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



  1. 4.5.1 Release is in-progress

    1. Test Plan <- opportunity for participation
  2. What is new from the last 3 weeks!!??
  3. LDCX report-out
    1. Distributed architectures for IRs
    2. Harmonizing Hydra and F4 AuthZ
    3. LDP Batch
    4. PCDM modeling - EDM/DPLA/IIIF/PCDM-Works
    5. F4 API specification
    6. F4 API-X
  4. Fedora Specification
    1. Messaging SPI <- close to done, need to tighten RDF serializations
    2. Atomic Batch Operations
    3. CRUD <- needs to include elements of RESTful HTTP API
    4. Resource Versioning <- not currently aligned with reference F4 impl, since that is just a copy/exposure of JCR which cannot be imported into specs
    5. Binary Fixity Checking <- needs help
    6. Authorization <- needs help
  5. ...
  6. Status of "in-flight" tickets

Ticket Summaries

  1. Please squash a bug!

  2. Tickets resolved this week:

  3. Tickets created this week:



Release 4.5.1

  1. A Fedora 4.5.1 release is in process. A release candidate will be released tomorrow.
  2. Three weeks will be alloted for testing, after which the final version of 4.5.1 will be released.
  3. New features in development now should be put in the Dev branch.
  4. Support for building on Windows needed? Yes, a couple of people in the community will test a Windows build.

Pregnant pause while Andrew reconnects.

Release Testing

  1. Review of 4.5.1 Test Plan at https://wiki.duraspace.org/display/FF/Release+Testing+-+4.5.1
  2. There are 11 manual tests listed. An additional content negotiation test could be useful. This test could retrieve content in various RDF formats. Can we automate these tests?

Test Automation

  1. Perhaps a brief shell script could maintain these tests in a cross-platform maintainable way. 
  2. Recommend a shell script for each manual test.
  3. Can someone write up a shell script that uses curl in order to do these manual tasks? Jared doing some test scripting now, Esme and others may assist.
  4. Beth: test scripts should be human readabale so that humans can execute these tests on their own. Adam: We still need a Test Compatibility Toolkit (TCK) to automate this no matter what. Still useful to have test cases target for humans to execute.
  5. Need JIRA tickets for test automation and to create a human readaable guide for manual testing by non-developers. (new JIRA Ticket: )
  6. Also need to run vagrant tests and consistency checks between Fedora, Solr and the triplestore.
  7. Test Camel messaging? Should test at least one camel message type.

Recent progress over the last three weeks
No comments

LDCX report-out

  1. Good presentations and discussions
  2. See links above in Agenda, items 3a-3f for detailed minutes for these sessions.
  3. 3a - Distributed repositories - shows value of Camel, messaging.
  4. 3b - Harmonizing Hydra and F4 AuthZ - encouraging. 
  5. 3c - LDP Batch - may want to implement as extension module, until HTTP/2 is more widely adopted. 
  6. HTTP/2 may be ready to be more widely adopted in 6-12 months.
  7. 3d - Cross repository modeling - encouraging.

Fedora Query interface

  1. Some people in the community, noted at LDCX and elsewhere, seem to want a query interface to Fedora. Others are not in favor of this idea as other components such as Solr are better suited to handle this. 
  2. May be difficult to implement a query interface as the back-end stores underneath Fedora differ. Would not want to just re-use the Modeshape search API. 
  3. Perhaps a general query interface based on messaging would work. 
  4. Need to better identify whether the use cases for Fedora search are synchronous or asynchronous.
  5. Overall Message - how can we make it easier for people to implement an async query capability? Need to simplify this.
  6. Perhaps API-X would be a suitable location to host this functionality.