Recommendations for handling complex values #43
Description
NOTE: this issue was renamed so that it could cover broader issues related to complex values.
Original issue posted by @ymgan :
The Humboldt Extension group encounters several terms where the list of values in a field have to be paired with list of values in another fields in order to be meaningful and we are struggling with the recommendation to deal with that. Please see an example below:
| eventID | targetTaxonomicScope | targetLifeStageScope |
|---|---|---|
| BROKE_WEST_RMT_006 | Myctophidae | Macrouridae | Artedidraconidae | Channichthydae | Nototheniidae | all | larvae and juvenile | larvae and juvenile | larvae and juvenile | larvae and juvenile |
This reads, in BROKE_WEST_RMT_006 Event, the targets are:
- all life stages of Myctophidae
- only larvae and juvenile of Macrouridae
- only larvae and juvenile of Artedidraconidae
- only larvae and juvenile of Channichthydae
- only larvae and juvenile of Nototheniidae
We do not know how to best represent this type of information. I did this in the Humboldt use case mapping exercise during the testing phase (dataset here: https://ipt.gbif.org/resource?r=brokewest-fish) based on the suggestion received. The only way to pair this information is based on its order in the cell. This could potentially impact multiple target__Scope fields in Humboldt Extension.
Similarly, eBird encounters the same issue for the Humboldt terms samplingEffortValue and samplingEffortUnit. There are multiple values for samplingEffortValue that each need to pair with a separate corresponding unit in samplingEffortUnit for a single record.
We will appreciate it if there will be guidance on this issue in the near future! Thank you very much!