search

RDA-DMP-Common / RDA-DMP-Common-Standard

Official outputs from the RDA DMP Common Standards WG
The Unlicense
65 stars 34 forks source link

Error in v1.0: dataset.language #21

Closed hmpf closed 4 years ago

hmpf commented 4 years ago

"ISO 6391-1" is probably a typo for "ISO 639-1". They are not country-codes but language codes.

Furthermore, how is "language of dataset" defined?

If it is a set of texts in a specific language, then ISO 639-1 is not good enough for linguistics. You'd need ISO 639-3 at a minimum, there are after all still more than 676 languages on this planet. But since a single dataset from linguistics can cover multiple languages (comparative datasets, for instance), a single field is not enough.

So: what is the actual purpose of dataset.language?

hmpf commented 4 years ago

This has been changed to ISO 639-3, so closing. But still needed: purpose of the field, in case of a dataset of multiple languages.