School of Computing, Napier University, Merchiston Campus, 10 Colinton Road, Edinburgh, U.K.
The availability, exchange and interpretation of taxonomic information (such as species check lists, distribution and identification data) is of critical importance to taxonomists, ecologists and other biologists and legislators, amongst others. This information is provided by a number of global and local taxonomic database services. These databases hold records, often based on valid scientific names for species, according to their own models of what constitutes a taxonomic 'entity' or concept (i.e. a species or higher level taxon). The need for a common mechanism for the providers of taxonomic information to exchange information with other providers and users of varying expertise in taxonomy was recognized at TDWG Lisbon 2003. Such a mechanism must adequately represent the data as modelled by the owners of the data, whilst facilitating integration with data provided according to different models of taxonomy. The Taxon Concept Schema was developed to meet the needs for an abstract model of taxonomic concepts, which can capture the various models represented and understood by the various data providers. The model is presented as an XML schema document that is proposed as a standard to allow exchange of data between different data models.
This process began with a series of consultation meetings covering the range of Taxonomic data providers and users specifically Vegbank, Taxonomer, SEEK, Berlin Model, GBIF, Nomencurator, Species 2000, ITIS and IPNI and informal discussion with APNI. From these meetings a Strawman Schema (TCS) was developed and presented at a meeting on Taxonomic Exchange Standards held at the eScience Centre in Edinburgh in May 2004 to which all interested parties were invited to attend and contribute. This schema focussed on taxonomic concepts and inherited the Names subschema from the ABCD Schema. The TCS was favourably received although some changes or extensions were suggested which were taken into account in preparation for presenting the draft schema (TCS v0.7) at TDWG Christchurch 2004 which received a mixed reception. Simultaneously an alternative schema (Linnean Core) to TCS was proposed to focus more specifically on name issues particularly as seen by nomenclators, however it was not planned to be developed further and was not presented at TDWG2004.
The main problem with v0.7 of TCS appeared to be that Names were being treated as part of a Concept and some data providers wanted to be able to deal with names independent of concepts. In addition it was felt that not all name issues had been addressed. A Linnean core sub group of the Names sub group was formed to undertake an in depth investigation into the name issues which were currently represented in the Name element and through Nominal concepts in the TCS. As a result Linnean Core was modified and extended and much discussion took place over the relationship between the two. After several revisions of TCS incorporating the findings of the Linnean core sub-group, the current Schema (v1.0) takes what we consider to be the best work from both Linnean core and TCS combined into a coherent schema which we believe can satisfy the needs of all taxonomic name and concept users or providers. The major difference from the schema presented at TDWG2004 being that Names have been elevated to top level elements and include specific valid relationships that can be held between names. The schema still aims to ensure that names are not used where concepts are the appropriate mechanism. Names in the TCS context have no taxonomic meaning other than being the label for a type specimen. Names with a meaning, implied or explicit, are concepts and should be described as such in the TCS. Concepts rather than names should be used for example in identifying specimens. V1.0 of the TCS schema will be presented at TDWG St Petersburg 2005 for voting as a TDWG standard by the members.