We provide continuing medical education in the form of interactive cases. Our database is available in a curated form through apps. There are two forms of publication: curated modules, which include multiple cases, and individual cases. I would guess that I register the database like a book with CrossRef, the curated learning modules like book chapters, and individual cases like a dataset.
Is there anything wrong with these levels of hierarchy? After all, this incidentally affects the costs associated with DOI registration.
By the way, we are in the process of adding appropriate landing pages to our apps so that we can work with DOI at all.
There’s no one perfect answer to how to register content that falls outside of the bounds of our named content types, especially when there are multiple layers of hierarchy involved.
That said, it would be most straightforward and make for the clearest metadata record if all three of those objects (database, modules, and cases) utilized the same content type.
For example, you could register the database as a Book Series, the modules as Books, and the cases as Book Chapters within their respective books.
Similarly, you could register the database as Report Series, the modules as Reports, and the cases as Content Items (essentially chapters) within the reports.
If you would strongly prefer to represent the cases as datasets, you can do that, but then you would have to explicitly insert Relationship metadata to connect each dataset to its parent/container module. Whereas, if you registered the cases as book chapters or report content items, the parent/child relationship with the modules (as either books or reports) would be automatically made.
Please let us know if you have any other questions.