geoCodeType.csv codelist contains duplicate codes.

[caprenter]

As an example - but there are more:

WPC Westminster Parliamentary Constituencies An ONS Electoral codelist http://statistics.data.gov.uk/id/statistical-entity/E14 http://statistics.data.gov.uk/id/statistical-geography/E14000530 WPC Westminster Parliamentary Constituencies An ONS Electoral codelist http://statistics.data.gov.uk/id/statistical-entity/N06 http://statistics.data.gov.uk/id/statistical-geography/N06000001 WPC Westminster Parliamentary Constituencies An ONS Electoral codelist http://statistics.data.gov.uk/id/statistical-entity/S14 http://statistics.data.gov.uk/id/statistical-geography/S14000001 WPC Westminster Parliamentary Constituencies An ONS Electoral codelist http://statistics.data.gov.uk/id/statistical-entity/W07 http://statistics.data.gov.uk/id/statistical-geography/W07000041

[timgdavies]

Ah. Intersting.

I believe that was generated from a SPARQL query against http://statistics.data.gov.uk/ which assumed they would not re-use the same abbreviation for different geographies.

From the look of it, they have different entities for geographies in Wales, Scotland, Northern Ireland and England - but the codes so not overlap (prefixed variously with E14, N06, S14 or W07) - so this can be treated as one vocabulary.

I suspect (but have not checked) the same is true for the other duplicate cases.

I suggest we remove the duplicates. We can either:

(a) Leave the documentationURL of the first one we find, and add a note to documentation for this codelist that any relatedEntities are also covered by same codelist identifier.

(b) Find alternative documentationURLs which cover the whole set of related entities (I couldn't find any good candidates on the ONS site from poking around)

[stevieflow]

AFAIK this is the only codelist "maintained by 360.

We also know that this is used by publishers

Actions:

• @morchickit to include reference to this in the discussions on codelists #23
• we should seek the authoritative source for this list (if it had changed)
• we should decide whether to include this list in the first release of the standard - or just link to a relevant source (but then be minded of #90 which discusses links in the schema)

[Bjwebb]

BTW, there's a suggestion from @timgdavies to substantially change this codelist - https://github.com/ThreeSixtyGiving/standard/issues/155#issuecomment-314807867

[morchickit]

See forum thread here: https://forum.threesixtygiving.org/t/codelists-and-360giving-standard/20