Page tree
Skip to end of metadata
Go to start of metadata

Attendees

General

Agenda

  1. Sprint b8 planning, see Sprints B8 and B9 Planning

Minutes

  1. Project and process introduction
    1. Be on IRC!
    2. Pivotal tracker
      1. Tickets, points, phases, etc
      2. Estimate, Start, Finish (add comment or PR link), Deliver, Accept
      3. If you create "tasks" within ticket, spin off into separate tickets
  2. Ticket walk-through and assignments

Priority Walk-through

AuthZ - osman, other?
  • Osman to contact Greg
  • Scott interested in helping with AuthZ documentation
Content Modelling - Scott, Greg, no-Adam
  • recipes from different institutions
  • no disseminators in F4
  • UWisc has ~20 content models
    • These content models depend on disseminators
    • Scott to document mapping mapping some of his existing models
  • Document how to create F4 content models
  • Greg to document mapping mapping some of his existing models
  • Note the fact that REST endpoint is generic (RDF-schema), not F3 custom
Large files - Eric
  • Read-only? Read-write?
  • Binaries only? Binaries and metadata?
    • Define metadata serialization? allow writing new objects
  • What happens when we serve a file through the REST-API
  • Eric, interested in async crawl to create checksums in background
    • Eric, 500-TB
  • Esme - in the cloud (not appropriate)
  • Scott - use case may not be a good fit
  • UNC - desire to have fixity
  • Where does the REST api fall down
Search - Eric, ?Ye, Scott
  • Chris: Must define properties for them to be indexed (internal caveat)
    • Supports dcterms
  • Note that internal search will not be expanded
  • External search, Ye: Solr documentation
    • Scott: interested in testing
    • Eric: interested in mapping RDF fields into Solr doc
      • Eric to add pivotal tickets
Versioning - Mike
  • Scott: interested in testing/documenting simple Versioning
Easy Deployment - ?Scott, other?
  • Create recipes for System-Properties
  • Create ticket for collecting user-needs for config management expectations (from stakeholders)
  • Send email to advisors
  • Osman: Will create deployment documentation
  • Chris: Maybe forking console plug-in https://github.com/jvanzyl/jetty-console/tree/master/src/main
  • Create ticket for creating top-level directory for all system properties
  • Esme: elevate to splash screen
    • Change tmpdir to CWD
Performance-single - Esme, ?Ben
  • Has post 0.5-TB test results (10,000 items, X-GB each)
    • Need to replicate test of production servers
      • UNC,
  • Scott to test performance with and without AuthZ
  • We need to start cleaning up documentation
  • Focus on "Getting Started" guide
  • Use apache module documentation as a guide
  • Profile: persist, authz

Developer Focus

Osman
  • AuthZ testing and documentation
  • Open to other tasks
Adam
  • RDF iteration
    • Waiting on Ben for Jersey2
  • Create a profile for testing (how does that map to configuration?)
Greg
  • Benchmarking on cluster
  • AuthZ
Scott
  • AuthZ testing
  • Content model examples
  • Verifying content models
  • Testing search
  • Stubbing "Getting Started" guide
  • Working with Adam on testing profiles
Frank
  • not on line
Eric
  • Benchmarking upper bounds of REST API
  • Add monitoring to FileSystem
  • Documentation of Solr, indexing fields from RDF
Ye
  • Search
Chris
  • Clustering tests
    • Real-world benchmarks
    • Interest beyond ingest
  • Admin search
  • HTTP-API polish
Ben
  • Jersey2
    • May be able to leverage Adam
Esme
  • Testing performance
  • User manual
  • Editable views
Mike
  • Versioning

Actions

  1. Flesh out your respective tickets
  • No labels