Based on ORCID or an institutional staff registry, the repository does a regular crawl for new metadata linked to authors of the institution. Once a new item is discovered that is not yet present in the repository (Duplicate detection!), a workspace item is automatically created with the metadata from the retrieved item.

After this item is created, an email goes out to the author asking him/her to REPLY TO THE EMAIL with the full text as an attachment. The message of the email clearly indicates which version of the file is acceptable, and that replying to the email also implies AGREEING TO THE COLLECTION LICENSE.

The email reply that contains the attachment gets processed automatically by the repository, and pushes the item further into the workflow (or straight into the repository if no workflow is defined for the collection).