Taxon WG Conference Call_4_March_2004

SEEK Taxon Working Group Conference Call, 4 March 2004, 12PM-1PM US EST

Participants

• Present: Kennedy, Peet, Stewart, Gales, Jones, Franz, Liu

Discussion

• Bob reported that the Taxon Working Group will meet May 10 and 11 in Edinburgh, prior to the E-Science meeting.
• Aimee, Robert, and Dave Thau have a draft of the TaxonPublicInterfaces diagram on the wiki and on CVS. A working glossary and definition of parameters to go with the diagram is also present. This diagram handles all of the old use cases and Nico's first draft. With the next version of the Use Cases, they will append a high level implementation plan to each case. They are currently working on high level implementation of each public interface, and the objects and methods needed for the entire system. This should be done pretty quickly, the next use cases document will point out whether we have missed any functionality, and we can re-commence coding. Problems with Python (a bug in the garbage collection process and an immature SOAP library) neccessitated changing languages. Java was chosen for its rich selection of libraries, ease of integration with Globus, garbage collection, etc. This gives them the opportunity to re-engineer in a more modular, flexible, abstract way.
• Jessie has been working on the Taxonomic Schema Standards
• Jessie and Robert Kukla have met with Donald Hobern of GBIF and Walter Berendson in Berlin. Both are doing a lot of similar work to SEEK. The Berlin group's code is all open-source. They are also trying to establish a standard with the ABCD schema. Some duplication of the DiGIR efforts is happening here.
• Data Sets
• We now have mosses and higher plants data from the Berlin group.
• We haven't received data from Richard Pyle yet although we have been in touch.
• Robert makes data sets available on the internet: http://146.176.162.1:10102/data-sets/
• Schema Mapping tools
• ABCD has an open-source schema mapping tool (in Python). It assumes a one-to-one relationship between the attributes of tables in the database and the elements in the XML schema.
• Species 2000 has Litchi (sp?) concept mapping tool - Jessie to find out more when she visits F Bisby's group.
• SEEK - student Jenny Wang under Joseph Guogen (sp?) has created a schema mapping tool (can get details from Matt and Bertram)
• Paula Huddleston at ITIS is still interested in collaboration, and is interested in meeting in Edinburgh. Matt said that we should be able to provide her with travel funds. Jessie will inform her of this and carry out tel call meetings prior to the edinburgh meeting.
• Jessie has plans to meet with people from Nomencurator (London), Species 2000 (Redding, England), and this then leaves ITIS in Edinburgh and APNI/IPNI to whom she has still to organise a meeting.
• GUIDs was one topic of discussion at these meetings. Reaction was not enthusiastic but if it could be done, will simplify life, GUIDs for pulbications, specimens, etc. GBIF is interested in getting publication info with GUIDs. Berlin hasn't used GUIDs, but has used ID for name, ID for publication, and concatenates the two for a concept ID. Political resistance may be an issue with main data providers, but this hasn't been the case so far with those visited.
• Bob
• Traveling to DC for NatureServe meeting next week to discuss interchange between Biotics model and Vegbank model.
• Plant Concepts for ITIS North America - now have about 7000
• Xinhua - has a summary of taxonomic data models mapping concepts, wants to move back and forth between IOPI, NatureServe, and Vegbank
• Nico - working on Use Cases, posted a document "ClassifyConcepts" on the SEEK Taxon Wiki that lists "concepts" according to (1) how much information content they carry, and (2) how much intellectual credit is associated with transferring them from one location to another.
• Jessie - MartinG will be working on visualization of concept data - compare larger structures of concept information, and explicit relationships between concepts. Thau suggested looking at Growl browser - to show networks, nodes.
• Dave Thau
• Implementations of GUIDs - DOI handle vs LSID system. LSID (ver 1.01) is less mature than handle system and there are some inconsistencies between clients and servers. LSID does not have to be maintained on the domain server but has responsibility for pointing to domain name with DNS.
• Working with Shawn on prototype of Taxon / SMS interaction
• RobertK, Thau, Matt, Jessie, Rich are to build a prototype GUID handle system
• Everybody please check TaxonMeeting_23_January_2004 and don't forget the tasks you are responsible for.
• Jessie will contact group leaders about E-Science registration for SEEK participants

Next Teleconference