Consistency and categorization of dataset labels

[Charlottevm]

For consistency sake and categorization of datasets we thought of a new way to name datasets both in ETSource and ETLocal.

What do you think of the following @redekok @mabijkerk ?

{COUNTRY}{CODE}{NAME}_{YEAR} (type and code are only relevant when it is not a country)

Examples: NL_GM345_veenendaal_2019 NL_netherlands_2019 VU_NH5_efate_2020 NL_RG01_drechtsteden_2019 NL_RES02_drechtsteden_2019

To do (concept):

• Look at impact on ETLocal, ETDataset, ETModel
• Migration in ETLocal
• Changing names in ETSource
• Changing datasets in ETDataset
• Communication with clients using the API

[mabijkerk]

We discussed this issue in the teammeeting and arrived at the following conclusions:

• We should (and do at the moment) allow multiple years of datasets to be active simultaneously, so the addition of the {YEAR} is required
• Some users just want to access the (latest) dataset, without having to specify the year. At least for the API we should therefore add a rule that selects the latest dataset if no year is specified. User would then be able to use the code NL_GM345_veenendaal and be automatically direct to NL_GM345_veenendaal_2019

[redekok]

Tiny little nitpick: make sure not to forget the zeros in the area codes. The geo ID for Veenendaal is GM0345 (not GM345). 🤓