GUID Use Cases

Biological Specimens

One of the key drivers for GBIF and TDWG to develop a GUID solution is the need to be able to refer reliably to specimens held within biological collections. The following use cases describe some of the situations in which they are needed:

• FindingSpecimens - retrieving data on a specimen even if the specimen or data are moved to a new location
• DetectingDuplicates - recognising when multiple data records reference the same specimen
• TrackingSourceRecords - recognising the source when aggregators have added value to a data record
• TrackingRecordCaching - tracking what services are caching or aggregating data harvested from a data provider
• IdentifiyingPreparations - recognising when the same specimen is available in different preparations
• IdentifyingDatasets - identifying datasets or individual data records used in analyses, reports or publications

The following links describe not an use case per se, but rather real scenarios of use of identifiers at collections:

• Use of identifiers at Missouri Botanical Garden
• Use of identifiers at Kew Botanical Gardens
• Use of identifiers at the Museum of Vertebrate Zoology - University of California at Berkeley
• Use of identifiers for germplasm - Nordic Gene Bank and description of the Germplasm Identifiers.

Taxon Concepts

Another important area of data management will be in tracking references to the taxon concepts in use by different authors. Potentially at least three different items may require some form of globally unique identifier: taxonomic/nomenclatural publications, taxon names and taxon concepts. The following use case describes the need for GUIDs for taxon concepts:

• ResolvingTaxonConcepts - determining whether different uses of taxon names refer to the same group of organisms
• IdentifyingTaxonomyForIdentifications - indicating the checklist or taxonomic revision used for identifications

Identification, analysis and descriptions

• Finding and retrieving descriptions
• Assessing comparability of terms for variable (character), value (e.g. categorical state), or modifier (e.g. "perhaps") concepts.
• Finding and retrieving definitions of descriptive terms

Schema Elements and Controlled Vocabularies

'GUIDs are also critical for managing schemas and their components (such as "concepts" in the DarwinCore?).' -- Comment from DaveVieglais (2005-10-21 01:12:01)

As Dave notes, another significant area in which GUIDs may be used is to provide better mechanisms for ensuring that collaborators in data exchange are using the same elements to structure their data (and also a shared set of possible values for any element). GUIDs can therefore help to form the semantic underpinning of biodiversity data exchange.

• ReferencingSchemaElements - ensuring that data exchange is based on a shared ontology
• ManagingControlledVocabularies - managing permitted values for elements used in data exchange

Data and Literature

GUIDs based ontologies will strongly improve scientific work by helping to find additional information which relates to a taxon, concept etc.

• FindingData - GUIDs can lead to data sets which relate to a distinct taxon (eg. observatory data, morphometric measurements etc.)
• ReplicatingData - GUIDs can support replicas of data (specimen data, observations, etc.) that are located in multiple different locations on the Internet for purposes of load balancing and failsafe retrieval
• FindingLiterature - Most electronically published documents do already use GUIDs (DOI)

Ontologies

Maybe this will be the most important use case in the future? Linking taxon names, specimen, taxon concepts, data literature etc by using GUIDs.

Annotations of composite multimedia objects

I have in mind JPEG2000 images, audio and video, but this is probably more general and no doubt GUIDs on components of compound objects are well discussed in the ontology literature. Here I write only about the specific needs we have arising in AnnotatingJPEG2000Images?. BobMorris

Categories

Comments

Posted by JessieKennedy - 2006-01-24 23:52:25

Looking at the use cases for specimens such as FindingSpecimens; DetectingDuplicates; TrackingSourceRecords; TrackingRecordCaching; IdentifyingDatasets - I would've thought there would be similar use cases for Taxon concepts.

I believe use cases like FindingSpecimens and FindingTaxonConcepts? are too much developed within a closed system. I believe we have to distinguish between a closed system, using custom-built software, and an open system involving other biodiversity and information players.

• Indexing engines like Google should be able to find a specimen.
• A specimen should be referable in a wiki
• A specimen should be referable in a journal (printed, online-pdf)
• A specimen should be referable in a Genbank record,

etc.

-- GregorHagedorn - 18 Jun 2007