Vocabulary codelists - "reserving" a number

(Steven Flower) #1


We’ve an interesting dilemma via research on the Open Ag Funding project.

We may want to propose an additional SectorVocabulary code, to assist publishers, for example. However, we would also plan a period of piloting and data exploration, before any proposal

Our (slight) difficulty is that the vocabulary codelists seem to be based on incremental integers - eg: SectorVocabulary currently has 1-6 codes, so we’re assuming the next on will be 7.

This could make it tricky to pilot anything. If we pilot with code 7, but could then find this is allocated to another purpose.

Would there ever be a situation whereby codes could be “pre-reserved” (or, “emerging”) , pending full proposal/acceptance? I appreciate that this in turn could generate gaps in the list(s)


Tech Paper: Codelist Management
(Bill Anderson) #2

Hi Steven

There is nothing in the standard that says that codelists MUST be based on incremental integers. If you send a formal request to support@iatistandard.org the Tech Team can ‘reserve’ this.



(Wendy Rogers) #3

Just to add that the Sector Vocabulary codelist (at V2.02) now has 1 to 10 in use so 11 would be the next number on the list to reserve or assign - http://iatistandard.org/202/codelists/SectorVocabulary/

(Steven Flower) #4

Thanks @bill_anderson & @Wendy

One other point of (related) clarification would be useful.

In 2.02 the attribute <vocabulary-uri> exists for <sector>: http://support.iatistandard.org/entries/105713163-Add-URI-attribute-to-elements-where-Reporting-organisation-vocabularies-are-used

The guidance seems to suggest that this should only be used when the vocab=99

If a publisher uses a vocabulary of 98 or 99 (i.e. ‘Reporting Organisation’), then the @vocabulary-uri attribute should also be used

I wonder if it feasible for different publishers to use vocab=99 and the same vocabulary-uri as an alternative way around the above issue? Or - do we always consider the 99 (Reporting Organisation) vocabulary to be unique and different across publishers?

(apologies if this is overly pushing the original definitions/design)

(Bill Anderson) #5

99 does not require the vocabulary-uri to be unique. If it is not unique - i.e. it is shared by a number of publishers - the question then arises as to how shared/sharable it is, and whether it should be recognised with its own vocab code.

(Wendy Rogers) #6

One other consideration here is that now that the attribute exists is it still necessary to retain ‘98’ on the Sector code vocabulary as multiple in-house codelists can be used and differentiated between by specifying an appropriate link in ?

I will also move this topic to the Standard Management -> Modifications, Additions, Improvements category for further consideration