2011Nov15

From Filtered Push Wiki
Jump to: navigation, search


Etherpad for meeting notes: http://firuta.huh.harvard.edu:9000/FP-2011Nov15

Agenda

  • NSF Workshop RFP
  • iDigBio presentation update: http://etaxonomy.org/mw/IDigBio_2011
  • Overview of tech meeting discussions
  • Pending Tech Issues
    • API for Query and for Cluster finding.
    • Tech group needs to schedule a time to discuss standardization of (within FP network) services. (For example, the Job Planner needs to find an "analysis" capability; how does it find the service, call the service, what does it get back, what will exceptions look like)
    • Tech group needs to make a decision on or set a date for decision for the query language for pub-sub for Apple Pie.
    • Tech group needs to make a decision on or set a date for decision for the domain objects supported for Apple Pie.
    • Tech group needs to decide on or set a date for decision for the scope, composition, and implementation of the "global cache."

Reports

  • Monday tech meeting on listing out next development tasks:
    • Bob and Paul need to determine how authorization will work in Apple Pie, whether it will be something DataONE can do, or otherwise issue 6
    • Paul has volunteered to create a screenshot mock-up of the proposed MorphBank client changes for "new determination." issue 3
    • Lei has volunteered to design a mock-up of the mapper gui for displaying an annotation as part of her ongoing work on prototyping the mapper issue 7.
    • Maureen is going to give Lei her build scripts for Specify, and also some sample data (she needs to do a smidgin of sample data cleaning first) Lei to set up specify for testing with the Mapper and FP services issue 11
    • Maureen is going to create a very light-weight web client and FP service from the existing demo project, so that we have a basis for testing the integration of the non-FP components issue 5
  • Paul
    • Mockup of Morphbank UI elements for make new determination. See attachements to: issue 3
    • Moved issues from SF Tracker to SF MantisBT, added TODO items from monday tech meeting as Mantis issues.
  • James
    • Submitted an internal grant proposal to prototype a botanical knowledge portal which will use FP as a backbone and produce an API for Semantic MediaWiki
    • Working on iDigBio presentation for upcoming meeting
  • Bob
  • Lei
    • began writing rule and code to parse annotation and test the idea of the mapper

Notes

Filtered Push Team Meeting 2011 Nov 15

Present: Bertram, Maureen, Paul, Bob, Jim, Lei, James

Agenda:

  • NSF Workshop RFP

No discussions yet. In Bob's court.

James: Conferred with iDigBio folks. Questions from iDigBio to be addressed in the presentation. They have a lot of presentations - limited to 15 minutes. Everyone needs to contribute to core messages. See wiki page. Probably needs a separate call.

Jim: In other context (EOL), made a single slide for each item on iDigBio's list of questions. Suggest accumulate a list of bullets for each question.

James: Concur, would be usefull to include a few images.

Jim: Concur, also like the idea of including illustrations.

Bob: Thematic networks, provide a suggestion of how FP could be used by one or more.

Paul: Modify SCAN diagram?

James: Like including Specify, Morphank, etc.

Bob: We don't want to be appearing to make a claim that this is so easy that it doesn't need funding.

Jim: Concerns or chalenges with integration - needs additional support to make it happen.

James: Help to set up a network of nodes, clients, etc.

Bob: All we are funded for are to have the tech personel set up ApplePie.

Paul: Will modify SCAN diagram to illustrate how FP could support a TCN.

James: Like diagram of workflow

James: Bob/Paul Address Ponts 4 (availablity/dependencies) and 6 (technology/standards/protocols)

Bob: Answers provided even if not talked about in slides. Hope outcome of workshop is to look at all the answers to these questions and see if there are functionalities/requirements that are similar in other parts of the iDigBio network, and assess if their choices are suitable/compatable for integration. (for example, pub/sub, is it important to use only one or a few things.).

Action Item: Everyone to address the wiki page http://etaxonomy.org/mw/IDigBio2011 today into tomorrow. James to email out to schedule an additional call this week if he isn't seeing what he needs/has questions by later tomorrow.

James: Strongly recommend including workflow, Quality Control.

  • Overview of tech meeting discussions

Tech meeting on monday. Overview of breakup of tasks following AppleCore ecosystem overveiw.

Decision to use MantisBT, started doing so.

Lei working on Mapper

Maureen working on refactoring prototype to use as lightweight FP service and web client for test environment.

Paul and Bob cleaning up SVN repository.

Bob working on authorization, set up discussion group on GBIF for discussion of authorization issues.

  • Pending Tech Issues
    • query API API for Query and for Cluster finding.

We will talk about this on Friday at the tech group meeting.

    • standardization of services Tech group needs to schedule a time to discuss standardization of (within FP network) services. (For example, the Job Planner needs to find an "analysis" capability; how does it find the service, call the service, what does it get back, what will exceptions look like)

This will probably come up in the meeting on the query API.

    • query language Tech group needs to make a decision on or set a date for decision for the query language for pub-sub for Apple Pie.

Bob: We declare this to be sparql for the moment, and putting in something else shouldn't be a problem if we need it. Should be easy to back out of this decision.

Bertram: Do we have a list of 20 questions that we would ask with such a language.

Bob: We don't but we can easily formulate from our use cases.

Bertram: Can work on this with Bob/Lei on formulating this list.

Paul: Concern with sparql as query language if our implementation of messaging uses JMS and filtering messages using the key-value pairs in the JMS metadata.

Bob: Suggest also formulating the questions as key-value questions, should be easy to interchange with RDF.

Action Item: Bob, Bertram, and Lei to formulate 20 questions from use cases/use case scenarioes (and compose as sparql and key-value criteria).

    • domain objects Tech group needs to make a decision on or set a date for decision for the domain objects supported for Apple Pie.

Put on next weeks' list.

    • global cache Tech group needs to decide on or set a date for decision for the scope, composition, and implementation of the "global cache."

This we leave til the end, after the other network components have been defined, and maybe developed.

    • global knowledge Tech group needs to decide on or set a date for decision for the scope, composition, and implementation of the "global knowledge."

This will be an on-going weekly discussion. Right now, it includes an annotation store, a known-duplicates store, and data providers' data stores. Probably also service registries.

For Friday: We will discuss the Query API (in a general sense). and Annotations and ApplePie rules (what is the content of annotations - data selectors etc).