Main aspects of the Species Profile Model and the TDWG architecture
Andreas Kohlbecker, Markus Döring, Andreas Müller
Abstract
The Species Profile Model (SPM) was created at the GBIF Species Model Workshop in Copenhagen, on April 16-18, 2007, to support the retrieval and integration of species data.
The SPM describes a root element referring to a TCS (Taxon Concept Schema) taxon concept, under which one or more facts (InfoItems) about the species can be listed. Each fact belongs to a category drawn from a set of controlled terms.
The aim of the SPM was to provide a flexible data model which allows for naive implementations, in which InfoItems can be freely tagged with terms from controlled vocabularies independent of one another.
However, the free tagging approach leads to problems when the SPM is used for data exchange, e.g., when using the SPM with TAPIR. The document resulting from a TAPIR request would consist solely of InfoItem elements having no child elements. To get the category for an InfoItem, TAPIR would need to send additional requests to retrieve the subelement category for each of these. Therefore, free tagging was replaced by a subclassing concept, in which a set of specialised InfoItem classes, each inheriting from an InfoItem super class, are created for each fact category. Documents now contain instances of these InfoItem subclasses, each named according to the category of data it contains.
At this time during the evolution of the SPM, it is crucial to shed light on how and whether the current version of SPM fits the needs of the use cases in scope. Identifying possible problems in a timely fashion will help us decide where to go from here.
The SPM describes a root element referring to a TCS (Taxon Concept Schema) taxon concept, under which one or more facts (InfoItems) about the species can be listed. Each fact belongs to a category drawn from a set of controlled terms.
The aim of the SPM was to provide a flexible data model which allows for naive implementations, in which InfoItems can be freely tagged with terms from controlled vocabularies independent of one another.
However, the free tagging approach leads to problems when the SPM is used for data exchange, e.g., when using the SPM with TAPIR. The document resulting from a TAPIR request would consist solely of InfoItem elements having no child elements. To get the category for an InfoItem, TAPIR would need to send additional requests to retrieve the subelement category for each of these. Therefore, free tagging was replaced by a subclassing concept, in which a set of specialised InfoItem classes, each inheriting from an InfoItem super class, are created for each fact category. Documents now contain instances of these InfoItem subclasses, each named according to the category of data it contains.
At this time during the evolution of the SPM, it is crucial to shed light on how and whether the current version of SPM fits the needs of the use cases in scope. Identifying possible problems in a timely fashion will help us decide where to go from here.