Attendees
General
- Indicates who took minutes -
- Call-in: Google-hangout at:
Meetings
7/7
Sprint planning
7/8
- Esme
- Looking at file system federation tickets
- Metadata updates only work in read/write mode
- Creating metadata/directories/files still causing problems in ModeShape
- ModeShape is failing when trying to append a child
- Will try to replicate the problem in a unit test and submit a ticket to the ModeShape community
- Moving on to bug tickets submitted by Indiana
- Longshou
- Working on 'Transparent filesystem persistence'
- Also working on 'NPE when submitting SPARQL query with FILTER'
- Mohamed
- Working on 'Create XACMLWorkspaceInitializer Unit Test'
- Andrew
- Will test Longshou’s pull request on expanding/contracting resource paths
- Prepping for Java client meeting on Thursday
- Attending meetings
7/9
- Esme
- Working federated filesystem
- Will be moving on to related tickets
- Longshou
- Finished sparql bug
- Will be breaking down ticket
- Need to configure destination directory for jcr/xml
- Mohamed
- Completed XACML unit tests
- Working on clustering
- Andrew
- Reviewed Longshou's and Mohamed's completed tickets
- Organizing with Artic as a beta-pilot
- Assessing LDP client libraries for Java client
- Will be reviewing Longshou's and Ben's PRs - longer tasks
- Please add topics to committers meeting
7/10
Committers Meeting
7/11
- Esme
- Working on file system federation bugs
- Seem to be ModeShape bugs
- Will work on file system federation bugs submitted by Indiana next
- Longshou
- Working on 'Attempt to delete non-assigned mixin throws error'
- Mohamed
- Working on 'Cluster: Mirrored content'
- Working on 'Create XACMLAuthorizationDelegate Unit Test'
7/14
- Mohamed
- Working on cluster performance issues
- Trying to get more powerful machines to use as cluster nodes
- Will know by tomorrow if those machines are available; if not, will use other machines
- 'Create XACMLAuthorizationDelegate Unit Test'
- Working on 'rdf:type triples are duplicated in output'
- Longshou
- Working on transparent persistence issues
- Working on 'Update GET fcr:sparql to return a SPARQL service description'
- Working on 'Update repository root RDF serialization to use the standard SPARQL predicate to point at the fcr:sparql service'
- Finished 'Update fcr:sparql requests with a form'
- Finished 'Attempt to delete non-assigned mixin throws error'
- Esme
- Working on file system federation tickets
- Can’t push forward any further without read/write federation
- Can’t implement until changes merged into ModeShape
- Will mark as finished pending the next ModeShape release
- Moving on to ontology tickets
- 'Ensure response predicates conform to ontology'
- 'Publish useful expositions of the ontology mapping from Fedora to LDP'
- Andrew
- At mini conference on Research Data on Friday
- Good event, mostly librarians, not overly technical
- Interesting project putting BibFrame over Fedora 4
- Will be out Wednesday-Friday
7/15
- Esme
- Working ontology tickets
- Will combine ontology files
- /rest returns lots of junk
- Create ticket for what gets returned in responses
- Some uris are simply invalid
- Working with modeshape bugs
- Longshou
- Working sparql and rdf tickets
- Mohamed
- RDF type duplicates
- Digging into jax-rs serialization
- mvn clean install ticket
- No news on VMs
7/16
- Longshou
- Finished 'Update GET fcr:sparql to return a SPARQL service description'
- Also finished 'Update repository root RDF serialization to use the standard SPARQL predicate to point at the fcr:sparql service'
- Nearly finished 'Document fcr:sparql "query via url-encoded"'
- Looking for new tickets to work on
- Esme
- 'SPARQL Update fails with PathNotFoundException when node path contains spaces'
- 'Get baseURL for JMS events from JAX-RS instead of requiring it be configured separately'
- 'Throw informative exception on invalid Federation'
- Mohamed
- Finished 'Fedora Repository HTTP API Module fails to build without 'clean' due to integration test failure'
- Working on 'Unique IDs in BasicRolesWriterIT and AbstractBasicRolesIT'
- May have some time left over to pick up another ticket
- Hardware resources should be available today to test cluster
7/17 (Sprint Wrapup)
- Sprint review
- Mohammed: Sprint went pretty well, the process was different from local process
- Found the daily standups useful for getting info from other people
- The RDF type duplicates ticket was very helpful for getting familiar with the code, since it involved several layers of the code.
- Testing tickets didn't help in that way, but were very useful for getting into the testing framework.
- Longshou: everything went fine.
- David: from several sprints:
- Our standups sometimes drag on, and we get into the weeds talking about issues
- Would be much better to identify who can help someone with a problem and let the standup move on.
- Not such a problem for a small number of developers, but would be a bigger problem with more developers on a sprint.
- We sometimes have a lot of active tickets, with one developer working several tickets.
- This sometimes happens when you start a new ticket and get a finished ticket reopened, or for inter-related tickets.
- But it makes it riskier that multiple tickets will be left open at the end of the sprint, and prevents other developers from being able to grab tickets.
- So it's good to keep an eye on this and try to get tickets finished.
- Esme: I enjoyed this sprint, and it was my first time doing significant work on Modeshape, reporting several bugs, opening tickets and PRs. I was pleased by how responsive and engaged they were, and am confident that our issues will be resolved.
- Though I'm not clear on their release timeframe, so some of my work is waiting on that.
