Filtered Push Meeting 2010 Jan 11
Agenda 1) Welcome for James Hanken.
2) Progress on developing workplan
Sequence diagram for commincation between three nodes
http://www.etaxonomy.org/mw/3nodes See: http://www.etaxonomy.org/mw/2011Jan04#Workplan From the Jan 4 meeting:
Evaluate new network architecture, examine path (incremental?) to three node architecture (This Friday).
Bob, Zhimin, and Paul have produced a sequence diagram for communication between three nodes. This informs requirements and goals for a three node network.
Bob: We need a usecase that is orthogonal to the current ones, map a workflow usecase onto the three node network and produce another sequence diagram. Paul: Workflow as service, or as client? Bertram: Both are possible, we have our specimen merge example, perhaps design or develop a workflow that could be invoked in the network. Discussion. Plan: Plan Friday, Paul, Bob, Zhimin meet to elaborate (1) a sequence diagram for a workflow usecase where Kepler is a network client and (2) where a workflow is embedded in the network.
Establish work area for examining data mapping and transformations from local collections databases to ABCD and DwC.
Paul, Zhimin started on some data to put in such a place. The canadians in Canadensis, and the Consortium of Northeast Herbaria are working on specific cases of mapping data from local databases onto darwin core. TDWG DarwinCore provides general guidance for field mappings, but James: The Canadensis folks are interested and there is a potential for developing a place for documenting particular examples of database designs and data mappings. May even fall under the domain of TDWG. Bob: Relevant to look at RDB to RDF. Jim: What range of examples of collections databases are we using? Paul: BBG data, Specify6, GBIF cache. James: We could do with an expanded range. Jim: Expanding range might be a route towards getting the software used. Paul: Perhaps good opportunity to begin discussion with people who have collections databases that we would like to interact with and request schemas and example data from them.
Discuss queries to the network.
Put off to discussion of the sequence diagrams.
3) Development priorities
(a) Bringing GBIF cache live - we need a capability to flag annotations as sandbox or intended for retention. Bob: We need to think about if this is a concern of the Annotation Ontology. Might be an automatic application of an annotation to an annotation where the annotation specifies management criteria - a metaanotation carrying management informaiton. We need to examine if AO will handle this use case. (b) Develop revised workplan. (c) Focus the Davis team on schema mapping
(d) Initial target of a three node network.
4) Exchange schema to database bidirectional mapping Paul to talk with Linda about getting a subset of data from one of the MCZ Muse databases for Bertram to explore data mapping and to use in one of his courses. 5)Project Admin. james ask TACC if we can use WebEx weekly. Bertram: (on travel Feb 16-18) Specimen database, schema mapping [ECS-166 datasets: using data_Darwincore.xls from Lei/Zhimin] downloaded mysql-schema.sql (typo in 'PrimaryDivision' .. missing quote) [ looking for a data modeling tool: ER, UML, ... ] ==> DB schema and instance from Paul ==> DRUID
Sequence diagram of a query in three nodes network Notes: http://firuta.huh.harvard.edu:9000/FP-2011Jan11