Science Environment for Ecological Knowledge
Ecoinformatics site parent site of Partnership for Biodiversity Informatics site parent site of SEEK - Home
Science Environment for Ecological Knowledge









 

 

 



Taxon Pre E Science 11 May 2004

11 May 2004 Taxon Working Group Meeting, Edinburgh, Scotland

Present:

  • Paula Huddleston
  • Bob Peet
  • Robert Kukla
  • Trevor Patterson
  • Dave Thau
  • Susan Gauch
  • Jessie Kennedy
  • Nico Franz
  • Robert Gales
  • Dave Vieglais
  • Aimee Stewart

Dave Thau - GUIDs What:

  • short name for complex entity
  • useful for locating info about
  • some items can have more than one GUID (GenBank nos, webURLs)
  • issuance is decentralized for some things

Why:

  • useful for SEEK internal components
  • systems and users external to SEEK
  • integration to other communities (some non-digital)
  • GUID goals: short, permanent, unique, resolvable

Why not use aggregator's ID (ITIS TSN)? ITIS TSNs are similar to GUIDs - minor errors can be corrected with same TSN, author changes get new TSN. SP2000 does not use publicly accessible GUID

What gets a GUID: Taxonomic concept, references, vouchers, data providers?, authors?, journals? Not Name!

When a new concept is added:

  • How do you define a concept for the system?
  • When is a concept new enough for GUID?
  • What minor changes are allowable?

How to get GUIDs? 2 serious candidates:

  • LSID - supports authentication, backed by big boys, uses web services protocols, decentralized, uses dns,
  • Handle system

Bob: Wants to have very simple concept - name, reference, date - no rank, no children, no hierarchy - to GUID. Wants to have "super-concept" all that info with GUID

Nico:

  • Names: "Not available" - suppressed by authority body, changes, resurrected, status changes, transfers?
  • Synonymies: all spellings, some are just misspellings, others different names and views, represented by '=' obscuring differences between the underlying reasons for them
  • TowardsConceptSynonymies

All = relations among names require additional interpretation to be useful as indicators of similarity and difference among synomymous concepts

Synonymies

  • congruent - same circumscription A = B
  • includes A > B
  • is included in A < B
  • overlaps A >< B
  • excludes A != B
  • uncertain A ? B

Revision Differences can be because

  • 1st author never looked at certain species to determine if they are part of the circumscription
  • or 2nd author looked at same species and adopted a different view

Synonymy relationships are directed arrows - directions refer to time. Our lineage relationships are really synonymy. Should be able to go both ways along arrows. Similarity algorithm to expand synonymy relationships.

Bob Peet: See document from SeekTaxon page. Bob will get data to us 1 week after Edinburgh so that everyone can understand the complex issues he addresses.

Susan Gauch:

  • Shawn will give us more complete definition of SMS APIs.
  • Next, developers will populate database with smaller more complete datasets - North American mammals, Bob's, German mosses
  • Short demo of the Taxon development progress - 5 APIs are implemented, those that are used to enter data have been postponed since we have the db populated with the data we currently have available (waiting on data above).

Shawn Bowers and Dave Thau

  • Data discovery prototype
    • resource discovery
    • data transformation
    • data integration
    • data query
  • SMS is glue (using ontological info)



Go to top   Edit this page   More info...   Attach file...
This page last changed on 30-Jun-2004 07:34:50 PDT by LTER.stekell.