Metadata Languages should be able to be disabled per collection, and set at the dataset level

[kaitlinnewson]

Overview of the Feature Request

This is a follow-up ticket to a Dataverse Community Slack conversation on Feb. 1st.

The Metadata Languages feature is currently configurable at the installation level and, when enabled, is not able to be disabled per collection. This becomes and issue in a multilingual Dataverse setting, particularly when a collection can contain datasets with more than one language. To improve this feature and make it more broadly usable, there are two suggestions:

• Allow for the setting to be turned off for each collection. In our case, we want to disable it at the root collection, otherwise it is inherited into all sub-collections.
• Allow for this to be set at the dataset level, instead of only the collection level. A collection may have datasets in more than one language, and this setting makes much more sense since it applies to a dataset. It could default to the collection-level setting, but be changed by the dataset creator.

What kind of user is the feature intended for?
(Example users roles: API User, Curator, Depositor, Guest, Superuser, Sysadmin)

Curators, dataset creators

What inspired the request?

We enabled this setting when upgrading at Scholars Portal, but turned the feature off because it was showing all sub-collections as using "English" metadata. This was an issue because we have bilingual and French institutions using our service, and were concerned it would be confusing from a UX perspective as well as it resulted in the metadata record saying the dataset was in English as soon as the feature was enabled ("This dataset has been configured to use English as the language for all metadata entries."), when it may not have been.

What existing behavior do you want changed?

The Metadata Languages feature (see above).

Any brand new behavior do you want to add to Dataverse?

Metadata languages set at the dataset level.