-
Notifications
You must be signed in to change notification settings - Fork 4
ClassIndividual
This class does not exist in the current Darwin Core (DwC) standard. There is a pending proposal (currently stalled) to add a class to DwC called Individual which would be the object of the existing term dwc:individualID (http://rs.tdwg.org/dwc/terms/index.htm#individualID), would accommodate resampling of the same individual or group over time, and would facilitate linking multiple Identifications to the same Individual.1 In this ontology, we have called the class (which corresponds to the proposed dwc:Individual) "IndividualOrganism"2 to prevent confusion with an "individual" in the sense of an OWL ontology (i.e. "objects in the domain in which we are interested"; see p. 10 of A Practical Guide to Building OWL Ontologies Using Protege 4 and CO-ODE Tools3, Edition 1.2).
Mirroring the use in dwc:individualID, an instance of the IndividualOrganism class could be a biological individual or a small population of the same kind of organism. Allowing IndividualOrganism instances to include small populations is a concession to the fact that it is sometimes inconvenient to distinguish or not possible to tell where one biological individual ends and another begins (e.g. clones, colonial organisms, densely packed populations). It also mirrors the common practice of collecting several biological individuals in the same sample (e.g. several grass plants in the same bundle or a clump of moss) or collecting individuals from the same small population and calling them "duplicates".4
The original intent when the DwC class was proposed was for a dwc:Individual to represent a particular taxon at the lowest level (i.e. having a homogeneous taxonomic identity). In the discussion that took place on the tdwg-content email list from Oct-Nov 2010, it was suggested that the class be more broadly defined to allow heterogeneous mixtures of biological individuals that would be identified to the lowest shared taxonomic identity. We suggest an approach that may accommodate such heterogeneous entities (See the wiki page TaxonomicHeterogeneity and the section below on "Views regarding the definition of 'Individual'" for more on this.)
1 http://lists.tdwg.org/pipermail/tdwg-content/2010-October/001675.html
2 http://lists.tdwg.org/pipermail/tdwg-content/2010-November/001870.html
3 http://owl.cs.manchester.ac.uk/tutorials/protegeowltutorial/
4 http://lists.tdwg.org/pipermail/tdwg-content/2010-August/000089.html , http://lists.tdwg.org/pipermail/tdwg-content/2010-August/000075.html , and http://lists.tdwg.org/pipermail/tdwg-content/2010-September/000056.html
The class IndividualOrganism has its origin in the need to facilitate the resampling of organisms. In traditional museum-based biodiversity informatics, sampling was accomplished through the collection of preserved specimens. Since the entire organism was often collected, resampling was usually not an issue. For that reason, early data models were centered around collection objects (i.e. specimens). 1 When the Darwin Core standard was developed, one objective was to facilitate the integration of observations with specimens. Since resampling of the same organism over time was relatively common, the DwC standard included the term dwc:individualID2 which was intended to facilitate resampling by providing a means to refer to an identifier for an individual organism or small population of organisms.
At around the same time, the Global Bioidiversity Information Facility (GBIF) stated the goal of capturing the potentially large pool of biodiversity information contained in records other than specimens: "Developing the network’s capacity to discover and mobilise a greatly expanded array of primary biodiversity data types – beyond specimens and observational records – is critical."3 In this context, the potential existed for multiple types of evidence (e.g. specimens, images, and tissue samples, perhaps collected from the same organism at different times and places) to be associated with a particular individual organism.
This rationale (articulated in Baskauf 20104) for including the individual organism as a resource class resulted in a proposal to add Individual as a class to the Darwin Core standard.5 The discussion which ensued on the tdwg-content email list following this proposal clarified the position of the proposed class in a broad data model, but also raised additional questions about the scope and use cases of the proposed class.
The centrality of IndividualOrganism in DSW also makes possible easy linkage from ecological samples, where repeat observations of individuals in field samples are often the core data (e.g., CTFS forest plots), to DwC-based taxonomic resources.
