This wiki retires in 2027; content deletion started in 2026. No planned cloud migration.
|
(Feedback from Australia)
This change was imported from CSPA LIM, as part of aligning GSIM 1.5 with the latter, rather than having a specific issue number in the development of GSIM 1.5.
A question was raised during ABS review, however, about the mandatory relationship to Measurement Unit given Value Domains may be Categorical/Enumerated.
It would be possible to define an arbitrary Measurement Unit (eg "Category") for categorical Value Domains but this adds no value given the fact the Value Domain itself is of sub-type Enumerated Value Domain.
A possible solution might be to move the relationship more specifically to Described Value Domain. Even there a Described Value Domain that contains a text string (or arguably a simple date rather than a time duration) does not have a meaningful Measurement Unit.
The ABS implementation addresses this by further sub-typing Described Value Domains and making the equivalent relationship mandatory for Numeric Value Domains.
A simpler solution for GSIM might be the approach used in the current version of ISO 11179 Part 3. ISO 11179 has Dimensionality (GSIM:MeasurementType) as 0..1 for Conceptual Domain and Unit of Measure (GSIM:MeasurmentUnit) as 0..1 for Value Domain.
2 Comments
Jenny Linnerud
22 Jan, 2019We support ABS that a mandatory relationship from Value Domain to Measurement Unit is not correct. It should be 0..1.
user-38d9f
24 Jan, 2019The cardinality between Value Domain and Measurement Unit is wrong in LIM.
Proposal: Change the cardnality to 0..1 Measurement Unit.