Database Interop Hackathon/Pre-Meeting

From Evolutionary Informatics Working Group
Jump to: navigation, search

The core developers of NeXML, CDAO, and PhyloWS met two weeks prior to the hackathon to better prepare the participating emerging standards for actual application at the hackathon.

Goals & Targets

Guidelines:

  • Choose a very narrow limited set of goals, and include time to
    • evaluate carefully what we have done
    • extrapolate and plan for the big meeting when we have 20 people to keep busy
  • Need to discuss how to accommodate all the projects using php.

Target ideas:

  • Implement services that would allow Mesquite to query and retrieve character data and trees from TreeBASE and from MorphBank.
  • Agreeing on how to implement semantic annotation of metadata types of the providers.
  • More fleshing out of PhyloWS specs.

Information to gather in advance:

  • Implementation details of the providers: which languages/libraries they use, what metadata they want to expose.

Notes from meeting

Thursday, Feb 19

Participants arrive.

Fri, Feb 20

  • 9-10am: Priorities, use-cases, documentation needs
  • 10-11.30am: Objectives, goals, and assessments (with Jonathon Cummings)
  • 12-1pm BBL seminar at NESCent by Arlin Stoltzfus & Rutger Vos
  • 1-3pm
  • Agenda and outline for conference call
  • Remote participation tools set up
  • 3-4pm: Teleconference with hackathon participants
  • 4-6pm: Continuing earlier work & wrap-up for the day
    • Information gathering form for data types for I/O and formats
    • PhyloWS: Review of current spec against SRU (with Ryan Scherle).

Sat, Feb 21

  • 9.30am-12.30pm
    • Expression of metadata and metadata semantics in NeXML
      • Representation of references
    • Re-using existing metadata vocabularies (such as DC)
  • 12.30-1.30pm Lunch
  • 1.30-5pm
    • Expression of metadata and metadata semantics in NeXML
      • Generalize the solution for representing references
      • Attaching GO terms to tree nodes
      • Linking OTUs to taxon identifiers and to specimen records
  • 5-6pm Dinner
  • 6-8pm

Sun, Feb 22

  • 9am-12pm
    • Contining expression of metadata
      • reworking reference using more precise format recommended for Dublin Core
      • oops, NESCent lost our changes from last night
      • re-wrote our section on representing EQ annotations on phenotypes
    • Wrapping up and follow-up to-do list
      • Survey (Arlin)
        • edit to add category for metadata types
        • edit to add questions about impementations (what is primary language for data processing, i.e., middleware in 3-tier scheme; what is the indexing methcanism for text searches; all custom code or uses toolbox or other code base)
        • send live survey to participants
      • solicit feedback on examples (Hilmar)
        • from TDWG colleagues on assigning specimens to collections
        • from Berkely Onto group on ??
        • from hackathon participants on whether examples are useful and relevant
      • refactor wiki
        • "implementations" relocate (Hilmar)
        • nexml (Rutger)
      • encourage participants to enter use cases on wiki -- ask Dave & Karla (Hilmar)
      • distribute CDAO information (Arlin)
  • 1:30 Brandon & Enrico departed
  • 1:30 Lunch
  • 2:30 to 4:30 pm
    • see "to do" list above
  • 4:30 Arlin departed

Logistics

The pre-meeting takes place Friday - Sunday, Feb 20-22, 2009 at NESCent in Durham, North Carolina.

Participants

Participant Affiliation Email Standard
Brandon Chisham New Mexico State University bchisham@cs.nmsu.edu CDAO
Hilmar Lapp NESCent hlapp@nescent.org PhyloWS
Enrico Pontelli New Mexico State University epontell@cs.nmsu.edu CDAO
Arlin Stoltzfus NIST, U. Maryland arlin.stoltzfus@nist.gov CDAO
Rutger Vos University of British Columbia rutgeraldo@gmail.com NeXML, PhyloWS