TG2-VALIDATION_DEGREEOFESTABLISHMENT_STANDARD about bdq HOT 16 OPEN

I think https://api.gbif.org/v1/vocabularies/DegreeOfEstablishment/concepts is OK as a source authority because it actually has an API, as long as it is understood that the actually vocabulary is maintained at https://dwc.tdwg.org/doe/ and the GBIF API is expected to remain up to date with that.

from bdq.

I think the better path is to convince GBIF to correct their implementations of the vocabularies. It was probably an oversight or an attempt to follow a capitalization pattern that was established before these vocabularies were added.

from bdq.

Should be made CORE - see comments under #268

from bdq.

Hi all

I think the better path is to convince GBIF to correct their implementations of the vocabularies. It was probably an oversight or an attempt to follow a capitalization pattern that was established before these vocabularies were added.

The GBIF concepts have been updated to lowerCamelCase to follow the TDWG convention for the following vocabularies:

https://registry.gbif.org/vocabulary/Pathway/concepts
https://registry.gbif.org/vocabulary/DegreeOfEstablishment/concepts
https://registry.gbif.org/vocabulary/EstablishmentMeans/concepts

from bdq.

Updated notes to change "fail" to more explicit "This test must return NOT_COMPLIANT if there is leading or trailing whitespace or there are leading or trailing non-printing characters. "

from bdq.

Source authority should be:

bdq:sourceAuthority default = "Degree of Establishment Controlled Vocabulary List of Terms" {[https://dwc.tdwg.org/doe/]} {GBIF vocabulary API [https://api.gbif.org/v1/vocabularies/DegreeOfEstablishment/]}

from bdq.

Changed Source Authority from

bdq:sourceAuthority default = "Darwin Core degreeOfEstablishment" {[https://dwc.tdwg.org/list/#dwc_degreeOfEstablishment]} {dwc:degreeOfEstablishment vocabulary API [https://api.gbif.org/v1/vocabularies/DegreeOfEstablishment/concepts]}

to

bdq:sourceAuthority default = "GBIF DegreeOfEstablishment Vocabulary" [https://api.gbif.org/v1/vocabularies/DegreeOfEstablishment]} {"dwc:degreeOfEstablishment vocabulary API" [https://api.gbif.org/v1/vocabularies/DegreeOfEstablishment/concepts]}

from bdq.

Source Authority should be

bdq:sourceAuthority default = "Degree of Establishment Controlled Vocabulary List of Terms" {[https://dwc.tdwg.org/doe/]} {GBIF vocabulary API [https://api.gbif.org/v1/vocabularies/DegreeOfEstablishment/concepts]}

from bdq.

Changed Source Authority from

bdq:sourceAuthority default = "GBIF DegreeOfEstablishment Vocabulary" [https://api.gbif.org/v1/vocabularies/DegreeOfEstablishment]} {"dwc:degreeOfEstablishment vocabulary API" [https://api.gbif.org/v1/vocabularies/DegreeOfEstablishment/concepts]}

to

bdq:sourceAuthority default = "Degree of Establishment Controlled Vocabulary List of Terms" {[https://dwc.tdwg.org/doe/]} {GBIF vocabulary API [https://api.gbif.org/v1/vocabularies/DegreeOfEstablishment/concepts]}

from bdq.

We have to address a case missmatch between the TDWG vocabulary and the GBIF API. The TDWG vocabulary is authoriative, and has both Label and Controlled Value with an initial letter lower case, e.g. "native", the GBIF API has Name and Label.Value in mixed case, e.g. "Native". A lookup in the GBIF API has a cross reference to the TDWG vocabulary by IRI (thus only the GBIF API needs to be consulted to check dwciri values (though not consistently between vocabularies)), but not by label, implementors using the GBIF API would still have to consult the TDWG vocabulary to obtain the actual controlled value for the dwc: term.

"native" should return Compliant, "Native" not compliant. referencing the GBIF API here does not assist either the specification or implementors.

We should probably remove the GBIF API from the source authority for this set of VALIDATION_..STANDARD tests (but include it for the related AMENDMENT..._STANDARDIZED tests, as it provides alternative terms that can be used for standardization.

This comment applies to #268, #269, #275, #276, #277, and #278 where the TDWG vocabulary has values with an initial lower case letter, and the GBIF API does not match, having values with an initial capital letter.

Examples:

Pathway:

http://rs.tdwg.org/dwcpw/values/p039 Controlled Value = hullFouling
vs
https://api.gbif.org/v1/vocabularies/Pathway/concepts/HullFouling/ name = HullFouling externalDefinitions[0]=http://rs.tdwg.org/dwcpw/values/p039

Degree of Establishment:

http://rs.tdwg.org/dwcdoe/values/d001 Controlled Value = native
vs
https://api.gbif.org/v1/vocabularies/DegreeOfEstablishment/concepts/Native/ name = "Native"
sameAsUris[0]='http://rs.tdwg.org/dwcdoe/values/d001

Establishment Means:

http://rs.tdwg.org/dwcem/values/e001 Controlled Value = native
vs
https://api.gbif.org/v1/vocabularies/EstablishmentMeans/concepts/Native/ name = "Native"
sameAsUris[0]=http://rs.tdwg.org/dwcem/values/e001

from bdq.

@tucotuco I agree. In the absence of that change, we should remove the reference to the GBIF API from the validations, and add a note for implementors to the ammendments.

from bdq.

@tucotuco yes, ideal would be an alignment of the text strings in the API to the standard, though GBIF doesn't necessarily have to change the values of the Names in their API, they only need to provide the TDWG IRI and ControlledVocabulary value in consistent places where we can tell implementors to look for them.

from bdq.

@timrobertson100

from bdq.

Thanks for the ping

It was probably an oversight or an attempt to follow a capitalization pattern that was established before these vocabularies were added.

It was indeed. Just for background, one design goal of the vocabulary server was to serve up various vocabularies (internal or external such as TDWG, IUCN etc) in a consistent manner through a unified API. When it comes to formats we're dealing with vocabularies that use lowerCamelCase, UpperCamelCase, lower_snake_case, CAPTIAL_SNAKE_CASE etc and - rightly or wrongly - the intent was to try and normalise things. That approach resulted in our following Concepts with UpperCamelCase and properties as lowerCamelCase similar to e.g. DwC at the time with Classes (HumanObservation).

I'll chat with people here and comment back on the feasibility of changing at our end - it may have implications I'm not aware of (e.g. external API users). As I understand it the options are either we change the name to lowerCamelCase or add a field controlledValue - if I'm mistaken please correct me.

More to follow...

from bdq.

from bdq.

GBIF vocabulary has now been aligned with Darwin Core. Thanks @timrobertson100

from bdq.