Olsen NewNationalThesaurus OfSpeciesNames abs0082

The new National Thesaurus of Species name...by Christian-Emil Ore and Stein A. OlsenThe Museum Project, University of Oslo: 

The new National Thesaurus of Species name... by Christian-Emil Ore and Stein A. Olsen The Museum Project, University of Oslo ..and an event oriented approach for the management of taxonomic information

The Norwegian Biodiversity Information Centre (NBIC) : 

The Norwegian Biodiversity Information Centre (NBIC) A national resource bank for biodiversity information Established in 2004, located in Trondheim Close cooperation with GBIF A primary task is to make a common species names thesaurus for species found in Norway. See www.artsdatabanken.no (Full English version ready 31st October.)

The Museum Project: 

The Museum Project Participants The university museums in Norway (Bergen, Oslo, Trondheim and Tromsø) Natural and cultural history collections Objectives Create a common database platform and interface Multi disciplinary search Digitize information from traditional archives and from the collections Methods Relational database Facsimiles of original documents XML for large texts Event oriented information architecture CIDOC-CRM (ISO 21127) compliant

The National Species Name thesaurus - requirements: 

The National Species Name thesaurus - requirements Given a name used in any (museum) collection, produce the correct valid name (according to Norwegian National Standard) As many synonyms as possible available in the system Names without any clear taxonomic content should be flagged and made unavailable to other users Should reflect a nomenclatural framework based on the international codes of nomenclature for botanic (ICBN) and zoology (ICZN) Conservative approach to changing nomenclature Standardized author names (Botany: Brummit & Powell, Zoology: ICZN) Store historical information about changes in taxonomic considerations and nomenclature

The working process: 

The working process Members taxonomic experts computer experts NBIC The Museum Project representatives from the Norwegian authority of environment. Starting point The Museum project’s conceptual model for a species name thesaurus based on CIDOC-CRM (ISO 21127)

Summary of the requirements and our suggested solution: 

Summary of the requirements and our suggested solution The species bank want a species name thesaurus that reflect the codes, keep track on different kinds of synonyms keep the history of the development of the names. We suggested a event oriented model compliant with CIDOC-CRM (ISO-21127) taxonomic and nomenclatural work are activities taking place in space and time by and done by actors

The CIDOC CRMTop-level Classes relevant for Integration: 

The CIDOC CRM Top-level Classes relevant for Integration E39 Actors (persons, inst.) E55 Types E28 Conceptual Objects E18 Physical Stuff E2 Temporal Entities (Events) E41 Appellations refer to / identifie at within (acc. Martin Doerr)

Scientific activity (Object documentation, collection management etc.): 

Scientific activity (Object documentation, collection management etc.) (acc. Karl Heinz Lampe, TDWG 2005)

The extension of this model towards taxon names and concepts: 

The extension of this model towards taxon names and concepts Parent taxon Involving in a Specified way (may be) Supported by is documented by Involving in a specified way

The respond to the suggested model: 

The respond to the suggested model Hard to understand. Complex. Do we really need this? All we need are names, and some relations between them. The job is to find the right relations between the names We don’t need any information about types (type specimens)

The questions that came up and the process: 

The questions that came up and the process How do we keep track on the different kind synonyms (homotypic and heterotypic)? A name can denote different taxa during its history. We need some notion about the taxon, not only the scientific names. Names and taxon are not one-to-one. We also need information about popular names

Information about names initial sketch: 

Information about names initial sketch Latin names relations -Name_id -name -Parent name -author -status -type_id Problem: How to ekspress homotypic and heterotypic synonyms?

Information about names, types and taxa(keeping track of the synonyms and common names): 

Information about names, types and taxa (keeping track of the synonyms and common names) Latin names Types Taxon Valid name -Name_id -name -Parent name -author -status -type_id Homotypic synonyms: Names belonging to the same type Heterotypic synonyms: Names belonging to the same taxon, but different type -type_id -taxon_id -taxon_id -valid_name_id Common names -popular_name_id -taxon_id Parent taxon_name

How to model different meanings (concepts) of a name at different times?: 

How to model different meanings (concepts) of a name at different times? Takes part with different roles Types of events: Splitting, change of meaning of name, Synonymization, change name

The CRM-compliant event oriented model that was presented in the beginning: 

The CRM-compliant event oriented model that was presented in the beginning Parent taxon Has (valid) Involving in a Specified way (may be) Supported by is documented by Involving in a specified way

What are the main differences between the two models?: 

What are the main differences between the two models? The CRM model is event oriented All operations go via events that incorporates taxa, specimens (types) and names with different roles. In the NBIC model the objects (names, types and taxa) can take part in events, like splitting, lumping, change of name, etc... The role of the type specimen In CRM: The type specimen is connected to the event that creates a new taxon. In the NBIC model: The type specimen is connected to the name The documentation In the CRM compliant model: Documents are connected to the events (one or more) In the NBIC model the document related fields are “hard coded” into the event, so there can only be one document connected to an event The parent taxon (-name) In the CRM compliant model the parent taxon is connected via taxon In NBIC model the parent taxon is connected via the taxon name

Conclusions: 

Conclusions The NBIC model can be mapped to the CRM compliant model The connections in the NBIC model can be derived in the CRM compliant model The NBIC model fulfills the requirements It makes it possible to difference between different kinds of synonyms It makes it possible to keep track on changing of names It makes it possible to detect pro parte synonyms via split events It involves vernacular names We think it is possible to capture all kinds of events taking part in taxonomy work, but it depends on how the different operations are being implemented But still: The Norwegian taxonomic environment and the computer scientists that work with biodiversity would benefit significantly on the TDWG-work on taxon names, concepts and ontologies. So I’ll see what I can do...thanks!