Time/Place

This meeting is a hybrid teleconference and slack chat. Anyone is welcome to join...here's the info:

Attendees

  1. Danny Bernstein
  2. Peter Eichman
  3. Bethany Seeger
  4. Jared Whiklo
  5. Andrew Woods
  6. Kevin Ford
  7. Michael Durbin
  8. Ben Pennell (star)
  9. Aaron Birkland


Agenda

  1. Announcements

  2. Sprint progress/checkin

  3. Issues to discuss: 

      1. Can we close it? 
    2. https://github.com/fcrepo/Fedora-API-Test-Suite/pull/208/files
      1. What does the spec require as far as patching binaries is concerned? 
    3. Specification update:
      1. Default acl's:  report from the spec committee.  Next steps.
    4. https://github.com/fcrepo/Fedora-API-Test-Suite/issues/182
    8. Dependency Update Discussion
  4. Additions to the draft October Fedora Newsletter
    1. What would the tech team like to highlight?
  5. <your agenda item here>

  6. Please squash a bug!


  7. Tickets resolved this week:


  8. Tickets created this week:


Minutes

  1. Announcements

    1. Danny sang a song about announcements
  2. Sprint progress/checkin
    1. CTS has a few small bugs and feature requests remaining, almost all the tests complete

      1. Jared is going to try 6.1 notification events.

      2. Still need the machinary to listen to the notifications.

      3. Modeshape sends out a lot of notifications that are "not exactly modifications"

      4. Is there an example of what is strictly expected in terms of format? It may be difficult to test

        1. There are a few MUSTs which are testable, such as identifier and activity stream namespace

      5. awoods: There are about 17 issues, do we want to defer them or do they need to be addressed now or can some be closed?

        1. danny will go through the tickets to try to make this determination, awoods will double check
  3. Issues to discuss
    1. Reviewing tickets in the sprint board https://jira.duraspace.org/secure/RapidBoard.jspa?rapidView=40

      1. 2730 - It may be operating system related

      2. 2778 - mostly done, tests mostly passing, but there is one that may or may not be related that is failing

      3. 2848 - the actual ticket is done. But there is an issue with date formatting, something that is formatting the dates is changing the significant digits which is actually changing the timestamp if a timestamp ends with a 0. So 140 milliseconds becomes 14 milliseconds in the memento.

        1. Danny - create another ticket for this second issue

        2. Bethany - sure, but we wouldn't want to merge this issue until the second ticket is addressed. Bethany still interested in keeping her name on the ticket

        3. Danny - maybe a second ticket isn't necessary since the issue is documented in the original ticket

      4. 2854 "Defining acl:accessToClass without an acl:default predicate has no benefit" - has been converted to a documentation ticket. Basically warn people that if they are doing something, it probably doesn't do what they think it does.

      5. cts208 - was merged yesterday

      6. cts206 (Inheritance and Default ACLs)
        1. The issue was that there is no definition for what the backstop ACL should be, and that there is no requirement that the default acl be exposed.
        2. peter: two suggestions:
          1. have a way to retrieve the effective authorization per node regardless of whether something is directly assigned,
          2. or just expose the backstop ACL explicitly off of the root node (this would just solve the initial testability issue)
        3. Peter will begin creating a sidecar modification for spec related to this.
        4. Do a 404 check for the time being.
      7. 2092 has existed for a year, may not be essential.
        1. IndirectContainer related. There were triples not being persisted.
        2. Medium gnarl, luke-warm interest, so may not be essential for version 5.0
  4. Newsletter
    1. Should we co-opt the existing newsletter, or send a separate tech update message?
    2. Updates on sprint progress, effort to remove modeshape dependencies, state of the 5.0 effort.
    3. Have a link from the newsletter to the tech update
    4. Danny will compile first edition of the tech update letter. "The road to no Modeshape"


Future Agenda Topics: