The data dumps in different formats currently reference sources and collections by their identifier, but those identifiers are meaningless to a data downloader.
To provider a fuller picture, the collections and sources possibly need matching data dumps? Alternatively, the fields could contain a unique identifier, such as the OpenAlex ID for the source and just the name of the collection.
Check all fields for meaningfulness/understandability.
Also, is the "job" column useful?
The data dumps in different formats currently reference sources and collections by their identifier, but those identifiers are meaningless to a data downloader.
To provider a fuller picture, the collections and sources possibly need matching data dumps? Alternatively, the fields could contain a unique identifier, such as the OpenAlex ID for the source and just the name of the collection.
Check all fields for meaningfulness/understandability.
Also, is the "job" column useful?