https://github.com/tdwg/tnc/tree/master/TDWG2018
Recent work on names is fragmented into individual and isolated project efforts. There are now more terms than concepts, more definitions than terms. Many, more-or-less, similar vocabularies, alternative frameworks and non-interoperable, application systems.
This may well be the natural process of evolution but we do need to make time now to discuss synthesis and a more collaborative approach to development of a workable standard for names. Can we work together to fill the gaps, and to resolve the competing requirements of aggregation and science, data and knowledge, product and research, Code-compliance and interoperability.
To raise awareness of the importance of names to biodiversity informatics and to re-invigorate a more collaborative phase of standards development with names within the TDWG umbrella.
This workshop will aim for consensus around the idea that we can work together within TDWG to evolve a pragmatic replacement for the existing names standards, and a map to get us there.
- Reconvene the Taxonomic Names and Concepts Interest Group
- Membership
- GitHub, tdwg/tnc
- Standards maintenance.
- Agreement on a plan for a replacement of the Taxonomic Concept Transfer Schema (TCS) that:
- is compliant with the TDWG Standards Documentation Standard and Vocabulary Maintenance Standard.
- provides for the needs of the various types of users and use cases of Taxonomic Names in the biodiversity informatics space
- as much as possible synthesises and takes benefit of the work that has been done in the TDWG community so far.
- Simple checklists, aggregators: GBIF, ALA, CoL; agitators: @rdmpage
- Species Information Systems
- Floras, Faunas etc.
- Systems that capture historical (as well as current) name usages: NSL, APNI, AFD
- Systems that try to capture phylogenetic relationships
- Entire TDWG community: everybody needs taxa for something
- Taxonomic Concept Transfer Schema (TCS)
- TCS 1.01 Schema Documentation can be found at http://tdwg.github.io/tcs/schema/1.01
- Darwin Core
- Darwin Core terms can be found at http://tdwg.github.io/dwc/terms/
Expectation reversal, but
- Taxa change
- AFD = taxonomic (opinion) product, not a Nomenclator, not a taxon store. - @rdmpage: WTAF
- Taxonomic Names are stable.
- Inappropriate use of taxon.
No "Names" standard.
- Darwin Core RDF Guide recommends separate vocabulary.
- Ad hoc Darwin Core extensions.
- Inappropriate use of the taxonomic concept
TCS
- Application Schema rather than Standard
- Designed to exchange with systems that look like TCS; not general information exchange
- Interchange of literature based concept definitions
- Difficult to distill, harder to extend.
- Syntax (XML) → path to RDF … JSON.
- Over engineered for normal usage
- Relationship object
- Path to RDF, JSON-LD
- How does tcs:TaxonConcept relate to dwc:Taxon?
- How does tcs:TaxonConcept relate to NameUsage and nsl:Instance?
Darwin Core
- Ambiguous wrt names
- Overloaded definitions
- Many interpretations.
- Overloaded usage.
- Classification terms.
- Hybrids
- Ranks.
- Path to RDF
Many different application profiles
- Simple vocabulary required.
A Domain Model?
- Agreement on definitions
- Code compliance
- Contract -> Ontology
Open Linked Data ++
- Identifiers
- knowledge graph
- RDF, JSON-LD
- schema.org
Services
???
- Bisby
- APNI
- Berlin Model
- IPNI
- ITIS, CoL
- Flora of Australia
- Tropicos
- LinneanCore (SDD)
- DwC, ABCD
- TCS
- TDWG ontology
- ALA-NSL (2008)
- txn:deVriese
- GBIF Checklist
- DwC RDF
- Nomen
- WFO
- OpenBiodiv
- BCO
- ColPlus
- TaxRef
- +++
A new standard or update/replace/extend existing standards?
Restart with ideas in Linnean Core (J.Cooper)
- A Basic vocabulary for the interchange and use of Taxonomic Names and "names".
- Darwin Core extensions NFB < TCS.
- Controlled vocabularies!
- Recommend classes.
- Guidelines.
Darwin Core Taxon
- leave "convenience" terms and internal links.
- remove external links (id terms) ?
- Add new terms?
- Darwin Core RDF Guide
A Domain Model or List?
- Normative or Guide?
- API specification for:
- Name
- Instance (name occurrence) according
- Usage ( potential/taxonomic concept)
- Tree ?
- Arrangement
(Core competencies / Functional Requirements / Use cases)
- Generic metadata model
- Vocabularies
- Full Code support
- ICN, ICZN, ...
- Code maintenance
- Domain owners?
- Interoperability
- Every name
- Every instance (name usage)
- Every relationship
- Every point of view
- Every version
- Every syntax
- Every application profile
- Common interchange format
- Support for publication
- Extensible
- Obey the TDWG laws for standards and vocabularies
- Independent of serialization
- Concise vocabulary
- Linked Data
- Shared vocabulary
- Application Profiles
- Simple enough for …. Anything
Some core "concepts" [ToDo: do we need to include "taxon" concept and "taxon"?]
A name is a "name", a designation, a label, a string used as a name or in place of a name: in any context.
A scientific name a name formulated (or attempted to be) according to some rules of nomenclature; uni/bi/tri.nomial_name, author, [year]
A taxonomic name is a name used in a taxonomic context, i.e. with a reference; often a scientific name.
A reference is a citable work: a publication, a resource, ... might be local.usage.
A taxonomic name instance is the occurrence of a name in a reference. A fact. One use of a taxonomic name establishing context: establishment of a name, a concept, a synonym, an assertion, a relationship; tax. nov., comb.nov., … , a misapplication, or assertion; a common name.
A taxonomic name usage is a collection of taxonomic name instances, establishing, annotating or describing a theory of a taxonomic entity - a taxonomic concept or OTU - often linking synonymic inclusions and adding annotations, description…
- issue#1: taxon concept vs concept label
A taxonomic tree is a hierarchical collection of selected, unique, taxonomic name usages - a taxonomy.
A taxonomic arrangement or named tree is a collection of taxonomic trees. A classification, checklist, Flora or Fauna.
A taxon :
A taxon concept :
Note: In a Names standard ‘scientific name’ is a type of taxonomic name rather than a property of a Taxon.
Bibliographic pattern: Where a name is found and how it is used in that place.
- Every taxonomic name instance.
-Just about names and taxa:
- Authors, References detail elsewhere - external resources
* Name
* Name Strings
* Code
* Type of name (scientific, vernacular, informal, cultivar…)
* Rank !
* Status !
* Authorship
* Primary reference ( incl. basionym) !
* Reference
* Type of reference !
* Part of Reference !
* dc:title, dc:creator, dc:created:...
* CitedAs
* External resource !
* Name Instance
* Type of instance !
* Name !
* Reference !
* Page citation
* Citing instance !
* Cited instance !
* Taxonomic tree
* Instance !
* Parent node !
* Included nodes !
* node !
* node !
* ...
* Annotations
* Profile
* Taxonomic arrangement
* Root node!
* node !
* Vocabularies
* Code of Nomenclature
* Status of a Name
* Code !
* Status
* Type of Name
* Type of Name Instance (usage/relationship)
* The Ranks of Names
* Code !
* Rank
* Type of Reference
* Type of external Resource
* Type of treenode
* Type of arrangement
Appropriateness
- RDF Guide
- "Convenience terms"
Extension of Darwin Core
- New/missing terms
- "Names" guide
- RDF Guide
- GBIF Checklist?
Cull Darwin core (TaxonConceptID)