1 See the ASC model at http://wiki.tdwg.org/twiki/bin/view/TAG/HistoricalDocuments
2 http://rs.tdwg.org/dwc/terms/index.htm#individualID
3 http://www.gbif.org/informatics/primary-data/ accessed 2011-04-14
4 https://journals.ku.edu/index.php/jbi/article/view/3664
5 http://code.google.com/p/darwincore/issues/detail?id=69
An extensive discussion of what the proposed dwc:Individual class should include1 was focused on the taxonomic scope of an Individual and the conceptual nature of what an Individual represents.
Initially, an Individual was described as consisting of a single "species or lower taxonomic level"2, but that was considered problematic because an Identification might be less definite than at the level of species 3 and because of issues related to the status of "a species"4. The definition was amended to indicate that an Individual represented a single Taxon. Subsequent discussion focused on the meaning of Identifications that were less precise than the lowest possible taxonomic level. If an Identification were made at a higher level, it could indicate that either: there was taxonomic homogeneity but the Identification could not (yet) be made to a lower level, or that the Individual was taxonomically heterogeneous and that the level indicated the lowest level of classification that applied to all organisms present in this "composite" individual.5 The issue of how to represent "composite" resources had been raised previously6 and broadening the definition of Individual to include taxonomically heterogeneous resources was a potential solution to this problem.7 In the event that an Individual was allowed to be taxonomically heterogeneous, a term such as individualScope would be needed to distinguish between the two circumstances where an Individual would be identified to a taxonomic level higher than the lowest possible level (i.e. indefinite identification vs. inclusion of multiple taxa determined to lowest common taxon).8 As the discussion progressed, it was agreed that the proposed dwc:Individual class should be allowed to contain "composite" entities that were taxonomically heterogeneous (although not everyone considered this to be the best solution9 10).
The difficult sticking point on the definition of an Individual was the breadth of resources that should be considered Individuals. The rather lengthy thread11 discussing whether organism parts (i.e. specimens and other Tokens derived physically from organisms) should be included in the definition of dwc:Individual was summarized succinctly: "Steve's idea for the Individual more closely resembles a many-to-many joining table in a database (ie doesn't serve much use other than connecting two tables/classes together - and doesn't normally relate to a 'real world' type of object). Whereas it seems Rich's idea is to relate it more to 'real-world' objects, such as samples, re-samples, etc, to allow tracking and connectability of the observed/collected/processed individuals."12 At this point the proposal for adding the class stalled as it became apparent that there was no consensus view of what an "Individual" represented conceptually and that the scope of an Individual would depend on the design of the model that included "Individual" as a class. In part, the creation of darwin-sw was a response to the need for a more concrete model that describes this and other classes, and their properties.12
In darwin-sw, we limit the scope of the IndividualOrganism class to resources that are taxonomically homogeneous. In an alternative model, we define another class for taxonomically heterogeneous resources (see the wiki page TaxonomicHeterogeneity for a description of this class and how it would be related to IndividualOrganism).
1 Thread begins with http://lists.tdwg.org/pipermail/tdwg-content/2010-November/001856.html and continues for at least 50 messages.
