2011Aug16

From Filtered Push Wiki
Jump to: navigation, search


Etherpad for meeting notes: http://firuta.huh.harvard.edu:9000/FP-2011Aug16


Agenda

  • Connecting Harvard and UCD grant admins
  • Progress on and needs to advance the TDWG Demo
  • Progress on an architecture implementation
  • Progress on the mapper

Reports

  • Maureen discussed FP network implementation for the demo with Lei, Bob, and Bertram for the Friday tech call. She is getting a Kepler web service running with help from Lei.
  • Bob more work on AOD manuscript and diagrams. See AOD_Diagrams. Participated in Friday Tech call.
  • Lei: studied the subscription/notification solution proposed by Maureen and discussed it in Tech meeting.
  • James: Met with Paul Hebert, Bob Hanner and Sujeevan Ratnasingham at the Univ. of Guelph to discuss DNA barcode identification services. They were very interested in discussing an API for their informatics platform BOLD. James was placed on a newly formed Advisory Committee for BOLD to facilitate this and other collaborations.

Notes

Filtered Push Team Meeting 2011 Aug 16

Present: Lei, Bertram, Paul, David, Bob, Maureen, James H., James M., Tim

Agenda:

  • Connecting Harvard and UCD grant admins
  • Progress on and needs to advance the TDWG Demo
  • Progress on an architecture implementation
  • Progress on the mapper

Notes:

  • Connecting Harvard and UCD grant admins

As of yesterday (Monday), grants officers at the two institutions had exchanged messages and are making necessary arrangements. Harvard's OSP is amending the original subcontract for second year funding.

  • Progress on and needs to advance the TDWG Demo
  • Progress on an architecture implementation
  • Progress on the mapper

Connecting Harvard and UCD grant admins: see notes above by Jim.

TDWG demo:

Lei provided Maureen with instructions on how to get the Kepler web service running. Maureen is following the instructions. There is a data set from the previous TDWG demo that Lei is looking for; she found a link to a wiki page but not the data. Lei will email the link to Paul and Paul will find the data set. Maureen will get the Kepler web service running and then work on the APIs between Kepler/FP.

Bob has a question that arose during Friday's call about what things will be FP messages. There are things that will be messages aside from annotations. There was an enumeration of message types in the prototype, but Bob didn't find it in the new architecture. Bob and Paul will look for it after the meeting.

Tim's work on the mapper: looking at the DWC star schema and its implications. It's probably a good time to re-evaluate our intent for the mapper, perhaps on the Friday tech call. Should we start making an implementation of Tim's demonstration, or should we have the mapper do something else?

Bob thinks we will be getting annotations with fragments of DWC. The annotation and data providers will not be obliged to compliance with much, the mapper will have more of that burden. Tim: ideally, data providers would only have to know how to map from their schema to DWC. Incoming changes would be represented to the data provider in DWC; the mapper would help find the annotated record in the database, show the user how the changes would affect the record, and store the changed record. The hope is to not require Java code on the part of the data provider.

Bob: that corresponds to our notion of annotations.

Tim: people can make annotations in different vocabularies, as long as you can map them to DWC. DWC seems to be the common language.

Bob: perhaps not for Apple Pie, but we might see more use of vocabularies from the OBO community (the people who are doing phylogenetic analysis). The annotation architecture agrees with Tim's understanding.

Tim: much of the schema dependencies could be handled with configuration

Bob: we could come up with a non-DWC example

Jim's questions for TDWG demo:

Demo sessions are poster sessions, which we attend for a specified period of time? Yes. Lei is the logical person to do the demo. We will also do a talk, which may lead to a demonstration, so we may be able to do a "canned" live demo in this case. [Bertram: Lei might not be available, in which case Tim could give the demo; Bertram has a conflict with the DataONE AHM and ProvenanceWG meeting there, so can't come either :( ]

Should we set advanced deadlines for a code freeze prior to the demo? Yes. Monday Oct. 3 is two weeks before TDWG. Sept. 26 is three weeks before the TDWG demo.


Abstract deadline is Sept 9th, see: http://www.tdwg.org/conference2011/submissions/

James: met with some people who are interested in an FP API. CBOL James is working on Apple Core in Montreal for the next week.

Bob: there will be a TDWG knowledge organization systems session.