Teleconference 04 December 2018

Participants

Steve Baskauf, Stan Blum, Jeff Gerbracht, Niels Klazenga, Richard Pyle, Cam Webb, Greg Whitbread. Apology: Markus Döring, Nico Franz

Draft spec.

The topic of the meeting was to discuss the very initial draft specification document at https://github.com/tdwg/tnc/blob/master/specification/tnu.adoc.

The following things were discussed/decided on:

General

• Document will be reworked to Google Doc, so it will be easier for people to edit and comment. There will be periodical exports to GitHub. (#23)
• Need to provide correct links to the TDWG LSID Ontology document (or remove the hyperlink entirely). The qualified name URIs give 404 errors.
• We’ll use dcterms:BibliographicResource for the Publication class in TCS
• We’ll adopt Darwin Core terms when they are fully equivalent. This came up under nomenclaturalNote, but applies to e.g. taxonRank and verbatimTaxonRank as well.
• TaxonConcepts came up again. While we are no longer making taxon concepts explicit in the specification itself, specification documentation should detail where the taxon concepts are in the proposed new specification.

TaxonomicName

• Change property name to taxonomicNameString.
• Include the parts the taxonomic name string is composed from. In TCS these are uninomial, genus, infragenericEpithet, specificEpithet and infraspecificEpithet.
• We will, for now, not include the parts that compose the (nomenclatural) authorship. Reasoning behind this is that, if you want to parse up authorship properly, it can become very complex.
• We’ll also include a taxonomicNameStringWithAuthor (exact label open for discussion). Example that was given is autonyms, where the authorship cannot be simply concatenated to the end of the taxonomicNameString. While it could be argued that autonyms do not have authors, if there is a use for the term, it should be in this standard.
  • Also in this discussion, a taxonomicNameParentPart property and a formatted name with RDFa bits and pieces came up, but no decision was made on these. That will be for another meeting
• All the name strings discussed above are canonical name strings. We’ll have a verbatimTaxonomicNameString property on the TaxonomicNameUsage class.
• We discussed vernacular names a bit. It was suggested that vernacular name could be a property of the TaxonomicName, as they are often published concurrently, but we concluded that vernacular names are probably better treated as different TaxonomicNameUsages. We agreed this requires more looking into (#22).
• We added a kindOfName property. Greg to provide the ‘name type’ vocabulary used in APNI (NSL). [Contribution from @mdoering after meeting:] GBIF and CoL+ use https://github.com/gbif/name-parser/blob/master/name-parser-api/src/main/java/org/gbif/nameparser/api/NameType.java#L21.
• We agreed that properties to do with typification are best placed in the TaxonomicNameUsage class (so are for a future meeting).
• Discussion ended (because of time) with whether “properties” in the TCS NomenclaturalNoteType (spellingCorrectionOf, basionym, basedOn, conservedAgainst, laterHomonymOf, sanctioned, replacementNameFor) are best treated as relationships between TNUs or as properties on a TaxonomicName object. Greg proposed to document current implementations and an ideal implementation on the Wiki. (#23).