2013May29

From FilteredPush
Jump to: navigation, search


Etherpad for meeting notes: http://firuta.huh.harvard.edu:9000/FP-2013May29

Agenda

  • Room Conflict: Need to change meeting time for the Fall Semester (12-1 Eastern 9-10 Pacific?)
  • Project/Package Refactoring
  • Annual NSF project report: proposed deadline of June 30th for all content
  • Progress on SPNHC demonstration
    • Packaging
    • Driver
  • Annotations
    • Progress on rewriting dwcFP, OAD, and example annotations.
  • MCZbase Driver
  • Kepler

Non-Tech

  • Recent Contacts
  • Annotations
    • Annotation MS
  • Collaborations
    • Specify/Symbiota
    • SCAN TCN
      • Niko is looking for a FP/SCAN update this week.
    • NEVP TCN

For Future meetings

Reports

  • Paul
    • Worked with Bob on finalizing revisions to annotation paper.
  • Maureen
    • Finished work factoring out business logic from glassfish deployment projects
    • Working on directly integrating driver code into annotation processor (skipping annotation parser & specifyweb servlets)

Notes

FilteredPush Team Meeting 2013 May 29

Present: Jim, Tianhong, Bob, Maureen, Paul, David, Bertram, James


  • Room Conflict: Need to change meeting time for the Fall Semester (12-1p Eastern 9-10a Pacific?)

Currently looks OK for Jim as well as everyone from last week.

  • Project/Package Refactoring

Maureen: Done, except for documentation.

  • Annual NSF project report: proposed deadline of June 30th for all content

Jim: NSF report due, but needs to go in through the research.gov site (not fastlane) which uses a different template. Have refactored the current content into that template, Sent out a google doc link (is in the FilteredPush folder in Google Drive) . We need to add additional information to update the report. Content from previous years is in blue. Add new material in red. Black and green are the template and instructions.

Target date to have this complete: June 26. Jim needs to submit in early July before he goes out of town.

Bob: FilteredPush Mailing list at cs.umb.edu may be having problems, please let Bob know if a mail that you send out to the list doesn't appear to get delivered rapidly.

  • Progress on SPNHC demonstration
    • Packaging

David: AnnotationProcessor and Specify on a Laptop, everything else on FP3. Maven configuration set up to use spring configuration to deploy to tomcat. Fall back position to deploy entire stack on a laptop in glassfish.

One thing that needs testing and updating are the Symbiota client and PHP helper.

    • Driver

Maureen: Working on integrating the driver into the annotation processor, skipping the authentication servelets.

  • Annotations
    • Progress on rewriting dwcFP, OAD, and example annotations.

Bob: Made small changes, mostly reconciling namespaces, mostly in examples. Manuscript examples and guidance examples (ontologies/oadExamples/*) are pretty much done. Working now on changing competency tests into sparql queries, and then validating them with test data case. Lots of examples were out of date about dwc and dwcFP namespaces.

  • MCZbase Driver

Maureen: Brendan hasn't had a chance to copy the database yet, may this afternoon.

Sven: Trying to restructure the components to make them independent of Kepler and Akka, allowing a maven build to work with either wrapped as services.

Bertram: What functionality has been implemented as Akka ac

Sven: Lightweight actors thus far georeferencing validator, taxon name validator, flowering time validator, run about 5 times faster in Akka than Kepler. Significant improvement by being able to parallelize requests to services.

Bertram: Possible to do the parallelization with a different design within Kepler? Is there a ready made director that would allow this paralellization in Kepler.

Sven: Possibly the tag data flow director, normal PN (including Comad) won't be easily changed to this kind of parallelization. Much easier to start from the begining.

Sven: Probaly ready to look at integration into FP maven in the next couple of weeks.

    • Canadensis library

Tianhong: Have written actors using that library, but aren't getting desired results from testing. One workflow with these actors, another without them, and getting different results. Investigating this.

Tianhong: Draft of some steps: http://wiki.filteredpush.org/wiki/Embedding_Kepler#Scientific_Name_Validator

James: Additional services: ITIS (lots of names, synonyms, references). ITIS has some services that return JSON. See: http://www.itis.gov/web_service.html

Paul: Quality is variable.

James: Need to pick (or grade) higher taxa.

Bertram: is GNI resolver down?

Maureen: Availablity of services may determine ability to curate a record.

Paul, currently seems to be working at: http://resolver.globalnames.org/

Todo: James, Paul, Tianhong to discuss further development of workflow here.

    • Provenance and rendering

David: Both levels of spreadsheet working, and updated validation state separation. Working on export to csv and xls. Also added to queries for analysies (on taxon etc).

Todo: David to put up some screenshots, Perhaps brief demo Friday for Bertram.

James: added on Find Duplicates page some information about consensus. See: http://wiki.filteredpush.org/wiki/Find_Duplicates#Consensus

Agenda for Friday: look at above link for finding duplicates to see what else we need to create tentative specs. Also typing of targets. Non-Tech

  • Recent Contacts

James sent out a brief reply to the Wouter Los and Michael Mirtl: EUDAT project (www.eudat.eu)

Three contacts to follow up on.

Ups priority for documentation.

  • Annotations
    • Annotation MS

Bob: All done with manuscript itself. Picky formatting will be deferred to later in process. Working on actual working competency questions on examples.

Target is to submit tonight or tomorrow.

Paul: Put typing of targets on agenda for Friday.

  • Collaborations
    • Specify/Symbiota
    • SCAN TCN
      • Niko is looking for a FP/SCAN update this week.
    • NEVP TCN

James: Need to think about workshop for CNH meeting.

Paul: Put some of the thinking about desired outcomes on Maureen's plate.

Tianhong: Are result types on wiki current?

David: Most recent json document should work, no additions to page.