2 http://code.google.com/p/darwincore/issues/detail?id=69
3 http://lists.tdwg.org/pipermail/tdwg-content/2010-November/001856.html
4 http://lists.tdwg.org/pipermail/tdwg-content/2010-November/001863.html
5 http://lists.tdwg.org/pipermail/tdwg-content/2010-November/001865.html
6 http://lists.tdwg.org/pipermail/tdwg-content/2010-September/000056.html and
http://lists.tdwg.org/pipermail/tdwg-content/2010-September/000055.html
7 http://lists.tdwg.org/pipermail/tdwg-content/2010-November/001821.html and subsequent posts of that thread
8 http://lists.tdwg.org/pipermail/tdwg-content/2010-November/001895.html
9 http://lists.tdwg.org/pipermail/tdwg-content/2010-November/001869.html
10 http://lists.tdwg.org/pipermail/tdwg-content/2010-November/001971.html
11 The majority of this thread includes these posts and subsequent responses:
http://lists.tdwg.org/pipermail/tdwg-content/2010-November/001912.html
http://lists.tdwg.org/pipermail/tdwg-content/2010-November/001946.html
12 http://lists.tdwg.org/pipermail/tdwg-content/2010-November/001956.html
In the development of darwin-sw, the role of an "individual" as a node (or many-to-many joining table if you prefer) to connect other classes (Occurrence, Identification, and Token) was separated from its role as documentary evidence (i.e. a Token). Thus it falls in a position between Occurrence and Identification1. Tokens are considered to be a separate class of "evidence" that is not necessarily associated with a single class2, but rather serves as evidence for instances of the Occurrence, IndividualOrganism,and Identification classes (see the wiki page for Token for more details). The focus of the IndividualOrganism class is not so much "what is an individual", but rather is defined to facilitate the use cases that require it.3 As such the class could have been named "TaxonomicallyHomogeneousUnit" since "things" other than individual organisms can fill the same functional role (e.g. clonal organisms that do not have distinct individuals and small populations of the same taxon where individuals are not easily distinguished). A TaxonomicallyHomogeneousUnit (= IndividualOrganism) is really any kind of "thing" that can fill the role of connecting classes as described above - there is no need to agonize over its exact nature or to try to extrapolate based on the name chosen for the class.
In the same spirit, a Token is defined based on its role as "evidence" rather than some absolute definition of its "nature". A Token can simultaneously serve as evidence for one or more of the Occurrence, IndividualOrganism,and Identification classes. It should also be pointed out that there is no requirement in RDF that a resource be an instance of only a single class (i.e. a resource may have more than one rdf:type). Therefore, it is possible that an individual organism (e.g. a LivingSpecimen like a tree in a botanical garden) may be an instance of both the IndividualOrganism and Token if it is documented in such a way that it can serve as evidence. (It should be noted that many classes in DSW are declared to be disjoint. So a resource could not validly be an instance of two disjoint DSW classes. However, IndividualOrganism and Token are NOT disjoint classes in DSW.) However, it is not required that this be the case - Tokens may be considered dsw:derivedFrom an IndividualOrganism. In the case of a whole, dead organism that is a PreservedSpecimen, it could be considered to have a single GUID and be both an IndividualOrganism and a Token, or to be a Token having a separate GUID dsw:derivedFrom from a more abstract IndividualOrganism with its own separate GUID.
It should also be noted that a Token can, and probably usually will also be an instance of some other class defined outside DSW. This typing would be based on the actual "kind" of thing that the Token was (e.g. dctype:StillImage, foaf:Document, etc.) and would serve as a means to infer the specific properties that one would expect that particular kind of Token to possess. See Token for additional information.
1 http://lists.tdwg.org/pipermail/tdwg-content/2010-October/001703.html
2 http://lists.tdwg.org/pipermail/tdwg-content/2010-November/001960.html which discusses whether Tokens should "hang" from Occurrences or Individuals. In darwin-sw, the answer is "both".
3 http://lists.tdwg.org/pipermail/tdwg-content/2010-November/001971.html
- individualOrganismRemarks Text literal
- hasOccurrence range Occurrence
- hasIdentification range Identification
- hasDerivative range is any token that is derived directly from the IndividualOrganism. May be a physical thing like a specimen, an information resource like a digital image, a text reference, etc.
- dcterms:hasPart range can include any physical object that was removed from the IndividualOrganism, e.g. preserved specimen, blood sample (see http://dublincore.org/documents/dcmi-terms/#terms-hasPart)
- foaf:depiction range dctype:StillImage (?) see http://xmlns.com/foaf/spec/#term_depiction