland
- Ben Pennel
- Rosie Le Faive
Part 2:
- Cancelled for today due to sprint activities
Agenda
- Highlights from last week
- Tomorrow's meeting
- Sprint Update:
- design discussions
- Transaction Aware Persistent Storage Layer (diagram)
- Flow Diagram Adding Binary to Existing AG
- Transaction Aware Persistence Layer with tx header
Transaction Aware DB/FS Persistence Layer
- Other use cases
- core development gateway tickets:
- Fedora 3→ 6 Migration work
- documentation review
- Java 11 issues?
- Your topic
Tickets
In Review
Please squash a bug!
Tickets resolved this week:
Tickets created this week:
Notes
The participants decided to cancel the second tech meeting from today, as there is an ongoing sprint.
- Highlights from last week:
- Tomorrow's meeting between Fedora and OCFL team regarding possibly large inentory files in OCFL
- regarding fedora implementation of OCFL persistency: no decission has been made so far regarding unversioned content/modifiable head
- Idea from Andrew Woods: instead of writing new OCFL version for every change, write all changes to head until explicit Fedora version is created
- would mean OCFL versions and Fedora versions are aligned again
- no head version, treat latest version as head (no extra directories)
- squashing would be possible
- Peter Winckles : with this approach it could get tricky to implement updates as atomic operation (create v1, copy to v2, write update to v2); but approach could work
- Fedora Leaders and OCFL editors must be contacted to check if this approach is feasable
- Do we need to support deleting of mementos if we implement a squashable HEAD?
- Sprint Update:
- design discussions
- Presentation of Flow Diagram Adding Binary to Existing AG by Ben Pennell
- core development gateway tickets
- FCREPO-3046: should be done later this day(?)
- Review of https://github.com/fcrepo/fcrepo-specification-atomic-operations/pull/8
- no need to limit transactions to single resources
- keep a lot of existing approach for transactions if possible
Actions