2011Jul26

From Filtered Push Wiki
Jump to: navigation, search


Etherpad for meeting notes: http://firuta.huh.harvard.edu:9000/FP-2011Jul26


Agenda

  • Project Programmer Position - who to advertise where?
  • TDWG
  • Path for Architecture to network implementation
  • Evaluation of embedded Kepler over prototype FP for TDWG demo
  • Mapper

Reports

  • Paul
    • Project programmer position is now publicly visible.
    • Got small subset (~110 records) of HUH data to Tim.
  • Lei
    • Wrote abstract about kuration package for IDCC11 (International Digital Curation Conference) and will circulate it soon in the FPush project group
  • Bob
    • Worked on exposition of Paul's AOD examples about multiplicity. Removed unnecessary OWL Declarations inserted by Protege.
  • Tim
    • Successfully tested a general strategy for performing selects and inserts on a collection database from Java in a schema-independent manner.

Notes

  • Project Programmer Position - who to advertise where?

Bob: Likely targets: Tdwg list, taxacom list.

Everyone should seek out potential candidates known to them personally.

Bob: Mailing list of former colleagues, most engineers in boston area. Will post at UMass, most have probably graduated or not available. Will contact a couple people known personally.

Paul: I'll take TDWG content and taxaxom.

Bertram: DBworld mailing list with job related sublist. Scientific workflow mailing list.

  • TDWG

Bob: James and Bob both want to participate in the invasive species activities, seeking to get the organizers to avoid conflicts between invasive species and annotation activities.

Paul: James has been moving the organization of a session and the annotation interest group working session with the TDWG meeting organizers.

  • Path for Architecture to network implementation

Maureen: The BOUML generated code is sparse and many turned into abstract classes rather than interfaces. How data is discovered by the network, how people join the network.

Bob: Going back to the generated code: The abstract classes might be an issue with the BOUML configuration. Motivation being to keep the BOUML diagrams consistent with the code.

Maureen: We can take a look. Bouml generated code doesn't build appropriate packages.

Bob: it might be that we can regard the round trip as nice idea that didn't work.

Maureen: Cleaned up code, made build script, documented.

Paul: Bob and maureen to spend an hour evaluating if BOUML round trip should be abandoned and code should be considered as authoritative.

  • Evaluation of embedded Kepler inside prototype FP for TDWG demo

Lei: On schedule for this week. Probably no big problems.

Paul: Lei to direct questions about analysis interface to Maureen, Paul and Bob.

Target review of this on Friday this week.

  • Mapper

Tim: Relaitvely easy to do a sequence of high level operations on the database that are mapped to sql specified in a sql document. Object models for this domain are very complicated. Only putting data into the databases and updating, not doing much in our own code, don't need an object model, just a means of transforming the data. XML document defines steps, with place holders for keys. Seems to be working with the toy database. Next step to do same set of java operations on a different schema. Code itself is pretty simple. XML document lists various sql statements corresponding to particular operations. Might need list of operations to carry out. Inserting a collector name in one database might be one table, might be serveral table in another database. Complexity not in java.

Would be good to review this code on Friday.

Paul: Place to create on the svn repository: in: http://filteredpush.svn.sourceforge.net/viewvc/filteredpush/ create FP-mapper with branches/tags/trunk.

Tim: Would it be possible to put BSD licences on the mapper code from Davis?

Paul, Bob: Should be possible. We'll need to look back at deliverables in the grant proposal.

Bob: Question for Jim; an ontology example is derived from email sent to Jim, can we get permission to incldue the email text.

Jim: Forward the text and I'll pass it on.

  • Abstract for 7th International Digital Curation Conference

http://www.dcc.ac.uk/events/idcc11