Originally this page was entitled Use Cases for the Global Names Index but has been amended to accommodate use cases that have multiple options for achieving the goal that do not require the Global Names Index (GNI).

• Provide links to species pages hosted by GBIF Participants within the GBIF data portal
• Implementation Steps
• GNI API Access
• Provide links to species pages hosted by GBIF Participants
• Dynamically embed verified links to selected biodiversity web resources
• Use Cases NOT supported by the GNI
• Resolution of index URIs
• Nomenclatural or Taxonomic syntax or semantics

Provide links to species pages hosted by GBIF Participants within the GBIF data portal

Goal: Provide a simple mechanism for members of the GBIF network to access and embed links to species pages hosted on sites of other GBIF participants within the GBIF data portal.

Example: The GBIF data portal currently provides a "taxon summary" page for nominal taxa that provides access to indexed primary occurrence data and some taxonomic information derived from a few select sources (primarily the Catalogue of Life). See http://data.gbif.org/species/13808249/ for an example of the African Lion, Panthera leo. GBIF is a network composed of many participating countries and organisations, many of which have their own biodiversity data portals and information relating to taxa. The discovery function of the data portal could be extended by providing links to those related pages from the taxon summary pages on the GBIF data portal.

The basic data requirements to meet this goal are for each GBIF participant seeking to publicise a species page to provide a list of taxon names or identifiers, and one or more URIs that point to human or machine-readable content. This requirement can be met by publishing the data as a basic species checklist that includes the GNA Alternative Identifiers extension.

This approach enables

• Indexes of species pages to be published as simple species checklists, a feature already supported in the IPT
• GBIF portal developers to directly access these files via existing methods
• Same solution meets the requirements of the Global Names Index
• File-level access to entire resource index
• Simultaneous publication of additional species data including higher taxonomy, synonymy, vernacular names and other content types supported by DwC + GNA extensions

Implementation Steps

1. Verify and Publish and document the GNA Alternative Identifiers extension
2. Register the extension
3. Host a sample participant data set in an IPT instance (NLBIF example)
4. GBIF develops a HIT indexing adapter for adding the data to the GBIF data portal
5. Reconcile sources to GBIF nub taxonomic identifiers
6. Portal modifications add a "species page links" panel to taxon pages.
7. Develop and Circulate document on "Linking Participant Species pages to the GBIF data portal"

GNI API Access

The GBIF data portal use case is not well supported by the Global Names Index which provides a lexically de-aliased index of links to species information. The main problems are:

• Access to species page links via the GNI is through the API. The API supports species-level queries. Direct access to published DarwinCore Archives and GBIF processing and indexing of "GNI indexes" is preferred to iteratively accessing indexes via web services either synchronously or asynchronously.

• The GNI registry currently doesn't support the means to filter a subset of registered resources for accessing.
• GNI metadata is limited. No means to tag resources for subsequent conditional access and selection ("Only give me links to species pages served by GBIF Participants for this taxon")

Provide links to species pages hosted by GBIF Participants

Goal: Provide a simple mechanism for members of the GBIF network to access and embed links to information related to a selected taxon that originate in non-member web portals and sites.

Example: The GBIF Strategic Plan calls for utilising name indexes to link information relating to taxa across disparate domains of biology. For example, a species page within the GBIF data portal currently displays data resources representing primary occurrence data. Additional information related to that species exists within other accessible data portals. For example the NCBI Genbank may hold molecular sequence data on a selected taxon. The GNI provides a simple way for building "linkouts" to other information resources.

Overall sequence as above

Requirements: Additional refinement of resource selection via the GNI API

Liabilities: Access via taxon-level API is CPU and time intensive.

Dynamically embed verified links to selected biodiversity web resources

Goal: Provide a web service that allows links to information about taxa to be dynamically embedded into a full-text web document via a species (or taxon) name in order to replace manual and less effective markup.

Example: Some academic journals (BioOne) provide links to the Integrated Taxonomic Information System (ITIS) by embedding a URI-based query-by-name to ITIS from the corresponding scientific name appearing in a publication (i.e., a scientific name in a article is hyperlinked to ITIS with a query to return information related to the name). This method often returns a negative result because ITIS holds no information on the taxon. Using the GNI, ITIS can publish an index that provides an explicit link to taxon information so that only taxa treated within the database are listed. The publisher can utilise tools such as LinkIT or the related NameLink services to automatically identify species names within an article, check the ITIS index, and only if a taxon appears in ITIS (or any indexed resource) embed a link to the ITIS site.

• ITIS exports a DwC archive file according to the GNA specificiation
• ITIS registers a GNI resource via the GNI interface and specifies the location of the DwC Archive via URI
• The GNI parses and validates the file and extracts and indexes the GNI specific data elements
• A publisher accesses the ITIS identifier via the GNI documentation and web services
• The publisher provides interfaces (synchronous for real-time rendering or asynchronous for pre-indexing) to utilise the NameLink service to dynamically embed links to ITIS within it's published online documents.

Use Cases NOT supported by the GNI

The Global Names Index currently does not support the following use cases

Resolution of index URIs

There is no requirement for a URI tied to a name to return any consistent output when it is resolved. A GNI URI may be linked to a web page, SOAP service, LSID returning arbitrary RDF, or any other conceivable response.

Nomenclatural or Taxonomic syntax or semantics

The GNI attempts to perform lexical de-aliasing of names (i.e., names are algorithmically grouped according to text-string similarity) so that misspelled names may be concatenated in a search and resources tied to homonyms can be kept separate in search responses.

• The GNI does not indicate if a name is correctly spelled.
• The GNI does not provide information regarding objective or nomenclatural synonymy
• The GNI does not provide information regarding taxonomic synonymy
• The GNI does not link vernacular names to scientific names
• The GNI does not provide higher classification information for a taxon name with the exception of the optional inclusion of a dwc:kingdom within the index.