2011Sep27

From FilteredPush
Jump to: navigation, search


Etherpad for meeting notes: http://firuta.huh.harvard.edu:9000/FP-2011Sep27

Check email for new Webex link. No Webex setup yet - failover to Skype.


Agenda

  • TDWG Demo
  • Architecture
  • Rules and configuration for standard Annotations for AppleCore
  • Symbiota
  • Possible collaboration with ADBC TCN's

Reports

  • Paul
    • Some more work on Bob's draft of annotation manuscript.
  • Lei
    • Studied the owl syntax, reasoner, swrl (semantic web rule language) and some owl processing softwares (Jena, Pellet) to explore how to define the rules about what the annotation will be expressed and in which way. These rules will be used in mapper to validate the incoming annotation and extract the required information.
  • Bob
    • Discussions about AO/AOD with Alex Garcia Castro soon to be at FSU
    • Began SWRL rules for DwC-based domain vocabularies in data annotations using AO/AOD. See DwCSWRL


Notes

FilteredPush Team Meeting 2011 Sept 27

Present: Bertram, Jim, Maureen, Bob, Paul

Agenda:

  • TDWG Demo
  • Architecture
  • Rules/Configuration for standard annotations for AppleCore instance.
  • Symbiota
  • Possible collaboration with ADBC TCN's

Notes:

  • TDWG Demo
  • Architecture
  • Rules/Configuration for standard annotations for AppleCore instance.
  • Symbiota
  • Possible collaboration with ADBC TCN's

Possible collaboration with the ADBC TCNs

Had received an overture from some southwest museums interested in a TCN, and have been discussing particulars of deliverables for developing a budget. However, they have been discussing budgets with NSF and whether they could fund us. Since our work does not involve scanning and digitization, it is unlikely our work would be funded to the extent we originally envisioned. The scope of work and budget for FP in this TCN proposal remains to be determined.

There is another similar group with a similar proposal, only northeast herbaria rather than southwest museums. Interestingly, both groups came independently to similar sets of target tools and technologies, Specify/ Morphbank/ Symbiota.

Hence the presence of Symbiota on the agenda today.

Anything relevant from meeting with Bob and Greg in Florida last week? One of the people that worked on AO will be working with Greg. Both Bob and Greg will be going to a semantic web workshop in Bonn after TDWG.

Symbiota

Replacing work we originally thought we would do in IPT? If so we could offer a FP annotation generation client built in to Symbiota. We should discuss how to integrate FP and Symbiota.

Bob and Paul will talk about a tentative technology stack, talk with James about AppleCore.

Network Architecture

Structure of messages. Since messages have to be passed in the demo we should revisit the message architecture. The question is "what is the next step that is needed to move forward in work for the network?" We need to know what the APIs are and who is supposed to implement them. We should revisit the sequence diagrams to see if they are sufficient, and revisit the classes to see that they are all there.

Bertram: Sequence diagrams = use case diagrams, or architecture diagrams? (neither, there are some diagrams that show network components as "columns" and flow of messages going between them, back and forth, to demonstrate one whole use case... there might be some in the wiki...http://etaxonomy.org/mw/File:Draft_login_sequence_diagram.png)

http://etaxonomy.org/mw/File:Henrys_story1.jpg

http://etaxonomy.org/mw/File:Analysis_sequence.png

http://etaxonomy.org/mw/File:Annnotation_Sequence_Proposed_in_network.png

We've identified messages and the messaging system as the next point of discussion for the architecture. Do we have the APIs and the messaging system properly composed?

Perhaps we could also talk about administration of the network nodes, and the format of global knowledge.

SWRL: http://en.wikipedia.org/wiki/Semantic_Web_Rule_Language

We expect to be collection annotations that assert new identifications and new georeferences. These fall between the annotation and domain vocabularies. There isn't anything in DWC that describes what should go into a new determination. There are terms that can be used, but no other specification. We could use rules to fill in that gray area.

Another (naive) question: what parts of AppleCore are being formalized Anticipated typical annotations, e.g. New Determinations, Georeferences, Changed Taxonomies, New specimen record on table.? Vocabulary terms and how they relate? The model itself? And what inferences are we making from the formalization?

Bob will send Lei the dwcFPModel.owl file.