New Term - identifiedBy

[tucotuco]

New term

• Submitter: Humboldt Extension Task Group
• Efficacy Justification (why is this term necessary?): Part of a package of terms in support of biological inventory data.
• Demand Justification (name at least two organizations that independently need this term): The Humboldt Extension Task Group proposing this term consists of numerous organizations.
• Stability Justification (what concerns are there that this might affect existing implementations?): None
• Implications for dwciri: namespace (does this change affect a dwciri term version)?: Needs ecoiri: equivalent.

Proposed attributes of the new term:

• Term name (in lowerCamelCase for properties, UpperCamelCase for classes): identifiedBy
• Term label (English, not normative): Identified By
• Organized in Class (e.g., Occurrence, Event, Location, Taxon): Event
• Definition of the term (normative): A list (concatenated and separated) of names of people, groups, or organizations who assigned the dwc:Taxon to the subject.
• Usage comments (recommendations regarding content, etc., not normative): Recommended best practice is to separate multiple values in a list with space vertical bar space ( | ).
• Examples (not normative): James L. Patton; Theodore Pappenfuss | Robert Macey
• Refines (identifier of the broader term this term refines; normative): None
• Replaces (identifier of the existing term that would be deprecated and replaced by this term; normative): None
• ABCD 2.06 (XPATH of the equivalent term in ABCD or EFG; not normative): DataSets/DataSet/Units/Unit/Identifications/Identification/Identifiers/IdentifiersText

[nielsklazenga]

Is this a new term, or is it the existing http://rs.tdwg.org/dwc/terms/identifiedBy? The definition is the same.

[tucotuco]

This is anew term in the eco: namespace an applies at the level of a dwc:Event. The same with eco:identificationReferences.

[nielsklazenga]

Any reason why dwc:identifiedBy cannot be used at the event level? Mind you, I am not objecting, just trying to understand. I am presuming dwc:identifiedBy is at the level of a dwc:Occurrence, but I do not understand the difference in this context. I guess I am just looking for an example of a situation where the two are different.

[nielsklazenga]

Just found an example in the Usage Guide:

1. dwc:eventID: BROKE_WEST_RMT_006 Anton Van de Putte recorded (samplingPerformedBy) and identified (identifiedBy) the catch from the target trawl sampling Event.

Clearly a different thing, but the definition is misleading, as the subject is the sample and that cannot always be assigned to a single dwc:Taxon. I think a different definition is required and perhaps a different local name.

[tucotuco]

The context for the usage here is that all of the Identifications of dwc:Organisms within a dwc:Event were done by the agent given in eco:identifiedBy. It is needed at the dwc:Event level for cases where there are absences or eco:excludedTaxa that would not show up in related dwc:Occurrences (an agent made a determination that no dwc:Organisms of a given dwc:Taxon were detected). That in itself doesn't mean that eco:identifiedBy is different from dwc:identifiedBy, it just serves a purpose that dwc:identifiedBy can't serve if it must be a property of a dwc:Occurrence. Since property assignment in Darwin Core is only organizational, that doesn't mean that eco:identifiedBy is different from dwc:identifiedBy either. They are effectively the same, with one applied to dwc:Events and one applied to dwc:Occurrences. I believe that the rationale here is to avoid confusing the two uses as there are currently no semantics to define a dwc:Occurrence as a special kind of dwc:Event.

[tucotuco]

Just found an example in the Usage Guide:

1. dwc:eventID: BROKE_WEST_RMT_006 Anton Van de Putte recorded (samplingPerformedBy) and identified (identifiedBy) the catch from the target trawl sampling Event.

Clearly a different thing, but the definition is misleading, as the subject is the sample and that cannot always be assigned to a single dwc:Taxon. I think a different definition is required and perhaps a different local name.

Specific recommendations are very welcome.

[nielsklazenga]

Thanks for the explanation @tucotuco. I can have a think about the definition but just adding the first part of your explanation above as a comment to the term will do the trick I reckon.

[tucotuco]

If that IS sufficient it is a nice solution. Thanks @nielsklazenga for the extra eyes on this. The Task Group's eyes (I include mine here) may have glazed over after three years of looking at these terms. ;-)

[timrobertson100]

@tucotuco - can you see any reasonable solution that would reuse the existing DwC term or is it simply unpractical, please?

Given DwC only provides a loose grouping of terms rather than providing a strict semantic model, is it problematic that a DwC term from the Identification group (not Occurrence) be used on an eco:Event? It seems a shame to mint and govern the exact same term in a different namespace. ~Note that identifiedByID would also be affected if that were to be introduced in HC~ (consciously omitted in HC relying instead on IRI terms).

Thanks.

(My comment is only intended to check options have been explored, not try and block progress)

[tucotuco]

@timrobertson100 With Darwin Core we have the luxury of undefined semantics. By this I mean that, though we organize dwc:identifiedBy under the dwc:Identification Class, that organization has no binding meaning. That leaves us free to have a dwc:Identification of a dwc:Occurrence, or of a dwc:Event, or of a dwc:Organism or of (the dwc:Organisms associated with) a dwc:Event. This would change if a semantic layer on top of the Darwin Core "bag of terms" were implemented saying, for example, that a dwc:Identification is always made on a dwc:Organism, but that is not currently the case and would require a more fully described structure to the data.

The relevant technical question right now is probably, "Does eco:identifiedBy actually mean something different, on its own, from dwc:identifiedBy?" Their definitions are identical, "A list (concatenated and separated) of names of people, groups, or organizations who assigned the dwc:Taxon to the subject." "Subject" leaves it open to apply to an instance of anything we like, so there should be no technical problem to use it in a dwc:Occurrence context or in a dwc:Event context.

The relevant practical question right now is, "Does it make any difference to a data publisher if the term is in the dwc: namespace or the eco: namespace?" I don't think so. It will likely be completely irrelevant to most people.

If a consensus can be reached that the dwc: term can serve perfectly well in place of the proposed eco: term, we can remove the new term from the proposal and replace it with the dwc: version in the Extension definition. In other words, the Humboldt Extension would still include the identifiedBy term, but would not require a new term to be created to fill this role.

The same considerations apply equally to the proposal for eco:identificationReferences.

[tucotuco]

In the Humboldt Extension Task Group meeting today it was agreed that the eco:identifiedBy term need not be created if the dwc:identifiedBy term is included in the extension in its place. However, doing so, The Task Group proposes that an addition be made to the comments of the dwc:identifiedBy term to reflect the use of the term in the Humboldt Extension context. A new issue to propose that change has been created and should be considered part of the Humboldt Extension public review milestone.