Closed jindrichmynarz closed 8 years ago
As you can see this is a fault of the source codelist itself
and occurs on multiple codes and some other codelists also. That's because some codes have the same description on multiple levels.
A possible solution to this could be to concatenate the description with the level so every label would be unique. Or we could leave them as is.
This is very, very common across budgetary classification in the real world, and is not a "fault" - it is a property of the data itself.
I couldn't agree more but also my thought on this is that these structures should be defined better, as this report also recommends
Well, yeah, the world should be different in many ways :). Until then....
OK, thanks for the explanation. This might make linking external datasets more ambiguous, because we would need to choose at what level to make a link based on a match on label shared by multiple concepts in the code list's hierarchy.
CPC contains several concepts that share the same preferred label. For example,
<http://data.openbudgets.eu/resource/codelist/cpc/98>
,<http://data.openbudgets.eu/resource/codelist/cpc/980>
,<http://data.openbudgets.eu/resource/codelist/cpc/9800>
, and<http://data.openbudgets.eu/resource/codelist/cpc/98000>
all share the preferred label"Domestic services"@en
. This is considered to be a bad practice (see here). Is this issue a problem of the source data or the conversion?